An Imageless Magnetic Resonance Diagnosis procedure for fast and affordable screening and follow-up

Motivation: Traditional MRI is costly and limited in availability, restricting its use for large-scale screening and follow-up. Combining high-efficiency pulse sequences, affordable hardware, and AI could address these limitations. Goal(s): Our goal was to assess the feasibility of Imageless MR Diagnosis (IMRD) through white matter lesion detection as a simulated case study. Approach: We optimized a fingerprinting-inspired acquisition of a single radial spoke and used the resulting data to train a deep-learning model for lesion detection. Results: Simulations using a single-gradient axis achieved an AUC greater than 0.95, indicating the feasibility of IMRD within short scan times (<1 minute). Impact: The use of Imageless MR sequences, combined with deep-learning methods, could offer a rapid, cost-effective screening technique suitable for large population-wise deployment. In simulations we show how white matter lesions could potentially be detected and characterized.