Brain tissue gadolinium retention in pediatric patients after contrast-enhanced magnetic resonance exams: pathological confirmation

Imaging tools for assessment of myocardial fibrosis in humans: the need for greater detail

Myocardial fibrosis is recognized as a key pathological process in the development of cardiac disease and a target for future therapeutics. Despite this recognition, the assessment of fibrosis is not a part of routine clinical practice. This is primarily due to the difficulties in obtaining an accurate assessment of fibrosis non-invasively. Moreover, there is a clear discrepancy between the understandings of myocardial fibrosis clinically where fibrosis is predominately studied with comparatively low-resolution medical imaging technologies like MRI compared with the basic science laboratories where fibrosis can be visualized invasively with high resolution using molecularly specific fluorescence microscopes at the microscopic and nanoscopic scales. In this article, we will first review current medical imaging technologies for assessing fibrosis including echo and MRI. We will then highlight the need for greater microscopic and nanoscopic analysis of human tissue and how this can be addressed through greater utilization of human tissue available through endomyocardial biopsies and cardiac surgeries. We will then describe the relatively new field of molecular imaging that promises to translate research findings to the clinical practice by non-invasively monitoring the molecular signature of fibrosis in patients.

Applying Artificial Intelligence to Mitigate Effects of Patient Motion or Other Complicating Factors on Image Quality

Artificial intelligence, particularly deep learning, offers several possibilities to improve the quality or speed of image acquisition in magnetic resonance imaging (MRI). In this article, we briefly review basic machine learning concepts and discuss commonly used neural network architectures for image-to-image translation. Recent examples in the literature describing application of machine learning techniques to clinical MR image acquisition or postprocessing are discussed. Machine learning can contribute to better image quality by improving spatial resolution, reducing image noise, and removing undesired motion or other artifacts. As patients occasionally are unable to tolerate lengthy acquisition times or gadolinium agents, machine learning can potentially assist MRI workflow and patient comfort by facilitating faster acquisitions or reducing exogenous contrast dosage. Although artificial intelligence approaches often have limitations, such as problems with generalizability or explainability, there is potential for these techniques to improve diagnostic utility, throughput, and patient experience in clinical MRI practice.