Decoding the human brain tissue response to radiofrequency excitation using a biophysical-model-free deep MRI on a chip framework

TL;DR

This paper introduces DeepMonC, a transformer-based MRI framework that rapidly decodes brain tissue response to RF excitation, enabling quick, comprehensive imaging with potential clinical and pathological insights.

Contribution

The study presents a novel deep learning framework that performs biophysical-model-free decoding of MRI signals, significantly reducing scan times and providing detailed tissue characterization.

Findings

94% faster than traditional protocols

Validated on healthy and patient data across sites

Automatically generates multiple MRI contrasts

Abstract

Magnetic resonance imaging (MRI) relies on radiofrequency (RF) excitation of proton spin. Clinical diagnosis requires a comprehensive collation of biophysical data via multiple MRI contrasts, acquired using a series of RF sequences that lead to lengthy examinations. Here, we developed a vision transformer-based framework that captures the spatiotemporal magnetic signal evolution and decodes the brain tissue response to RF excitation, constituting an MRI on a chip. Following a per-subject rapid calibration scan (28.2 s), a wide variety of image contrasts including fully quantitative molecular, water relaxation, and magnetic field maps can be generated automatically. The method was validated across healthy subjects and a cancer patient in two different imaging sites, and proved to be 94% faster than alternative protocols. The deep MRI on a chip (DeepMonC) framework may reveal the molecular composition of the human brain tissue in a wide range of pathologies, while offering clinically attractive scan times.