Accelerated Body Diffusion-Weighted MRI Using Artificial Intelligence

Whole-body diffusion-weighted MRI (WBDWI) is a non-invasive tool used for staging and response evaluation in oncologic practice and is at the core of emerging response criteria in advanced prostate and breast cancers.

WBDWI is a sensitive tool that radiologists can use to review the extent of disease and is achieved using a series of sequential imaging stations from the head to the mid-thigh. WBDWI accounts for more than 50% of the acquisition time of conventional whole-body MRI studies with a 1-hour duration. Despite national and international guidance for using whole-body MRI, a recent UK survey indicated that only 27% of UK radiology departments were offering a whole-body MRI service with a lack of scanner availability cited by 50% of respondents as the main challenge to service delivery. In the context of the ever-increasing capacity pressures on MRI departments, reducing acquisition times would facilitate the wider adoption of clinical WBDWI, reduce costs, and improve the patient experience.

DWI is also embedded into consensus MRI protocols across almost all tumour types including primary prostate and breast cancers, metastatic liver disease, gynaecological cancers \& GI cancers, where acquisition time savings could also be beneficial. The investigators have previously published accelerated DWI with deep learning based denoising filters (quickDWI), which can provide up to 50% reduction in whole-body MRI acquisition times. The goal of the deep-learning algorithm is to remove the noise in these subsampled images, producing an image with acceptable clinical quality.

The aim of this investigation is to extend this work by testing quickDWI within a larger retrospective data cohort, incorporating other cancers such as disease of the abdomen and pelvis, primary prostate cancer, liver metastases, and pancreatic cancer.