A research roadmap has just been published in the journal Radiology, which highlights knowledge gaps in the literature. This was produced following a workshop held in August of last year, hosted by the National Institutes of Health (NIH), the Radiological Society of North America (RSNA), the American College of Radiology (ACR) and The Academy for Radiology and Biomedical Imaging Research (The Academy). The event aimed to foster collaboration in applications for diagnostic medical imaging, identify knowledge gaps and develop a roadmap to prioritise research needs.
“The scientific challenges and opportunities of AI in medical imaging are profound, but quite different from those facing AI generally,” said lead author Curtis P. Langlotz. “Our goal was to provide a blueprint for professional societies, funding agencies, research labs, and everyone else working in the field to accelerate research toward AI innovations that benefit patients."
Current applications of AI in radiology include medical image reconstruction, noise reduction, quality assurance, triage, segmentation, computer-aided detection, computer-aided classification and radiogenomics. Although there is huge potential for AI within medical imaging, it is still in its early stages.
Key research priorities highlighted in the report include: new image reconstruction methods that efficiently produce images suitable for human interpretation from source data, automated image labelling and annotation methods, including information extraction from the imaging report, electronic phenotyping, and prospective structured image reporting, new machine learning methods for clinical imaging data, such as tailored, pre-trained model architectures, and distributed machine learning methods, machine learning methods that can explain the advice they provide to human users (so-called explainable artificial intelligence) and validated methods for image de-identification and data sharing to facilitate wide availability of clinical imaging data sets.
The report describes a number of solutions that would help to create more publicly available, validated and reusable data sets against which to evaluate new algorithms and techniques, noting that to be useful for machine learning these data sets require methods to rapidly create labelled or annotated imaging data.
In laying out the foundational research goals for AI in medical imaging, the authors stress that standards bodies, professional societies, governmental agencies, and private industry must work together to achieve these goals for patients, who stand to benefit from the innovative imaging technologies that will result.
