The integration of deep learning (DL) in breast MRI has revolutionized the field of medical imaging, notably enhancing diagnostic accuracy and efficiency. This review discusses the substantial influence of DL technologies across various facets of breast MRI, including image reconstruction, classification, object detection, segmentation, and prediction of clinical outcomes such as response to neoadjuvant chemotherapy and recurrence of breast cancer. Utilizing sophisticated models such as convolutional neural networks, recurrent neural networks, and generative adversarial networks, DL has improved image quality and precision, enabling more accurate differentiation between benign and malignant lesions and providing deeper insights into disease behavior and treatment responses. DL’s predictive capabilities for patient-specific outcomes also suggest potential for more personalized treatment strategies. The advancements in DL are pioneering a new era in breast cancer diagnostics, promising more personalized and effective healthcare solutions. Nonetheless, the integration of this technology into clinical practice faces challenges, necessitating further research, validation, and development of legal and ethical frameworks to fully leverage its potential.

Stroke is a major cause of disability and death in adults worldwide. In clinical setting, time efficient MRI protocols including diffusion weighted images, fluid attenuated inversion recovery, fast T2*-weighted images and MR angiography aim to establish ischemic stroke diagnosis, visualize vessel occlusion and determine the extent of ischemia damage distinguishing between the salvageable ischemic penumbra and the infarct core. Thus, MRI plays a pivotal role in diagnosis, treatment decision-making, and prognostic assessment, although prediction based on these elements remains limited and can be improved. We reviewed the added-values of alternative MRI methods such as sodium(²³Na) MRI and ¹H MR spectroscopic imaging that open new temporal and pathophysiological windows on ionic dys-homeostasis and metabolism alteration in the context of ischemic stroke and reperfusion. Insights on the timeline of the cellular events accessible using these alternative methods and perspectives to improve clinical outcome prediction of ischemic stroke patients are discussed.

Purpose: To evaluate how the relationship between respiratory interval (RI) and temporal resolution (TR) impacts image quality in free-breathing abdominal MRI (FB-aMRI) using golden-angle radial sparse parallel (GRASP).

Methods: Ten healthy volunteers (25.9 ± 2.5 years, four women) underwent 2 mins free-breathing fat-suppression T1-weighted imaging using GRASP at RIs of 3 and 5s (RI₃ and RI₅, respectively) and retrospectively reconstructed at TR of 1.8, 2.9, 4.8, and 7.7s (TR_1.8, TR_2.9, TR_4.8, and TR_7.7, respectively) in each patient. The standard deviation (SD) under the diaphragm was measured using SD maps showing the discrepancy for each horizontal section at all TRs. Two radiologists evaluated image quality (visualization of the right hepatic vein at the confluence of the inferior vena cava, posterior segment branch of portal vein, pancreas, left kidney, and artifacts) at all TRs using a 5-point scale.

Results: The SD was significantly higher at TR_1.8 compared to TR_4.8 (P < 0.01) and TR_7.7 (P < 0.001), as well as at TR_2.9 compared to TR_7.7 (P < 0.01) for both RIs. The SD between TR_4.8 and TR_7.7 did not differ for both RIs. For all visual assessment metrics, the TR_1.8 scores were significantly lower than the TR_4.8 and TR_7.7 scores for both RIs. The pancreas and left kidney scores at TR_2.9 were significantly lower than those at TR_7.7 (P < 0.05) for RI₅. Additionally, the left kidney score at TR_1.8 was lower than that at TR_2.9 (P < 0.05) for RI₃. All scores at TR_2.9, TR_4.8, and TR_7.7 were similar for RI₃, while those at TR_4.8 and TR_7.7 were similar for RI₅.

Conclusion: Prolonging the TRs compared to RIs enhances image quality in FB-aMRI using GRASP.

Purpose: To investigate the predictive performance of radiomic features extracted from breast MRI for upgrade of ductal carcinoma in situ (DCIS) to invasive carcinoma.

Methods: This retrospective study included 71 women with DCIS lesions diagnosed preoperatively by biopsy. All women underwent breast dynamic contrast-enhanced (DCE) MRI of the breast, which included pre-contrast and five post-contrast phases continuously with a time resolution of 60s. Lesion segmentation was performed manually, and 144 radiomic features of the lesions were extracted from T2-weighted images (T2WI), pre-contrast T1-weighted images (T1WI), and post-contrast 1st, 2nd, and 5th phase subtraction images on DCE-MRI. Qualitative features of mammography, ultrasound, and MRI were also assessed. Clinicopathological features were evaluated using medical records. The least absolute shrinkage and selection operator (LASSO) algorithm was applied for features selection and model building. The predictive performance of postoperative upgrade to invasive carcinoma was assessed using the area under the receiver operating characteristic curve.

Results: Surgical specimens revealed 13 lesions (18.3%) that were upgraded to invasive carcinoma. Among clinicopathological and qualitative features, age was the only significant predictive variable. No significant radiomic features were observed on T2WI and post-contrast 2nd phase subtraction images on DCE-MRI. The area under the curves (AUCs) of the LASSO radiomics model integrated with age were 0.915 for pre-contrast T1WI, 0.862 for post-contrast 1st phase subtraction images, and 0.833 for post-contrast 5th phase subtraction images. The AUCs of the 200-times bootstrap internal validations were 0.885, 0.832, and 0.775.

Conclusion: A radiomics approach using breast MRI may be a promising method for predicting the postoperative upgrade of DCIS. The present study showed that the radiomic features extracted from pre-contrast T1WI and post-contrast subtraction images in the very early phase of DCE-MRI were more predictable.

Purpose: This study was conducted to evaluate whether thin-slice 2D fat-saturated proton density-weighted images of the shoulder joint in three imaging planes combined with parallel imaging, partial Fourier technique, and denoising approach with deep learning-based reconstruction (dDLR) are more useful than 3D fat-saturated proton density multi-planar voxel images.

Methods: Eighteen patients who underwent MRI of the shoulder joint at 3T were enrolled. The denoising effect of dDLR in 2D was evaluated using coefficient of variation (CV). Qualitative evaluation of anatomical structures, noise, and artifacts in 2D after dDLR and 3D was performed by two radiologists using a five-point Likert scale. All were analyzed statistically. Gwet’s agreement coefficients were also calculated.

Results: The CV of 2D after dDLR was significantly lower than that before dDLR (P < 0.05). Both radiologists rated 2D higher than 3D for all anatomical structures and noise (P < 0.05), except for artifacts. Both Gwet’s agreement coefficients of anatomical structures, noise, and artifacts in 2D and 3D produced nearly perfect agreement between the two radiologists. The evaluation of 2D tended to be more reproducible than 3D.

Conclusion: 2D with parallel imaging, partial Fourier technique, and dDLR was proved to be superior to 3D for depicting shoulder joint structures with lower noise.

Purpose: We aimed to investigate the changes in intravoxel incoherent motion (IVIM) and diffusion parameters between in vivo and post-mortem conditions and the time dependency of these parameters using two different mouse tumor models with different vessel lumen sizes.

Methods: Six B16 and six MDA-MB-231 xenograft mice were scanned using 7 Tesla MRI under both in vivo/post-mortem conditions. Diffusion weighted imaging with 17 b-values (0–3000 s/mm²) were obtained at two diffusion times (9 and 27.6 ms). The shifted apparent diffusion coefficient (sADC) using 2 b-values (200 and 1500 s/mm²), non-Gaussian diffusion and IVIM parameters (ADC₀, K, f_IVIM) were estimated at each of the diffusion times. The results were evaluated by repeated measures two-way analysis of variance and post hoc Bonferroni test.

Results: In B16 tumors, f_IVIM significantly decreased with post-mortem conditions (from 12.6 ± 6.5% to 5.2 ± 1.9%, P < 0.05 at long diffusion time; from 11.0 ± 2.4% to 4.6 ± 2.7%, P < 0.05 at short diffusion time). In MDA-MB-231 tumors, f_IVIM also significantly decreased (from 8.8 ± 3.8% to 2.6 ± 1.1%, P < 0.05 at long; from 7.9 ± 5.4% to 2.9 ± 1.1%, P < 0.05 at short). No diffusion time dependency was observed (P = 0.59 in B16 and P = 0.77 in MDA-MB-231). The sADC and ADC₀ values tended to decrease and the K value tended to increase after sacrificing and when increasing the diffusion time.

Conclusion: The f_IVIM values dropped after sacrificing, confirming that IVIM MRI is a promising quantitative parameter to evaluate blood microcirculation. The presence of residual post-mortem f_IVIM values suggested that the influence of water molecule diffusion in the blood lumen may contribute to the IVIM effect. Diffusion MRI parameter’s time dependency and those changes after sacrificing could possibly provide additional insights into diffusion hindrance mechanisms.

Purpose: Chronic obstructive pulmonary disease (COPD) is a complex multisystem disease associated with comorbidities outside the lungs. The aim of this study was to measure changes in metrics of pulmonary gas exchange function and brain tissue metabolism in a mouse model of COPD using hyperpolarized ¹²⁹Xe (HP ¹²⁹Xe) MRI/MR spectroscopy (MRS) and investigate the relationship between the metrics of lung and brain.

Methods: COPD phenotypes were induced in 15 mice by 6-week administration of cigarette smoke extract (CSE) and lipopolysaccharide (LPS). A separate negative control (NC) group was formed of 6 mice administered with saline for 6 weeks. After these 6-week administrations, the pulmonary gas exchange function parameter f_D (%) and the rate constant, α (s⁻¹), which are composed of the cerebral blood flow F_i and the longitudinal relaxation rate 1/T_1i in brain tissue, were evaluated by HP ¹²⁹Xe MRI/MRS.

Results: The f_D of CSE-LPS mice was significantly lower than that of NC mice, which was in parallel with an increase in bronchial wall thickness. The α in the CSE-LPS mice decreased with the decrease of f_D in contrast to the trend in the NC mice. To further elucidate the opposed trend, the contribution of T_1i was separately determined by measuring F_i. The T_1i in the CSE-LPS mice was found to correlate negatively with f_D as opposed to the positive trend in the NC mice. The opposite trend in T_1i between CSE-LPS and NC mice suggests hypoxia in the brain, which is induced by the impaired oxygen uptake as indicated by the reduced f_D.

Conclusion: This study demonstrates the feasibility of using HP ¹²⁹Xe MRI/MRS to study pathological mechanisms of brain dysfunction in comorbidities with COPD.

Purpose: Gadolinium-based contrast media (GBCM) may affect apparent diffusion coefficient measurements on diffusion-weighted imaging. We aimed at investigating the effect of GBCM and inter-reader variation on intravoxel incoherent motion (IVIM) parameters in breast lesions.

