To clinically use the physical quantities obtained by magnetic resonance imaging (MRI) as biomarkers, quantification of the quantities and standardization of the measurement methods are necessary. The Quantitative Imaging Biomarkers Alliance (QIBA), led by the Radiological Society of North America, is proceeding such requirements. For functional MRI (fMRI), dynamic contrast-enhanced MRI (DCE-MRI), dynamic magnetic susceptibility contrast MRI (DSC-MRI), musculoskeletal (MSK), diffusion-weighted imaging (DWI), and magnetic resonance elastography (MRE), documents called "profile” summarizing required specifications has been created and is being prepared for clin ical use. On the other hand, arterial spin labeling (ASL), diffusion tensor imaging (DTI), and fat proton density for fat fraction (PDFF) are still in the early stages of standardization. There are also other physical quantities that can be biomarkers, such as temperature, pH, and magnetization transfer coefficient (MTC). In this paper, the requirements and the latest trends including developments of standard phantoms for quantification and standardization of MRI-based biomarkers are discussed.
With the worldwide increase in patients with chronic liver disease, shear wave elastography, which measures liver stiffness noninvasively using ultrasound, is widely performed in clinical practice. On the other hand, different diagnostic criteria have been reported for each device due to bias in measurements between devices. The inter-operator measurement variations and differences in stiffness compared with MR elastography have also been experienced, leading to confusion in its clinical use for diagnosis. Japan-QIBA, an imaging biomarker standardization committee, reported that the 95% confidence interval of the inter-device bias measuring viscoelastic phantoms was 11% for using the convex probe, and that the difference in frequency bands used in the measurements and analysis caused a discrepancy in stiffness between the US and MR elastography. Standardization of attenuation imaging for the diagnosis of fatty liver is also being considered in QIBA. It is expected that standardization of ultrasound biomarkers will improve the accuracy of ultrasound diagnosis of chronic liver disease in the future.
Positron emission tomography (PET) can measure in vivo biological processes at the molecular level because to use biological element as radioactive tracers. In Japan, a PET scanner was developed in 1979, and PET research has a history of about 40 years. Recently, PET is not only an essential imaging modality for diagnosis of various diseases in clinical practice but also a molecular imaging technology for drug discovery research. However, the image quality and quantitative of PET images acquired with PET scanners depends on the scanner model, injection dose, scanning duration, and other details of the data acquisition protocol. It also depends on body size, and larger subjects generally give poorer image quality with the same injected activity per weight. There is, therefore, related academic societies have created several standardization guidelines of PET imaging. This paper outlines the history of standardization of PET imaging in Japan and introduces trends in Europe and the United States.
Compton cameras have the potential to become a novel imaging modality in nuclear medicine. One of the key issues in the medical application of Compton camera is the lack of performance measurement methods. In the case of SPECT and PET, there are standard performance measurement methods that are utilized in research and development of new imaging systems. Defining the standard performance measurement methods may facilitate research and development of a clinical Compton camera. We have organized an interdisciplinary research group and work towards developing performance measurement methods for medical Compton cameras. This report overviews this work.
In a pinhole SPECT system, degradation of the spatial resolution occurs in reconstructed images depending on the pinhole aperture. Conventionally, blurring caused by such pinhole apertures has been corrected by obtaining a system matrix based on the detection probability or by increasing the number of projection rays in a ray driven method. However, the former has the drawback that the computational load increases when the pixel size is small or the number of pinholes increases, while the latter has the problem that the spatial resolution cannot be sufficiently improved when the pinhole diameter increases. Therefore, in this study, we proposed a method to reduce the deterioration of the spatial resolution due to pinholes using deep learning. Specifically, we implemented convolutional neural networks (U-net and U-net++) that learns projection data with reduced spatial resolution as input data and ideal infinitesimal pinhole projection data as training data to suppress the degradation of the spatial resolution. Simulation results showed that the peak signal-to-noise ratios of the corrected image with the U-net or U-net++ based method were 17.48 and 17.92 dB, respectively, and that with the conventional 21-rays method was 16.82 dB. It was clarified that the proposed methods can improve the spatial resolution compared with the conventional method.
In this short article, we describe the diffusion MRI analysis software developed by the authors. The technical details of the analysis method are left to the references, and the situation and the background at the time of development, distribution and dissemination as freeware, and subsequent progress are described. Based on this experience, some personal views on software development in medical imaging research are presented.