Synthetic MRI using two types of scan data, namely MP2RAGE and Multi TE, calculates T₁ value and T₂ value by relaxation time analysis using a Bayesian estimation method. Making use of the NIST phantom quantitative value as a true value, we examined the influence of setting conditions on the quantitative phantom value of MP2RAGE, which is an index of the T₁ value, and that of multi TE, which is an indicator of the T₂ value.

　MP2 RAGE improves the accuracy of measured values by including T₁, the value to be measured, in the range of TI 1 and TI 2. Enabling Multi TE improves the accuracy of the measured values by setting a short TE of about 10 ms.

　In case of clinical use, the quantitative value of the target is unclear, so a setting according to the imaging site is necessary.

　Magnetic resonance (MR) conditional implanted medical devices have become widely used, making it necessary to consider safety for MR imaging (MRI). Accordingly, compliance with the conditions listed in the package insert is essential. The package insert for an MR conditional implanted device indicates that the time-varying magnetic field should be less than 60 T/s. However, the field may exceed 60 T/s in normal operation mode. Therefore, operational conditions may deviate from the conditions specified in the package insert without the practitioner being aware. Practices that deviate from the recommendations of the package insert may compromise safety. Even in normal operation mode, it is important to understand that the time-varying magnetic field may exceed the recommendations of the package insert. As such, MRI should be performed under the conditions specified by the package insert to ensure patient safety.

　We correlated fat content surrounding breast cancer using the IDEAL-IQ method with histologic prognostic factors. There was a significant correlation between the fat content calculated with the IDEAL-IQ method and the histologic prognostic factors (lymph node metastasis and MIB-1 index). Fat content surrounding breast cancer calculated with the IDEAL-IQ method may be a useful prognostic biomarker for breast carcinoma.

　Spiral magnetic resonance imaging enables fast and efficient acquisition, although it suffers from aliasing artifact in cases of undersampling. It is still challenging to achieve fast and robust reconstruction for undersampled spiral datasets since iterative approaches are generally required. In this study, we developed a straightforward reconstruction method for spiral imaging using the non-uniform Fourier transform and convolutional neural networks (CNN). Two CNNs were used for estimating the sensitivity maps for parallel imaging and removing aliasing artifact caused by the undersampling. Simulation and volunteer studies were conducted to show the usefulness of our method. The results implied that the proposed method enables better performance than a conventional reconstruction method.

　We analyzed the efficacy of time-intensity curve and texture analysis of dynamic contrast enhanced study (DCE), and also mean apparent diffusion coefficient (ADC) and histogram analysis of diffusion-weighted imaging in a neoadjuvant chemotherapy (NAC) non-effective group and an NAC effective group of patients with breast cancer. Based on texture and histogram analysis, it was speculated that early response was predictable.

　Compressive sensing (CS) is an effective approach for fast magnetic resonance imaging (MRI). To improve the reconstruction accuracy and computational speed, we propose a novel deep architecture using deep learning. Experiments on MR image reconstruction demonstrate that proposed method significantly accelerates the reconstruction time, with image quality comparable to that of traditional iterative reconstruction.