Methods: A total of 89 patients referred to 3T breast MRI with at least one histologically verified lesion were included. IVIM data were acquired using a single-shot echo planar imaging sequence before and after GBCM administration. D (true diffusion coefficient), D* (pseudo-diffusion coefficient) and f (perfusion fraction) were calculated and measured by two readers (R1, R2). Inter-reader and intra-reader agreements were assessed by intraclass correlation coefficients (ICCs) and Bland–Altman plots.

Results: D was comparable before and after GBCM administration and between readers. D* and f decreased after GBCM administration and showed a lower agreement between readers. Intra-reader agreement before and after GBCM administration was almost perfect for D for both R1 and R2 (ICC 0.955 and 0.887). The intra-reader agreement was substantial to moderate for D* (ICC R1 0.708, R2 0.583) and moderate for f (ICC R1 0.529 and R2 0.425). Inter-reader agreement before GBCM administration was almost perfect for D (ICC 0.905), substantial for D* (ICC 0.733), and moderate for f (ICC 0.404); after contrast media administration, it was almost perfect for D (ICC 0.876) and substantial for D* (ICC 0.654) and f (ICC 0.606). Bland–Altman plots revealed no significant bias.

Conclusion: Administration of GBCM seems to have a stronger effect on D* and f values than on D values. This should be considered when applying IVIM in clinical practice.

Purpose: To compare the utility of thin-slice fat-suppressed single-shot T2-weighted imaging (T2WI) with deep learning image reconstruction (DLIR) and conventional fast spin-echo T2WI with DLIR for evaluating pancreatic protocol.

Methods: This retrospective study included 42 patients (mean age, 70.2 years) with pancreatic cancer who underwent gadoxetic acid-enhanced MRI. Three fat-suppressed T2WI, including conventional fast-spin echo with 6 mm thickness (FSE 6 mm), single-shot fast-spin echo with 6 mm and 3 mm thickness (SSFSE 6 mm and SSFSE 3 mm), were acquired for each patient. For quantitative analysis, the SNRs of the upper abdominal organs were calculated between images with and without DLIR. The pancreas-to-lesion contrast on DLIR images was also calculated. For qualitative analysis, two abdominal radiologists independently scored the image quality on a 5-point scale in the FSE 6 mm, SSFSE 6 mm, and SSFSE 3 mm with DLIR.

Results: The SNRs significantly improved among the three T2-weighted images with DLIR compared to those without DLIR in all patients (P < 0.001). The pancreas-to-lesion contrast of SSFSE 3 mm was higher than those of the FSE 6 mm (P < 0.001) and tended to be higher than SSFSE 6 mm (P = 0.07). SSFSE 3 mm had the highest image qualities regarding pancreas edge sharpness, pancreatic duct clarity, and overall image quality, followed by SSFSE 6 mm and FSE 6 mm (P < 0.0001).

Conclusion: SSFSE 3 mm with DLIR demonstrated significant improvements in SNRs of the pancreas, pancreas-to-lesion contrast, and image quality more efficiently than did SSFSE 6 mm and FSE 6 mm. Thin-slice fat-suppressed single-shot T2WI with DLIR can be easily implemented for pancreatic MR protocol.

Purpose: Voxel-based morphometry (VBM) is widely used to investigate white matter (WM) atrophy in patients with progressive supranuclear palsy (PSP). In contrast to high-resolution 3D T1-weighted imaging such as magnetization-prepared rapid acquisition with gradient echo (MPRAGE) sequences, the utility of other 3D sequences has not been sufficiently evaluated. This study aimed to assess the feasibility of using a 3D fast low-angle shot sequence captured as a localizer image (L3DFLASH) for VBM analysis of WM atrophy patterns in patients with PSP.

Methods: This retrospective study included 12 patients with pathologically or clinically confirmed PSP, and 18 age- and sex-matched healthy controls scanned with both L3DFLASH and MPRAGE sequences. Image processing was conducted with the Computational Anatomy Toolbox 12 in statistical parametric mapping 12. In addition to the atrophic WM pattern of PSP on VBM, we assessed the WM volume agreement between the two sequences using simple linear regression and Bland–Altman plots.

Results: Despite the slightly larger clusters on MPRAGE, VBM using both sequences showed similar characteristics of PSP-related WM atrophy, including in the midbrain, pons, thalamus, and precentral gyrus. In contrast, VBM showed gray matter (GM) atrophy of the precuneus and right superior parietal lobule exclusively on L3DFLASH. Unlike the measured values of total intracranial volume, GM, and cerebrospinal fluid on MPRAGE, the value of WM was larger on L3DFLASH. In contrast to the total intracranial volume, brainstem, and frontal and occipital lobes, the correlation with WM volume in other regions was relatively low. However, the Bland–Altman plots demonstrated strong agreement, with over 90% of the values falling within the agreement limits.

Conclusion: Both MPRAGE and L3DFLASH are useful for detecting PSP-related WM atrophy using VBM.

Purpose: We developed new deep learning-based hierarchical brain segmentation (DLHBS) method that can segment T1-weighted MR images (T1WI) into 107 brain subregions and calculate the volume of each subregion. This study aimed to evaluate the repeatability and reproducibility of volume estimation using DLHBS and compare them with those of representative brain segmentation tools such as statistical parametric mapping (SPM) and FreeSurfer (FS).

Methods: Hierarchical segmentation using multiple deep learning models was employed to segment brain subregions within a clinically feasible processing time. The T1WI and brain mask pairs in 486 subjects were used as training data for training of the deep learning segmentation models. Training data were generated using a multi-atlas registration-based method. The high quality of training data was confirmed through visual evaluation and manual correction by neuroradiologists. The brain 3D-T1WI scan–rescan data of the 11 healthy subjects were obtained using three MRI scanners for evaluating the repeatability and reproducibility. The volumes of the eight ROIs—including gray matter, white matter, cerebrospinal fluid, hippocampus, orbital gyrus, cerebellum posterior lobe, putamen, and thalamus—obtained using DLHBS, SPM 12 with default settings, and FS with the “recon-all” pipeline. These volumes were then used for evaluation of repeatability and reproducibility.

Results: In the volume measurements, the bilateral thalamus showed higher repeatability with DLHBS compared with SPM. Furthermore, DLHBS demonstrated higher repeatability than FS in across all eight ROIs. Additionally, higher reproducibility was observed with DLHBS in both hemispheres of six ROIs when compared with SPM and in five ROIs compared with FS. The lower repeatability and reproducibility in DLHBS were not observed in any comparisons.

Conclusion: Our results showed that the best performance in both repeatability and reproducibility was found in DLHBS compared with SPM and FS.

Purpose: To develop a new method to generate synthetic MR spectroscopic imaging (MRSI) data for training machine learning models.

Methods: This study targeted routine MRI examination protocols with single voxel spectroscopy (SVS). A novel model derived from pix2pix generative adversarial networks was proposed to generate synthetic MRSI data using MRI and SVS data as inputs. T1- and T2-weighted, SVS, and reference MRSI data were acquired from healthy brains with clinically available sequences. The proposed model was trained to generate synthetic MRSI data. Quantitative evaluation involved the calculation of the mean squared error (MSE) against the reference and metabolite ratio value. The effect of the location of and the number of the SVS data on the quality of the synthetic MRSI data was investigated using the MSE.

Results: The synthetic MRSI data generated from the proposed model were visually closer to the reference. The 95% confidence interval (CI) of the metabolite ratio value of synthetic MRSI data overlapped with the reference for seven of eight metabolite ratios. The MSEs tended to be lower in the same location than in different locations. The MSEs among groups of numbers of SVS data were not significantly different.

Conclusion: A new method was developed to generate MRSI data by integrating MRI and SVS data. Our method can potentially increase the volume of MRSI data training for other machine learning models by adding SVS acquisition to routine MRI examinations.

Purpose: To investigate the visibility of the lenticulostriate arteries (LSAs) in time-of-flight (TOF)-MR angiography (MRA) using compressed sensing (CS)-based deep learning (DL) image reconstruction by comparing its image quality with that obtained by the conventional CS algorithm.

Methods: Five healthy volunteers were included. High-resolution TOF-MRA images with the reduction (R)-factor of 1 were acquired as full-sampling data. Images with R-factors of 2, 4, and 6 were then reconstructed using CS-DL and conventional CS (the combination of CS and sensitivity conceding; CS-SENSE) reconstruction, respectively. In the quantitative assessment, the number of visible LSAs (identified by two radiologists), length of each depicted LSA (evaluated by one radiological technologist), and normalized mean squared error (NMSE) value were assessed. In the qualitative assessment, the overall image quality and the visibility of the peripheral LSA were visually evaluated by two radiologists.

Results: In the quantitative assessment of the DL-CS images, the number of visible LSAs was significantly higher than those obtained with CS-SENSE in the R-factors of 4 and 6 (Reader 1) and in the R-factor of 6 (Reader 2). The length of the depicted LSAs in the DL-CS images was significantly longer in the R-factor 6 compared to the CS-SENSE result. The NMSE value in CS-DL was significantly lower than in CS-SENSE for R-factors of 4 and 6. In the qualitative assessment of DL-CS images, the overall image quality was significantly higher than that obtained with CS-SENSE in the R-factors 4 and 6 (Reader 1) and in the R-factor 4 (Reader 2). The visibility of the peripheral LSA was significantly higher than that shown by CS-SENSE in all R-factors (Reader 1) and in the R-factors 2 and 4 (Reader 2).

Conclusion: CS-DL reconstruction demonstrated preserved image quality for the depiction of LSAs compared to the conventional CS-SENSE when the R-factor is elevated.

Purpose: To assess right heart diastolic energy loss (EL) as a cardiac workload and evaluate its association with major cardiac events (MACE) in adult patients with pulmonary atresia with an intact ventricular septum (PAIVS).

Methods: We retrospectively enrolled and compared 30 consecutive adult patients (18 with PAIVS and 12 with pulmonary stenosis [PS] as controls) who underwent right ventricular (RV) outflow tract reconstruction and 4D flow MRI. EL, conventional parameters on MRI, and the severity of tricuspid regurgitation (TR) on echocardiography were assessed. We also evaluated the association between MACE including arrhythmias, heart failure, surgical intervention, and imaging parameters in adults with PAIVS.

Results: Patients with PAIVS were younger, had a higher diastolic EL/cardiac output (CO) ratio, and had a more significant TR than those with PS (controls). However, RV volume, ejection fraction (EF), and pulmonary regurgitation (PR) severity did not differ between the two groups. Higher RV end-diastolic pressure (EDP) and lower cardiac index (CI) correlated with the diastolic EL/CO in patients with PAIVS. Univariate logistic analysis demonstrated that older age and a higher diastolic EL/CO ratio were important factors for MACE in adults with PAIVS (P = 0.048, 0.049).

Conclusion: A higher diastolic EL/CO ratio was associated with a higher RV EDP and lower CI. A high diastolic EL/CO ratio is also associated with MACE in adults with PAIVS. Even in adults with normal RV volume and EF, the right heart EL was elevated, suggesting an excessive right-sided cardiac workload that integrated both afterload and preload beyond the RV size in adult patients with PAIVS.

Purpose: High b-value acquisition and diffusion-weighted imaging with background suppression (DWIBS) are desirable in high-specificity breast cancer diagnosis on non-contrast-enhanced magnetic resonance imaging; however, this inherently results in a lower signal-to-noise ratio (SNR). Compressed sensitivity encoding (C-SENSE), which combines SENSE with compressed sensing, improves the SNR by reducing noise. Recent technological improvements allow us to incorporate this acceleration technique into echo-planar imaging, called echo-planar imaging with C-SENSE (EPICS). This study aimed to compare image quality and reliability of the apparent diffusion coefficient (ADC) between DWIBS obtained using SENSE and EPICS in patients with small breast cancers.

Methods: Thirty-seven patients with pathologically confirmed breast cancer underwent DWIBS, and images were reconstructed using both conventional SENSE (SENSE-DWIBS) and EPICS (EPICS-DWIBS). Two board-certified radiologists independently evaluated lesion conspicuity (LC) and noise using a 5-point grading scale. The same 2 radiologists independently measured SNR, contrast-to-noise ratio (CNR), and the mean cancer ADC. The Pearson coefficient and Bland–Altman plot were applied to assess the accuracy of ADCs.

Results: LC scores were higher with EPICS than with SENSE, reaching significance for one reviewer but not the other reviewer. Noise ratings on visual evaluation were significantly lower with EPICS than with SENSE (P < 0.001 for both reviewers). SNR was significantly higher with EPICS than with SENSE (P < 0.005 for both reviewers). CNR was significantly higher with EPICS than with SENSE (P < 0.001 for both reviewers). Bland–Altman plots of cancer ADCs using EPICS-DWIBS and SENSE-DWIBS showed excellent concordance, with a bias of 0.026 × 10⁻³ mm²/s and limits of agreement ranging 0.054 × 10⁻³ mm²/s; the Pearson’s correlation coefficient was 0.997 (P < 0.0001).

Conclusion: EPICS enhances breast DWIBS image quality, with improved SNR and CNR and reduced noise levels. The ADCs of breast cancers obtained using EPICS were almost perfectly correlated with those obtained using conventional SENSE.

Purpose: This study aimed to evaluate the feasibility of single-shot echo planar diffusion-weighted imaging with compressed SENSE (EPICS-DWI) for pancreas assessment by comparing with single-shot echo planar DWI with parallel imaging (PI-DWI).

Methods: This multicenter prospective study included 27 consecutive participants with untreated pancreatic ductal adenocarcinoma (PDAC) (15 men; mean age, 67 ± 10 years) who underwent pancreatic protocol MRI including both PI-DWI and EPICS-DWI. Two radiologists independently and randomly reviewed the high b-value DWI images and qualitatively assigned confidence scores for overall image quality, image noise, pancreas conspicuity, and PDAC conspicuity using a 5-point scale. One radiologist measured the PDAC-to-pancreas contrast-to-noise-ratio (CNR) on high b-value DWI images and the apparent diffusion coefficient (ADC) value of PDAC. Qualitative and quantitative parameters were compared between PI-DWI and EPICS-DWI using the Wilcoxon signed-rank test.

Results: The confidence scores for overall image quality (P < 0.001 in both radiologists) and image noise (P < 0.001 in both radiologists) were higher in EPICS-DWI than in PI-DWI. The pancreas conspicuity was better in EPICS-DWI than in PI-DWI in one of the radiologists (P = 0.02 and 0.06). The PDAC conspicuity was comparable between PI-DWI and EPICS-DWI (P > 0.99 in both radiologists). The PDAC-to-pancreas CNR was higher in EPICS-DWI than in PI-DWI (P = 0.02), while the ADC value of PDAC in PI-DWI was not significantly different compared to that in EPICS-DWI (P = 0.48).

Conclusion: The image quality and PDAC-to-pancreas CNR was improved in EPICS-DWI compared to PI-DWI. However, the conspicuity and ADC value of PDAC were comparable between PI-DWI and EPICS-DWI.

Purpose: This study aimed to compare MRI findings among benign, borderline, and malignant ovarian seromucinous neoplasms.

Methods: We retrospectively analyzed MRI data from 24 patients with ovarian seromucinous neoplasms—seven benign, thirteen borderline, and six malignant. The parameters evaluated included age, tumour size, morphology, number, height, apparent diffusion coefficient (ADC) values, T2 ratios, time-intensity curve (TIC) descriptors, and TIC patterns of the mural nodules. Additionally, we examined the T2 and T1 ratios of the cyst contents, tumour markers, and the presence of endometriosis. We used statistical tests, including the Kruskal–Wallis and Fisher–Freeman–Halton exact tests, to compare these parameters among the three aforementioned groups.

Results: The cases showed papillary architecture with internal branching in 57% of benign, 92% of borderline, and 17% of malignant cases. Three or fewer mural nodules were seen in 57% of benign, 8% of borderline, and 17% of malignant cases. Compared to benign and borderline tumours, mural nodules of malignant neoplasms had significantly increased height (P = 0.015 and 0.011, respectively), lower means ADC values (P = 0.003 and 0.035, respectively). The mural nodules in malignant cases also demonstrated significantly lower T2 ratios than those in the benign cases (P = 0.045). Most neoplasms displayed an intermediate-risk TIC pattern, including 80% benign, 83% borderline, and 60% malignant neoplasms, and no significant differences were observed.

Conclusion: Most benign and borderline tumours exhibited a papillary architecture with an internal branching pattern, whereas this feature was less common in malignant neoplasms. Additionally, benign tumours had fewer mural nodules compared to borderline tumours. Malignant neoplasms were characterized by mural nodules with increased height and lower ADC values than those in benign and borderline tumours. Interestingly, all three groups predominantly exhibited an intermediate-risk TIC pattern, emphasizing the complexity of diagnosing seromucinous neoplasms using MRI.

Purpose: Fresh blood imaging (FBI) utilizes physiological blood signal differences between diastole and systole, causing a long acquisition time. The purpose of this study is to develop a fast FBI technique using a centric k_y – k_z k-space trajectory (cFBI) and an exponential refocusing flip angle (eFA) scheme with fast longitudinal restoration.

Methods: This study was performed on 8 healthy subjects and 2 patients (peripheral artery disease and vascular disease) with informed consent, using a clinical 3-Tesla MRI scanner. A numeric simulation using extended phase graph (EPG) and phantom studies of eFA were carried out to investigate the restoration of longitudinal signal by lowering refocusing flip angles in later echoes. cFBI was then acquired on healthy subjects at the popliteal artery station to assess the effect of varying high/low flip ratios on the longitudinal restoration effects. In addition, trigger-delays of cFBI were optimized owing to the long acquisition window in zigzag centric k_y – k_z k-space trajectory. After optimizations, cFBI images were compared against standard FBI (sFBI) images in terms of scan time, motion artifacts, Nyquist N/2 artifacts, blurring, and overall image quality. We also performed two-way repeated measures analysis of variance.

Results: cFBI with eFA achieved nearly a 50% scan time reduction compared to sFBI. The high/low flip angle of 180/2 degrees with lower refocusing pulses shows fast longitudinal restoration with the highest blood signals, yet also more sensitive to the background signals. Overall, 180/30 degrees images show reasonable blood signal recovery while minimizing the background signal artifacts. After the trigger delay optimization, maximum intensity projection image of cFBI after systole-diastole subtraction demonstrates less motion and N/2 artifacts than that of sFBI.

Conclusion: Together with eFA for fast longitudinal signal restoration, the proposed cFBI technique achieved a 2-fold reduction in scan time and improved image quality without major artifacts.

Purpose: To assess the utility of apparent diffusion coefficient maps (ADC) for diagnosing myometrial invasion (MI) in endometrial cancer (EC).

Methods: This retrospective study included 164 patients (mean age, 56 years; range, 25–89 years) who underwent preoperative MRI for EC with <1/2 MI or no MI between April 2016 and July 2023. Five sequences were evaluated: T2-weighted imaging (T2WI), diffusion-weighted imaging (DWI), ADC, dynamic contrast-enhanced T1-weighted imaging (DCE-T1WI), and contrast-enhanced T1WI (CE-T1WI). Three experienced radiologists independently assessed the sequences for MI. For ADC, MI was determined if the endometrial-myometrial junction-tumor boundary had disappeared. Additionally, the assessment of MI was performed using the combination of T2WI, DWI, and ADC, as well as T2WI, DCE-T1WI, and CE-T1WI. The sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve (AUC) for the presence of MI were calculated and compared between the sequences and combinations. Inter-reader agreement was assessed using kappa (κ) statistics.

Results: The sensitivity of ADC was significantly higher than T2WI (P < 0.001) and DCE-T1WI (P = 0.018) for one reader and significantly higher than CE-T1WI (P = 0.045 and 0.043) for two readers. The specificity of ADC was significantly lower than T2WI (P = 0.015 and < 0.001) and CE-T1WI (P = 0.031 and 0.01) for two readers and significantly lower than DCE-T1WI (P = 0.031) for one reader. The AUC of ADC was significantly higher than T2WI (P = 0.048) and DCE-T1WI (P = 0.049) for one reader. The combination including ADC showed higher positive predictive value for all three readers compared to any sequence or combination including contrast enhancement. Additionally, ADC demonstrated the highest agreement rates.

Conclusion: ADC had high sensitivity for MI and the highest agreement rate among all sequences. Thus, this sequence, combined with other sequences, can be crucial for a comprehensive evaluation of MI.

Purpose: T₂ values are hypothesized to be reduced where protein accumulates in the cerebrospinal fluid (CSF). We aimed to verify the accuracy of Carr-Purcell-Meiboom-Gil (CPMG) pulses and non-negative least squares (NNLS) analysis in visualizing protein concentrations by mapping the T₂ values.

Methods: We first dissolved 1.2g of bovine serum albumin powder in 4 mL of artificial CSF to purify an albumin solution with a concentration of 4.5 mM. Artificial CSF was added thereto, and eight types of albumin solutions, with concentrations of 0.002–4.5 mM, were purified. We acquired this albumin solution with CPMG pulses and NNLS, decomposed the T₂ values per pixel, and derived 25 T₂ component values of 60–2000 ms. We assessed the change of T₂ values by the difference in albumin concentration of a single voxel. Finally, we used the method to assess T₂ values from two patients, one with a subdural hematoma and one with a suprasellar cystic tumor. T₂ component values were plotted graphically, presented individually, and created in color maps.

Results: T₂ component values for albumin concentrations ranging from 0.056 to 4.55 mM showed different T₂ peaks, whereas, for concentrations 0.002 to 0.019 mM, the peaks were similar heights and overlapped. Peak width was similar for all concentrations. The color maps successfully reflected the changes in T₂ values across both RGB color patterns. T₂ components for albumin samples with 2.5 mM and 6.1 mM concentrations within a single voxel were represented separately and reflected the ratio of the two samples in nine different regions of interest within one slice. In the clinical cases, the T₂ component map imaged differences in albumin concentrations, similar to those observed in the albumin samples.

Conclusion: The present method with CPMG sequences and NNLS provide adequate images to differentiate accumulating protein concentrations in the CSF, even at the level of a single pixel.

Purpose: To evaluate the clinical value of early renal changes in type 2 diabetes mellitus (T2DM) using multiparameter MRI.

Methods: The study included 41 diabetics (normoalbuminuria: n = 23; microalbuminuria: n = 18) and 30 healthy controls. All subjects underwent intravoxel incoherent motion diffusion-weighted imaging (IVIM), blood oxygen level dependent (BOLD) and arterial spin labeling (ASL) examinations. One-way analysis of variance was used to compare MRI parameters among the three groups. Pearson correlation analysis was used to evaluate the relationship between MRI parameters and estimated glomerular filtration rate (eGFR) and albumin–creatinine ratio (ACR). Receiver operating characteristic analysis was performed to assess the diagnostic performance.

Results: There were statistical differences in cortical D, D*, f, renal blood flow (RBF) and medulla D, D*, f, R2* among the three groups (P < 0.05). The cortical or medullary D, cortical f, and RBF were significantly positively correlated with eGFR (all P < 0.01). The cortical or medullary D, D*, f, cortical RBF were negatively correlated with ACR (all P < 0.05).

To evaluate early kidney changes and degree of diabetes, cortical combined D and RBF (AUC [area under the curve]  = 0.796 and 0.947, respectively) was better than single D or RBF (all P > 0.05); medullary combined D and R2* (AUC = 0.899 and 0.923, respectively) was better than single D or R2* (all P > 0.05), except single D (P = 0.005) in differentiating normoalbuminuria group from control group.

Conclusion: The early changes of renal diffusion and perfusion, oxygenation level, and blood flow in T2DM could be evaluated noninvasively and quantitatively using IVIM, BOLD and ASL. Renal medullary combined IVIM-derived D and BOLD-derived R2* and cortical combined IVIM-derived D and ASL-derived RBF were better for evaluating early renal changes in T2DM.

Purpose: The aim of our study was to investigate the technical accuracy of susceptibility-weighted imaging (SWI) and quantitative susceptibility mapping (QSM) created to detect intramammary-like calcifications depending on different TEs, volume, and type of calcification samples at 1.5T.

Methods: Jello-embedded particles of blackboard chalk and ostrich eggshell ranging in size from 4 to 25 mm² were used to simulate intramammary calcifications after testing different base substances and calcifications for their suitability to be used in breast phantoms. Breast phantoms were systematically examined using CT and an optimized 3D multi-echo gradient echo pulse sequence with following parameters: TR/TE, 22/1.88–15.52 ms in 1.24 ms increments; reconstructed voxel, 0.5 × 0.5 × 1.1 mm³; receiver bandwidth, 1120 Hz/Px; flip angle, 15°; integrated parallel imaging technique with a GeneRalized Autocalibrating Partial Parallel Acquisition (GRAPPA) factor of 2/24; and a total acquisition time of 3:00 min. A qualitative evaluation of the dependence of the visualization of calcification samples on volume and TE value was followed by a calculation of the SNR, the contrast-to-noise ratio (CNR) and the creation of SWI and QSM in the sense of a (semi)-quantitative analysis of the images.

Results: Jello proved to be a suitable base substance for preparing breast phantoms for SW MRI. Blackboard chalk and ostrich eggshell proved to be suitable for mimicking intramammary-like calcifications. The decrease in the median SNR of the blackboard chalk samples was significantly higher than the corresponding value of the ostrich eggshell samples over the entire TE range (47.5 to 17.0 vs. 16.0 to 6.56, P < 0.0001). The increase in the median CNR of the blackboard chalk samples was significantly higher than the corresponding value of the ostrich eggshell samples over the entire TE range (2.46 to 35.0 vs. 20.2 to 36.8, P = 0.007). With increasing TE value, the signal void volume of the calcification particle increases in the magnitude images as well as in SWI and QSM. Due to the blooming effect, the median gradients of the TE-based changes in signal void volumes were higher in SWI than in magnitude images and in QSM, regardless of the type of calcification particle examined. The maximum magnetic susceptibility of ostrich eggshell samples varied in a TE range of 1.88 to 15.52 ms from −7.2 to −2.51 ppm and that of blackboard chalk from −2.0 to −1.7 ppm. Compared to the manually measured volumes of the calcification particles, both MR-based measurements and CT examinations overestimated the actual sample size. The (non)-significant overestimation in the MRI-data is dependent on the set TE. The CT-based hyperdense volumes were overestimated compared to the corresponding manually measured sample volumes in a range of 109.8%–315.2% for ostrich eggshell samples (P = 0.016) and in a range of 39.9%–156.4% for blackboard chalk samples (P = 0.69).

Conclusion: Our systematic in-vitro investigation of magnitude images, SWI, and QSM revealed that various set TE values, different volumes, and compositions of calcifications have a significant impact on visualizing intramammary(-like) calcifications.

Purpose: This study investigated the ability of three-dimentional motion-sensitized driven-equilibrium prepared T₁-weighted fast spin echo (3D MSDE-FSE) imaging to identify distal dural rings (DDRs) and paraclinoid aneurysms (ParaC-ANs) and differentiate between intradural and extradural ParaC-ANs and compared it with that of established MR cisternography-based techniques.

Methods: 3D MSDE-FSE images were acquired along with fast imaging employing steady state acquisition (FIESTA), and time-of-flight magnetic resonance angiography (TOF-MRA) on a 3T MRI system in 53 patients with unruptured and untreated ParaC-ANs. Two radiologists applied a 3-point scale to rate the clarity with which the DDR (53 left and 53 right) and ParaC-ANs (total of 55) were depicted in the 3D MSDE-FSE and FIESTA images. The clarity scores, which were determined by averaging the scores of the 2 assessors, on the 3D MSDE-FSE and FIESTA images were compared using the Wilcoxon signed-rank test. Furthermore, the same radiologists classified the ParaC-ANs as intradural, extradural, or transitional on 3D MSDE-FSE images. A third radiologist independently classified the ParaC-ANs as intradural, extradural, or transitional based on the FIESTA and MRA fusion images. The kappa coefficient was used to compare this classification with that based on 3D MSDE-FSE images.

Results: The Wilcoxon signed-rank test revealed no significant difference between 3D MSDE-FSE images and FIESTA images in the scores for the clarity of depiction of the DDRs (P = 0.119). However, the scores for the clarity of the depiction of the ParaC-ANs were significantly greater for the 3D MSDE-FSE images than for the FIESTA images (P < 0.001). The kappa coefficient for comparison of classification based on 3D MSDE-FSE images and FIESTA and MRA fusion images was 0.82.

Conclusion: 3D MSDE-FSE imaging has the potential to differentiate between intradural and extradural ParaC-ANs by directly recognizing the DDR.

Purpose: The purposes of this study were 1) to improve vessel visibility of our MR sequence by modifying k-space filling and 2) to verify the usefulness of applying artificial intelligence (AI) for volume isotropic simultaneous interleaved bright- and black-blood examination (VISIBLE) with compressed sensitivity encoding (CS) in autodetecting brain metastases.

Methods: We modified 3 sequences of VISIBLE (Centric, Reversed Centric, and Startup Echo 30). The Centric sequence is a prototype. The Reversed Centric filled the k-space in a reversed centric manner to improve vessel visibility. The Startup Echo 30 implemented dummy echoes to further improve vessel visibility. Vessel visibility was evaluated in one slice at the level of the centrum semiovale. The sensitivity, specificity, the area under the curve (AUC), and false positives of detecting brain metastases using AI were evaluated among 3 sequences. Statistical comparisons were performed using a one-way analysis of variance, followed by Friedman and Dunn’s multiple comparison tests.

Results: The number of visualized vessels was significantly lower in the Centric (39.3 ± 9.7, P < 0.05) and Reversed Centric (44.2 ± 9.8, P < 0.05) methods than in the magnetization-prepared rapid gradient echo (49.3 ± 9.1) but comparable in the Startup Echo 30 method (44.9 ± 8.8, P > 0.05). No significant differences existed in sensitivity, specificity, and AUC among the 3 methods. False positives achieved using the Reversed Centric method were significantly fewer (54 false positives) than those achieved using the Centric (85 false positives) and Startup Echo 30 (68 false positives) methods (P = 0.0092).

Conclusion: Vessel visibility was improved by modifying the k-space filling, which may reduce false positives. The AI model for VISIBLE with CS achieved good performance in autodetection of brain metastases. The AI model for VISIBLE with CS can help radiologists diagnose brain metastases in clinical practice.

Purpose: To re-evaluate images recovered from JCOG0911, a randomized phase 2 trial for newly diagnosed glioblastoma (nGBM) conducted by the Japan Clinical Oncology Group (JCOG) Brain Tumor Study Group.

Methods: The correlation between tumor volumes and survival was evaluated, followed by progression-free survival (PFS) analysis by independent central review based on Response Assessment in Neuro-Oncology (RANO) criteria using MRI recovered from 118 nGBM patients enrolled in the JCOG0911 trial. A radiomic analysis was also performed to identify radiomic features predictive of nGBM prognosis.

Results: The distribution of the Gd-enhancing and T2-weighted image/fluid attenuated inversion recovery-high intensity lesions mainly occupied white matter. JCOG0911 consisted of more subjects with right-sided lesions. The median extent of resection of the Gd-enhancing lesions was 99%. The overall survival showed a nonsignificant negative trend with postoperative Gd-enhancing lesion volume (P = 0.22), with the hazard ratio increasing in parallel with its volume. The median PFS after registration was 302 and 308 days for local Response Evaluation Criteria in Solid Tumors (RECIST)-based and central RANO-based diagnoses. However, an apparent discrepancy was observed between the two in the early phase, presumably due to the misdiagnosis of pseudoprogression by local RECIST-based diagnosis. Radiomic analysis identified 28 radiomic features predictive of nGBM prognosis, 5 of which were those previously identified in a separate cohort. The constructed radiomics-based prognostic model stratified the cohort into high- and low-risk groups (P = 0.028).

Conclusion: Novel analytical methods that could be incorporated into future clinical trials were successfully tested. RANO and RECIST may not differ in progression calls if pseudoprogression is appropriately handled.

Purpose: A cyst-like structure near superior sagittal sinus (Arachnoid Cuff Exit Site cysts: ACES cysts) has been reported in MRI. The purpose of this study was to investigate the association between presence of ACES cysts and cognitive function, as assessed using mini-mental state examination (MMSE) scores.

Methods: We retrospectively analyzed patients who underwent head MRI for dementia screening. Differences in patient ages and MMSE scores between patients with and without ACES cysts were examined using the Mann-Whitney U test. Correlations between patient ages and MMSE scores were examined for patients with and without ACES cysts using Spearman’s rank correlation coefficient. Multivariate logistic regression analysis was performed to examine the influence of presence or absence of ACES cysts on MMSE score.

Results: A total of 112 patients (male: 28, female: 84) were included for the analysis. The patient ages ranged from 66 to 94 years (median: 83 years). MMSE scores ranged from 6 to 30 (median: 24). ACES cysts were detected in 57 patients (50.9%). There was no significant difference in patient ages between the patients with and without ACES cysts (P = 0.058). The patients with ACES cysts showed significantly lower MMSE scores compared to the patients without ACES cysts (P < 0.001). In the patients with ACES cysts, there was no significant correlation between patient ages and MMSE scores (ρ = −0.178, P = 0.185), whereas a significant negative correlation was observed in the patients without ACES cysts (ρ = −0.347, P = 0.001). The presence of ACES cysts was determined as an independent predictor for the lower MMSE score (odds ratio = 15.2, 95% confidence interval = 5.59–41.4, P < 0.001).

Conclusion: The presence of the ACES cysts showed significant association with lower MMSE score. ACES cysts might be involved in the pathological processes affecting cognitive function.

Purpose: This study aims to assess thalamocortical tract fiber injury using diffusion-tensor imaging (DTI) and to characterize metabolic alterations in the dorsal thalamus with proton magnetic resonance spectroscopy (MRS) in patients with cervical spondylotic myelopathy (CSM).

Methods: A prospective study involved 98 CSM patients and 66 age-matched controls without neurological disease, recruited from May 2021 to December 2021. Neurological function was evaluated using the modified Japanese Orthopaedic Association (mJOA) score. DTI and MRS were analyzed by 2 experienced radiologists, with statistical analysis performed via SPSS (v.26), including t-tests, chi-square tests, and Pearson’s correlation.

Results: DTI revealed significantly lower fractional anisotropy (FA) and higher radial and axial diffusivity in the bilateral postcentral gyrus of CSM patients compared to controls (P < 0.001). MRS showed decreased ratios of N-acetylaspartate (NAA)/creatinine (Cr), choline (Cho)/Cr, and myo-inositol (mI)/Cr in the dorsal thalamus of CSM patients (P < 0.001). Correlation analysis indicated a moderate association between mJOA scores and NAA/Cr and mI/Cr ratios, and a weaker correlation with FA and Cho/Cr.

Conclusion: CSM patients exhibit significant thalamocortical disruptions and metabolic changes in the dorsal thalamus, correlating with clinical severity. These findings suggest that DTI and MRS could provide valuable insights into the extent of neural damage in CSM.

Purpose: To investigate the potential of 4D flow MRI-derived pulmonary hemodynamic parameters as sensitive markers for chronic obstructive pulmonary disease (COPD) patients with right ventricular dysfunction (RVD).

Methods: We enrolled 15 COPD patients combined with RVD and 43 non-RVD participants, all of them underwent pulmonary function tests, thoracic CT and cardiac MR examinations, and the image post-processing analysis was completed. After comparing the 2 groups, the average flow velocity of the main pulmonary artery (Vavg-MPA) and the right pulmonary artery (Vavg-RPA) were identified as statistically significant confounding factors, propensity score matching was used to pair patients controlling for these 2 parameters. Univariate and multivariate logistic regression analyses were performed to assess the pulmonary hemodynamic parameters obtained from 4D flow MRI that could serve as sensitive markers for identifying COPD patients with RVD based on the matched participants dataset.

Results: Fourteen COPD patients combined with RVD and 29 non-RVD participants were successfully matched. Logistic regression analysis showed that the decreased systolic pressure drop along the MRA-RPA tract (odds ratio [OR]: 0.31; 95% confidence interval [CI]: 0.12–0.78; P =0.013) and the presence of vortex (OR: 8.82; 95% CI: 1.11–70.36; P =0.040) were identified as independent risk factors for RVD in COPD patients.

Conclusion: Pulmonary hemodynamic parameters derived from 4D flow MRI, specifically the systolic pressure drop along the MPA-RPA tract and the presence of vortex in the main pulmonary artery, can serve as sensitive indicators for predicting right ventricular dysfunction in COPD patients.

Purpose: To compare diagnostic reliability between an intravoxel incoherent motion (IVIM) imaging and an arterial spin labeling (ASL) in assessment of renal blood flow in rodents.

Methods: We first evaluated 3 different fitting methods on 5 datasets of diffusion-weighted imaging (DWI) with 10 b-values (0–1000 s/mm²) for bi-exponential analysis in IVIM imaging to calculate pseudo-diffusion parameters. Coefficient of variation (CV) for each parameter and correlation among the parameters was assessed to test the robustness of the 3 fitting methods. Subsequently, DWI and ASL methods were performed before and 14 days after onset of acute kidney injury (AKI) in a rat model. Temporal change before and after AKI onset in the pseudo-diffusion parameters in 3 fitting methods was compared with that in the renal blood flow (RBF) derived in the ASL method.

Results: The CVs in all IVIM parameters were the lowest in the fitting method that estimated pseudo-diffusion parameters after a fixed true-diffusion was determined where the pseudo- and true-diffusion coefficients had no correlation. The RBF substantially reduced (~50%, P < 0.001) due to the AKI onset; however, no pseudo-diffusion parameters in any of 3 fitting methods could not detect the change. Further, any pseudo-diffusion parameters showed no correlation with the RBF.

Conclusion: Pseudo-diffusion parameters in the IVIM concept were not reliable to estimate RBF in the study. Since the kidney has a unique profile in the “tissue flow”, our data indicate that study design and interpretation of results needs to be carefully considered when IVIM imaging is used for evaluation of blood flow in tissue, especially in the kidney.

Purpose: To assess the institutional variability in ultrafast dynamic contrast-enhanced (UF-DCE) breast MRI using time-resolved angiography with stochastic trajectories (TWIST)-volumetric interpolated breath-hold examination (VIBE) and compressed sensing (CS)-VIBE sequences acquired at 2 different institutions with different patient populations and contrast injection protocols.

Methods: UF-DCE MR images of 18 patients from site A acquired using a TWIST-VIBE sequence, and UF-DCE MR images of 18 patients from site B acquired with a CS-VIBE sequence, were retrospectively evaluated and compared. The 2-site patient cohort was matched for patient age, background parenchymal enhancement, malignancy or benignity, and lesion size. Qualitative assessments included noise, blurring, poor fat suppression, aliasing artifact, motion artifact, lesion conspicuity, lesion morphology, time-intensity-curve smoothness, and vessel delineation. For quantitative assessment, the bolus arrival time was evaluated for each lesion, and its diagnostic performance in discriminating between benign and malignant lesions was examined using receiver operating characteristics analysis.

Results: Thirteen malignant and five benign lesions were included from each site. Qualitative evaluation revealed that poor fat suppression and aliasing artifacts were visible in images from site A with TWIST-VIBE (P = 0.004 and P < 0.001), whereas motion artifacts were present in images from site B with CS-VIBE (P = 0.04). Lesion morphology assessments (P < 0.001) and vessel delineation (P < 0.001) were superior for images from site B with CS-VIBE. Bolus arrival time was significantly longer with TWIST-VIBE than with CS-VIBE, for both benign and malignant lesions (P < 0.001). The area under the receiver operating characteristics curve was 0.55 for site A and 0.69 for site B (P = 0.39).

Conclusion: Both acquisitions allowed evaluation of breast lesions with good lesion conspicuity and time-intensity-curve smoothness, whereas CS-VIBE was superior to TWIST-VIBE for morphological evaluation of breast lesions and depiction of blood vessels in the breast. Injection rate appears to have a significant impact on semi-quantitative parameters derived from UF-DCE MRI.

Purpose: Hyperthermia is a treatment that applies heat to damage or kill cancer cells and can be also used for drug deliveries. It is important to apply the heat into the specific area in order to target the cancer tissue and avoid damaging healthy tissue. For this reason, the development of heat applicators that have the capability to deliver the heat to the target area is vital. In this study, we present an optimization of an array coil for brain hyperthermia that can be used in combination with MRI, such that the heat can be delivered to the cancer area.

Methods: The array coils were based on optimizing loop coils by varying the capacitor’s position along the perimeter. The optimization was performed using electromagnetic simulations, by computing the electric field (E) and temperature inside of the brain and targeting tumor tissues for focus temperature application. The coils were compared with a general–use symmetric coil array for head heating.

Results: The optimization of the coil array was able to focus electric field and make temperature rise at the cancer areas. The temperature in Tumor 1 before and after standard and the proposed method optimization was 43.6°C, 48.3°C, and 42.5°C and for Tumor 2 the temperatures were 44.2°C, 43.1°C, and 42.9°C, respectively. Although the standard optimization method exhibits higher temperatures, it also had higher temperatures outside the tumors area.

Conclusion: We demonstrated the optimization of array coils with different capacitor positions to obtain focused heating temperatures.

Purpose: The purpose of this study was to assess the extent to which improvement in sleep with lemborexant contributed to changes in interstitial fluid dynamics.

Methods: The 3 methods including diffusion tensor image analysis along the perivascular space (DTI-ALPS), dynamic contrast-enhanced method to assess tissue vascular permeability (Ktrans), and choroid plexus volume (CPV) were used. Correlations between these imaging indices and sleep parameters (latency to persistent sleep [LPS], wake after sleep onset [WASO], total sleep time [TST], and sleep efficiency [SE]) were evaluated using Pearson correlation analysis. Additionally, multiple regression analysis and linear mixed model analysis were employed to assess the relationship between baseline sleep status and imaging parameter changes. MRI and sleep assessments were performed before treatment initiation (week 0, w0) and at 12 weeks after lemborexant administration (week 12, w12).

Results: The ALPS-index was inversely correlated with LPS and positively correlated with TST and SE at w0. In multiple regression analysis, ALPS-index was lower when sleep parameters other than LPS were poor at w0. A linear mixed model analysis suggested that poor sleep status in LPS and SE at w0 may have an effect on greater ALPS-index. In the evaluation of Ktrans measurement, the single regression analysis showed a statistically significant correlation between the reduction in Ktrans and the shortening in LPS. Examination of CPV and sleep parameters showed a significant negative correlation between TST and CPV at w0 and w12. Multiple regression analysis also showed that TST of w12 had a significant effect on CPV at w12.

Conclusion: Our results suggested that poor sleep status is related to the greater change of ALPS-index and CPV improvement after lemborexant administration may be related to in part to sleep parameter improvement.

Purpose: To reduce motion artifacts (MA) and noise in brain MRI using deep learning to promote clinical utility.

Methods: T1-weighted (T1W), T2-weighted (T2W), and fluid attenuated inversion recovery (FLAIR) images of the brain (including sagittal, coronal, and axial sections) of 20 healthy volunteers were collected using a 3.0T MR system. Simulated images with horizontal and vertical phase directions exhibiting varying white noise and MA (n = 115200) were created for each sequence and trained in deep learning (36000 pairs), validation (200 pairs), and testing (200 pairs, 2000 pairs) datasets. Images with MA and noise and images without MA and noise were included. A training model was constructed to remove noise and MA. The model’s ability to remove noise and MA was evaluated by the structural similarity index (SSIM) and peek signal to noise ratio (PSNR). The SSIM and PSNR between the ground-truth and output images were calculated (SSIMout, PSNRout), and the SSIM and PSNR between the ground-truth and input images were calculated (SSIMinp, PSNRinp). The ratio of SSIMinp to SSIMout was then evaluated as the improvement ratio of SSIM (IMPRs) and the ratio of PSNRinp to PSNRout as the improvement ratio of PSNR (IMPRp). In addition, 10 radio technologists performed visual evaluation.

Results: The SSIMout were >0.95 and 33 dB, respectively, for T1W, T2W, and FLAIR images with different contrasts. The mean value of SSIMinp was 0.72. Noise and MA removal effects were observed in images, with an average value of 72 dB. Visual evaluation revealed that the reduction effect in the output image was higher than that observed in the input image.

Conclusion: The method proposed herein, which uses separate training models for the T1W, T2W, and FLAIR sequences, is a valuable approach for removing MA and noise, independent of the imaging direction and artifact orientation. An improvement in image quality was achieved by generating a large amount of training data through simulation.

Purpose: Recently, a novel deep learning (DL)-based technique for reconstructing highly undersampled MR data (DL-Speed, DLS) has been developed, which demonstrated superior performance over compressed sensing. This study aimed to achieve high-resolution double inversion recovery (DIR) imaging using DLS (DLS-DIR) and compare its diagnostic performance in the detection of juxtacortical multiple sclerosis (MS) lesions with that of conventional DIR (C-DIR).

Methods: We retrospectively analyzed MRI data from 25 patients with MS who underwent a comprehensive imaging protocol, including 3D fluid-attenuated inversion recovery (FLAIR), C-DIR, and DLS-DIR. A voxel size of 1.3 × 1.3 × 1.4 mm³ with a scan duration of 3 mins 55s were used for C-DIR, and isotropic 0.7 mm voxels with a scan time of 4 mins 23s were employed for DLS-DIR. Two neuroradiologists assessed the juxtacortical MS lesions during 2 separate reading sessions (one with C-DIR and the other with DLS-DIR). Lesions were categorized as subcortical white matter lesions, intracortical lesions, or mixed lesions involving both subcortical white and gray matter. The lesion counts per region were compared between the imaging techniques using the Wilcoxon signed-rank test.

Results: DLS-DIR detected a significantly higher number of juxtacortical MS lesions compared to C-DIR (Radiologist A: 211 lesions vs. 164 lesions; Radiologist B: 209 lesions vs. 157 lesions, P < 0.05). DLS-DIR also identified more intracortical lesions (Radiologist A: 22 additional lesions, Radiologist B: 34 additional lesions, P < 0.05) and more mixed lesions (Radiologist A: 46 additional lesions, Radiologist B: 42 additional lesions, P < 0.05).

Conclusion: The DLS technology enables high-resolution, whole-brain DLS-DIR imaging within a 5 mins acquisition time, which can be seamlessly incorporated into routine clinical workflows. This approach substantially enhances the detection and evaluation of juxtacortical MS lesions.

Purpose: This study aimed to quantitatively evaluate whether non-echoplanar diffusion-weighted MRI (non-EP DWI) lesion size and detectability can predict mastoid extension and complications such as labyrinthine fistula and dural exposure in middle ear cholesteatoma.

Methods: This retrospective study included 120 lesions with surgically confirmed middle ear cholesteatoma. Non-EP DWI was performed within 6 months preoperatively and evaluated for lesion detectability and size measurements, including maximum axial diameter, maximum axial area, and volume. Surgical findings were used to assess mastoid extension, labyrinthine fistula, and dural exposure.

Results: Of the 120 lesions, 30 were undetectable and 90 were detectable on non-EP DWI. Undetectable lesions had significantly less mastoid extension or labyrinthine fistula compared to detectable lesions (P < 0.001 − P = 0.006). The undetectable finding on non-EP DWI for identifying mastoid extension-negative lesions showed a sensitivity of 0.59, specificity of 0.95, and for labyrinthine fistula-negative lesions showed a sensitivity of 0.29, specificity of 1.00. Among the 90 detectable lesions, all size parameters (maximum axial diameter, maximum axial area, and volume) were significantly larger in cases with positive mastoid extension and positive labyrinthine fistula compared to negative cases (P < 0.001 − P = 0.005). For dural exposure, the maximum axial diameter and maximum axial area were significantly larger in positive cases (P = 0.002), but volume did not differ significantly. Optimal diagnostic cut-off values were determined for mastoid extension (8.9 mm diameter and 56 mm² area, both with sensitivity 0.89 and specificity 0.97), labyrinthine fistula (82 mm² area, sensitivity 0.47, specificity 1.00), and dural exposure (14.3 mm diameter, sensitivity 0.59, specificity 0.87; 112 mm² area, sensitivity 0.68, specificity 0.73).

Conclusion: Non-EP DWI lesion size and detectability can predict important operative findings in middle ear cholesteatoma. Undetectable lesions on non-EP DWI indicate a lack of mastoid extension or labyrinthine fistula, while larger detectable lesions correlate with increased risks of extension and complications.

Purpose: To investigate whether significant differences exist in labyrinthine lymph fluid signal intensities on non-contrast-enhanced 3D real inversion recovery (3D-real IR) images between patients with and without significant endolymphatic hydrops (EH), potentially enabling the non-contrast detection of EH.

Methods: Thirty-nine patients suspected of having EH underwent 3D-real IR MRI before and 4 hours after intravenous administration of a single dose of gadobutrol. Signal intensities of the cerebrospinal fluid (CSF), perilymph, and endolymph were manually measured on pre-contrast images using ROIs. Patients were categorized into 2 groups based on post-contrast imaging: those with significant EH and those without. Normalized signal intensities (nSIs) of the endolymph were calculated and compared between the groups using the Mann–Whitney U test.

Results: The nSIs of the vestibular endolymph on non-contrast 3D-real IR images were significantly lower in the group with significant EH compared to the group without EH (P  < 0.05), suggesting T1 prolongation in the vestibular endolymph of patients with significant EH. However, considerable overlap was observed in the nSIs between the 2 groups, and significant EH did not always result in T1 prolongation. No significant differences were found in the nSIs of the perilymph or CSF between the groups.

Conclusion: The study suggests that T1 prolongation occurs in the vestibular endolymph in cases of significant EH. This finding indicates the potential for developing non-contrast MRI methods to detect EH and underscores the importance of understanding the mechanisms behind T1 changes in the endolymph. Further research with larger patient cohorts and inclusion of healthy control subjects is necessary to validate these results and to elucidate the pathophysiology of EH in Ménière’s disease.

Purpose: This study aims to evaluate the predictive value of MRI features for pathologic complete response (pCR) and survival outcomes in patients with breast cancer (BC) undergoing neoadjuvant chemotherapy (NAC).

Methods: A retrospective analysis was conducted on 168 BC patients treated with NAC between 2018 and 2022. Pre-NAC breast MRI scans were evaluated for enhancement patterns, time-intensity curve (TIC), peritumoral edema, and background enhancement. Both pre- and post-NAC MRIs were assessed for Epeak %, mean apparent diffusion coefficient (ADC) value, and ADC ratio (mean ADC of lesion/contralateral normal breast parenchyma). Survival outcomes were analyzed using Kaplan−Meier and Cox regression models.

Results: pCR was achieved in 34% of patients. MRI demonstrated a sensitivity of 74% and a specificity of 86% in predicting pCR, with an overall accuracy of 82%. The post-NAC percentage of initial peak enhancement (Epeak) was significantly lower in the pCR group (P < 0.001). Multivariate analysis identified a pre-NAC Epeak ≤ 96 (hazard ratio [HR]: 6.26, P < 0.001) and a post-NAC Epeak > 188 (HR: 18.40, P < 0.001) as independent risk factors for disease-free survival. Additionally, a lower pre-NAC ADC ratio (≤0.65) was associated with poorer overall survival (HR: 2.8, P: 0.041). Pre-NAC peritumoral edema, background enhancement, and TIC were not significant predictors of survival outcomes.

Conclusion: MRI features, including Epeak % and ADC ratio, are important predictors of pCR and survival outcomes in BC patients undergoing NAC. Incorporating these biomarkers into clinical practice may improve treatment planning and optimize patient outcomes.

Purpose: Glymphatic system impairment has been suggested in previous animal and human studies regarding traumatic brain injury (TBI). Diffuse axonal injury (DAI) is an important pathological feature of TBI and is frequently diagnosed in patients with moderate to severe TBI. This study evaluated the glymphatic system function in patients with DAI using diffusion tensor imaging analysis along the perivascular space (DTI-ALPS), a non-invasive technique.

Methods: A total of 162 patients with TBI, including 84 with DAI and 78 without DAI, underwent MRI with DTI within 6 months of the date of injury. The ALPS index was calculated to assess the glymphatic system activity and compared between patients with and without DAI. Analysis of covariance (ANCOVA) was used to compare the ALPS index between patients with DAI grades 1, 2, and 3. Correlation analysis was performed between the ALPS index, DAI grade, and Glasgow Coma Scale (GCS) score.

Results: Patients with DAI (1.29 ± 0.17) had a significantly lower ALPS index than those without DAI (1.42 ± 0.19, P < 0.001). The ALPS index differed significantly between patients with different DAI grades (ANCOVA, P < 0.001). The ALPS index and DAI grades were negatively correlated (r =−0.47, P < 0.001). The ALPS index and GCS scores showed a weak positive correlation (r = 0.174, P = 0.027).

Conclusions: Patients with DAI have a lower ALPS index, indicating impaired glymphatic system activity, which is more severe in patients with a higher DAI grade. These findings broaden the understanding of the pathophysiology of DAI and help predict patients’ prognoses and recovery trends.

Purpose: Diffusion models (DMs) excel in pixel-level and spatial tasks and are proven feature extractors for 2D image discriminative tasks when pretrained. However, their capabilities in 3D MRI discriminative tasks remain largely untapped. This study seeks to assess the effectiveness of DMs in this underexplored area.

Methods: We use 59830 T1-weighted MR images (T1WIs) from the extensive, yet unlabeled, UK Biobank dataset. Additionally, we apply 369 T1WIs from the BraTS2020 dataset specifically for brain tumor classification, and 421 T1WIs from the ADNI1 dataset for the diagnosis of Alzheimer’s disease. Firstly, a high-performing denoising diffusion probabilistic model (DDPM) with a U-Net backbone is pretrained on the UK Biobank, then fine-tuned on the BraTS2020 and ADNI1 datasets. Afterward, we assess its feature representation capabilities for discriminative tasks using linear probes. Finally, we accordingly introduce a novel fusion module, named CATS, that enhances the U-Net representations, thereby improving performance on discriminative tasks.

Results: Our DDPM produces synthetic images of high quality that match the distribution of the raw datasets. Subsequent analysis reveals that DDPM features extracted from middle blocks and smaller timesteps are of high quality. Leveraging these features, the CATS module, with just 1.7M additional parameters, achieved average classification scores of 0.7704 and 0.9217 on the BraTS2020 and ADNI1 datasets, demonstrating competitive performance with that of the representations extracted from the transferred DDPM model, as well as the 33.23M parameters ResNet18 trained from scratch.

Conclusion: We have found that pretraining a DM on a large-scale dataset and then fine-tuning it on limited data from discriminative datasets is a viable approach for MRI data. With these well-performing DMs, we show that they excel not just in generation tasks but also as feature extractors when combined with our proposed CATS module.

Purpose: To assess the effects of industry-developed deep learning reconstruction with super resolution (DLR-SR) on single-shot turbo spin-echo (SshTSE) images with thickness of 2 mm with DLR (SshTSE^2mm) relative to those of images with a thickness of 5 mm with DLR (SSshTSE^5mm) in the patients with pancreatic cystic lesions.

Methods: Thirty consecutive patients who underwent abdominal MRI examinations because of pancreatic cystic lesions under observation between June 2024 and July 2024 were enrolled. We qualitatively and quantitatively evaluated the image qualities of SshTSE^2mm and SshTSE^5mm with and without DLR-SR.

Results: The SNRs of the pancreas, spleen, paraspinal muscle, peripancreatic fat, and pancreatic cystic lesions of SshTSE^2mm with and without DLR-SR did not decrease in compared to that of SshTSE^5mm with and without DLR-SR. There were no significant differences in contrast-to-noise ratios (CNRs) of the pancreas-to-cystic lesions and fat between 4 types of images. SshTSE^2mm with DLR-SR had the highest image quality related to pancreas edge sharpness, perceived coarseness pancreatic duct clarity, noise, artifacts, overall image quality, and diagnostic confidence of cystic lesions, followed by SshTSE^2mm without DLR-SR and SshTSE^5mm with and without DLR-SR (P  <  0.0001).

Conclusions: SshTSE^2mm with DLR-SR images had better quality than the other images and did not have decreased SNRs and CNRs. The thin-slice SshTSE with DLR-SR may be feasible and clinically useful for the evaluation of patients with pancreatic cystic lesions.

Purpose: To evaluate the effect of model-based deep-learning reconstruction (DLR) compared with that of compressed sensing–sensitivity encoding (CS) on cine cardiac magnetic resonance (CMR).

Methods: Cine CMR images of 10 healthy volunteers were obtained with reduction factors of 2, 4, 6, and 8 and reconstructed using CS and DLR. The visual image quality scores assessed sharpness, image noise, and artifacts. Left-ventricular (LV) end-diastolic volume (EDV), end-systolic volume (ESV), stroke volume (SV), and ejection fraction (EF) were manually measured. LV global circumferential strain (GCS) was automatically measured using the software. The precision of EDV, ESV, SV, EF, and GCS measurements was compared between CS and DLR using Bland–Altman analysis with full-sampling data as the gold standard.

Results: Compared with CS, DLR significantly improved image quality with reduction factors of 6 and 8. The precision of EDV and ESV with a reduction factor of 8, and GCS with reduction factors of 6 and 8 measurements improved with DLR compared with CS, whereas those of SV and EF measurements were not different between DLR and CS.

Conclusion: The effect of DLR on cine CMR’s image quality and precision in evaluating quantitative volume and strain was equal or superior to that of CS. DLR may replace CS for cine CMR.

Purpose: Aortopathy, characterized by aortic dilatation caused by cystic medial necrosis, typically develops in adulthood but has been observed at a young age in patients with tetralogy of Fallot. We hypothesized that some patients with tetralogy of Fallot or double-outlet right ventricle experience early-onset aortic disturbed flow. This study aimed to identify and analyze the causes of disturbed flow using 4D flow MRI.

Methods: This study included 24 patients who underwent 4D flow MRI at our institution between January 2022 and September 2024. MRI and cardiac catheterization were performed during follow-up. 4D flow MRI was used to detect disturbed flow and investigate its underlying causes.

Results: The mean age of patients with tetralogy of Fallot or double-outlet right ventricle was 12 years (range, 1 to 37 years). Of the 24 participants, 11 (46%) exhibited disturbed flow. Patients with disturbed flow had significantly higher Valsalva Z-scores (4.7 ± 3.1 vs. 2.4 ± 1.4, P = 0.013) and a significantly narrower left ventricular outflow tract-ascending aorta angle (113.5 ± 11.6 vs. 127.1 ±  6.7°, P = 0.002). Wall shear stress and energy loss were not significantly different between the 2 groups.

Conclusion: Aortic disturbed flow may occur in patients with tetralogy of Fallot or double-outlet right ventricle, regardless of age, suggesting an association with the left ventricular outflow tract-ascending aorta angle. The occurrence of disturbed flow at a young age should be noted as it may contribute to the future progression of aortopathy.

Purpose: Acupuncture is believed to significantly modify neural circuits in the brain. However, the effects of moxibustion stimulation remain unclear. Therefore, we used functional MRI to investigate brain activation sites induced by moxibustion stimulation using an electric moxibustion device that mimics Japanese Tonetsu-kyu half-grain-sized direct moxibustion.

Methods: Twenty-two healthy adult participants underwent 6 rounds of 7-s moxibustion stimulations on the right acupuncture point ST36 using electric moxibustion during functional MRI measurement. The maximum output temperature of electric moxibustion was 67.5°C. However, the contact surface temperature was adjusted to 58.6 ± 0.1°C using cooking wrap to avoid small burns caused by heating. The ON time was divided into 3 periods: ON1, 2s from the start of moxibustion stimulation (<45°C); ON2, 5s from 2s after the start of output to the end of stimulation (>45°C); and ON3, 3s after the completion of stimulation. Each block was designed with all options other than ON set to OFF.

Results: Common and different activations were observed in all ON times. During stimulation, common activation was observed in the insula, S1, and supramarginal gyrus. Activation in the central operculum, frontal operculum and supplementary motor area was observed only in the ON1 condition, while activation in the frontal pole, cerebellum, and right S2 was observed only in the ON2 condition. Using electric moxibustion that mimics a grain-sized direct moxa cone, common and different activations were confirmed from the start of output to 45°C and above 45°C, and the activation was sustained after the completion of stimulation.

Conclusion: This study showed that moxibustion could affect almost the same areas of pain-related regions. Based on the findings of this study, further research on moxibustion-induced brain activation may help elucidate the mechanism of its therapeutic effects.

Purpose: To study the potential advantages of phosphorus magnetic resonance spectroscopy (³¹P-MRS) in differentiating advanced from mild fibrosis in non-alcoholic fatty liver disease (NAFLD) and early diagnosis at high field strength MR (9.4 Tesla).

Methods: Fibrosis of normal and carbon tetrachloride (CCl₄)-treated male rats was staged into: none (F0), perisinusoidal or periportal (F1), perisinusoidal and portal/periportal (F2), bridging fibrosis (F3) and cirrhosis (F4) by Sirius Red staining. The degree of steatosis and inflammatory activity were also graded based on Hematoxylin and Eosin staining. Rats were divided into different groups by different stages of fibrosis (F0, F1–2, F3–4) and laboratory blood tests were performed to verify the degree of liver injury. ³¹P-MRS was performed at 9.4T MR to obtain signal peaks of different phosphorus metabolites and the changes of the ratios between the peaks were observed.

Results: At 9.4 T, phosphoethanolamine (PE), phosphocholine (PC) and glycerophosphorylethanolamine (GPE), glycerophosphorylcholine (GPC) could be separated respectively from the peaks of phosphomonoesters (PME) and phosphodiesters (PDE), meanwhile nicotinamide adenine dinucleotide phosphate (NADPH) and uridine diphosphate glucose (UDPG) showed up as well. The marker of cell membrane metabolism, in F1–2, PME/PDE (P < 0.001), PC/GPE (P < 0.01), PC/GPC (P < 0.05) and PC/(PME + PDE) (P < 0.05) decreased while GPE/(PME + PDE) (P < 0.05) and GPC/(PME + PDE) (P < 0.05) increased significantly. In F3–4, there was a recovery trend of most ratios, especially for PC/(PME + PDE) (P < 0.05). As for the main ratio related to energy metabolism, β-ATP/P_total (P < 0.05) decreased in the early stage of the disease (F1–2) and this decline was maintained in advanced stage (F3–4). NADPH/P_total (P < 0.01) and β-ATP/Pi (inorganic phosphate) (P < 0.05) ratio was lower in F3–4 comparing with F0.

Conclusion: ³¹P-MRS can generally stage the liver fibrosis by comparing the ratios of the phosphorus metabolites resonance peaks at 9.4 T and more importantly it can be used for early diagnosis.

Purpose: Pulmonary valve regurgitation after repaired Tetralogy of Fallot (TOF) or double-outlet right ventricle (DORV) causes hypertrophy and papillary muscle enlargement. Cardiac magnetic resonance imaging (CMR) can evaluate the right ventricular (RV) dilatation, but the effect of trabecular and papillary muscle (TPM) exclusion on RV volume for TOF or DORV reoperation decision is unclear.

Methods: Twenty-three patients with repaired TOF or DORV, and 19 healthy controls aged ≥15, underwent CMR from 2012 to 2022. TPM volume is measured by artificial intelligence. Reoperation was considered when RV end-diastolic volume index (RVEDVI) >150 mL/m² or RV end-systolic volume index (RVESVI) >80 mL/m².

Results: RV volumes were higher in the disease group than controls (P α 0.001). RV mass and TPM volumes were higher in the disease group (P α 0.001). The reduction rate of RV volumes due to the exclusion of TPM volume was 6.3% (2.1–10.5), 11.7% (6.9–13.8), and 13.9% (9.5–19.4) in the control, volume load, and volume α pressure load groups, respectively. TPM/RV volumes were higher in the volume α pressure load group (control: 0.07 g/mL, volume: 0.14 g/mL, volume α pressure: 0.17 g/mL), and correlated with QRS duration (R α 0.77). In 3 patients in the volume α pressure, RV volume included TPM was indicated for reoperation, but when RV volume was reduced by TPM removal, reoperation was no indicated.

Conclusion: RV volume measurements, including TPM in volume α pressure load, may help determine appropriate volume recommendations for reoperation.

Purpose: To evaluate the feasibility of high-spatial-resolution hepatobiliary phase (HBP) imaging using optimized integrated combination with the compressed sensing and parallel imaging technique (Compressed SENSE).

Methods: Sixty consecutive participants underwent liver MRI and breath-hold HBP imaging using enhanced T1 high-resolution isotropic volume excitation (eTHRIVE; SENSE factor, 1.7; slice thickness/gap, 4/-2 mm; and acquisition time, 20s), eTHRIVE with Compressed SENSE (CS-eTHRIVE_4mm; C SENSE factor, 3.45; slice thickness/gap, 4/-2 mm; and acquisition time, 10s), and thin-slice eTHRIVE with Compressed SENSE (CS-eTHRIVE_2mm; C SENSE factor, 3.45; slice thickness/gap, 2/0 mm; and acquisition time, 20s). The signal intensity ratio (SIR) and signal-to-noise ratio (SNR) of the liver on each HBP image were calculated. The image quality and conspicuity of hypointense nodules on HBP images were qualitatively assessed. Then, the sensitivity for detecting hypointense nodules was calculated. The quantitative and qualitative parameters of three HBP images were compared.

Results: The SIR of the three HBP images did not differ (P = 0.36). The SNR of CS-eTHRIVE_2mm was lower than that of eTHRIVE and CS-eTHRIVE_4mm (P < 0.001). CS-eTHRIVE_2mm had a better image quality than eTHRIVE and CS-eTHRIVE_4mm (P < 0.001). CS-eTHRIVE_2mm (97.5%) had a significantly better sensitivity for detecting hypointense nodules on HBP image than eTHRIVE (86.4%) and CS-eTHRIVE_4mm (89.0%) (P = 0.001‒0.006).

Conclusion: CS-eTHRIVE_2mm had an excellent image quality and lesion detectability due to its high-spatial-resolution.

Purpose: This study aimed to compare the differences in the imaging findings for dedifferentiated liposarcoma (DDLS) and myxoid liposarcoma (MLS).

Methods: The study included 30 patients with histopathologically confirmed DDLS and 13 patients with MLS. All DDLSs and MLSs were diagnosed immunohistochemically using MDM2 and DDIT3 staining, respectively. Conventional MRI, CT, and ¹⁸F-fluorodeoxyglucose-positron emission tomography/CT findings were retrospectively evaluated and compared between the 2 pathologies.

Results: The median age of patients with DDLS was higher than that of patients with MLS (74 vs. 46 years, P < 0.01). In 10 DDLSs and 7 MLSs with fatty areas, the well-differentiated liposarcoma-like fatty areas were more common in DDLS than in MLS (70% vs. 14%), whereas septal/linear fatty areas were less common in DDLS than in MLS (30% vs. 86%, P < 0.05). The T2-hyperintense area of non-fatty area was less common in DDLS than in MLS (50% vs. 92%, P < 0.05), and the tumor-to-muscle signal intensity ratio of non-fatty areas on T2-weighted images was lower in DDLS than in MLS (3.18 vs. 5.92, P < 0.01). Apparent diffusion coefficient value was lower in DDLS than in MLS (1.29 vs. 2.10 × 10⁻³mm²/sec, P < 0.01). Unenhanced CT attenuation of non-fatty area was greater in DDLS than in MLS (33 vs. 19 Hounsfield unit, P < 0.01).

Conclusion: MRI features are valuable in differentiating MLS from DDLS. Younger age, septal/linear fatty areas, and high signal intensity of non-fatty areas on T2-weighted images were useful for diagnosing MLS.

Purpose: To evaluate short T₂ components potentially reflecting calcification or other susceptibility-affected tissue components in atherosclerotic plaques, using multicomponent analysis with ultrashort TE (UTE) MRI.

Methods: A phantom experiment was conducted using a 4-echo UTE sequence, mimicking the sample as a small amount of calcification found intra-voxel. The phantom included 6 samples containing varying concentrations of hydroxyapatite (calcification) and mayonnaise (lipid–water emulsion). Data acquired from the UTE sequence were compared with those obtained using a conventional multi-echo gradient-echo (mGRE) method.

Results: Bi-exponential analysis of UTE data successfully separated short- and long-T₂* components, with ranges of 0.44–4.81 ms and 4.29–24.37 ms, respectively. Short T₂* values derived from UTE showed minor changes with increasing hydroxyapatite concentration. Using bi-exponential analysis of mGRE data, short and long T₂* values ranged from 0.17–0.77 ms and 6.16–39.20 ms, respectively. For mono-exponential fitting of mGRE data, T₂* values ranged from 4.84–38.32 ms. In all datasets, 1/T₂* increased with hydroxyapatite concentration. The signal fraction of short T₂* components in the UTE dataset decreased as hydroxyapatite concentration increased. A clinical scan of 1 patient with an atherosclerotic plaque yielded mean short and long T₂* values of 0.12 ± 0.35 ms and 33.22 ± 17.25 ms, respectively.

Conclusion: T₂* analysis using UTE data enabled the separation of mixed calcification and mayonnaise (lipid–water emulsion) within a sample into 2 components and detected short T₂* components that may reflect calcification-related susceptibility effects, without directly indicating calcification. Multicomponent T₂* analysis with UTE-MRI is a promising technique for evaluating calcification and other short T₂* components in atherosclerotic plaques.

Purpose: Among patients with hepatitis B virus-associated liver cirrhosis (HBV-LC), there may be differences in the hepatic parenchyma between those with and without hepatocellular carcinoma (HCC). Proton MR spectroscopy (¹H-MRS) is a well-established tool for noninvasive metabolomics, but has been challenging in the liver allowing only a few metabolites to be detected other than lipids. This study aims to explore the potential of ¹H-MRS of the liver in conjunction with deep learning to differentiate between HBV-LC patients with and without HCC.

Methods: Between August 2018 and March 2021, ¹H-MRS data were collected from 37 HBV-LC patients who underwent MRI for HCC surveillance, without HCC (HBV-LC group, n = 20) and with HCC (HBV-LC-HCC group, n = 17). Based on a priori knowledge from the first 10 patients from each group, big spectral datasets were simulated to develop 2 kinds of convolutional neural networks (CNNs): CNNs quantifying 15 metabolites and 5 lipid resonances (qCNNs) and CNNs classifying patients into HBV-LC and HBV-LC-HCC (cCNNs). The performance of the cCNNs was assessed using the remaining patients in the 2 groups (10 HBV-LC and 7 HBV-LC-HCC patients).

Results: Using a simulated dataset, the quantitative errors with the qCNNs were significantly lower than those with a conventional nonlinear-least-squares-fitting method for all metabolites and lipids (P ≤0.004). The cCNNs exhibited sensitivity, specificity, and accuracy of 100% (7/7), 90% (9/10), and 94% (16/17), respectively, for identifying the HBV-LC-HCC group.

Conclusion: Deep-learning-aided ¹H-MRS with data augmentation by spectral simulation may have potential in differentiating between HBV-LC patients with and without HCC.

Purpose: To evaluate the efficacy of radiomic analysis applied to pretreatment gadolinium ethoxybenzyl diethylenetriamine pentaacetic acid (Gd-EOB-DTPA)-enhanced MRI (Gd-EOB-DTPA-MRI) for predicting the response to transcatheter arterial chemoembolization (TACE) for hepatocellular carcinoma.

Methods: Data and images from 40 consecutive patients (28 men, 12 women) who underwent pretreatment Gd-EOB-DTPA-MRI and a total of 52 TACE procedures for 75 non-treated hepatocellular carcinomas were retrospectively analyzed. Two radiologists manually outlined lesions on pretreatment arterial- and hepatobiliary-phase hepatic images to extract radiomic features. The radiomics data from one observer were randomly divided into a training dataset and a validation dataset in the ratio of 7:3. Radiomic features extracted using least absolute shrinkage and selection operator (LASSO) binomial regression applied to the training dataset and that showed intraclass correlation coefficients (ICC) >0.7 were used to construct a radiomic model. The predictive performance of the model was evaluated using receiver operating characteristics curves. Lesions classified as showing a complete or partial response according to the modified RECIST criteria were allocated to a response group.

Results: There was no significant difference in Child–Pugh score, tumor marker values, or TACE procedure between response and non-response groups. Six radiomic features were selected using the LASSO binomial regression and 5 of them showing an ICC >0.7 were used to establish the radiomic model. The area under the curve of the radiomic model was 0.89 for the training dataset, 0.83 for the validation dataset, and 0.83 for the other observer’s data. The sensitivity and specificity for the prediction of tumor response to TACE were 78% and 92% for the training dataset; 71% and 50% for the validation dataset; and 75% and 79% for the other observer’s data.

Conclusion: The pretreatment Gd-EOB-DTPA-MRI-based radiomic model is useful for predicting the response to TACE of hepatocellular carcinoma.