XAI-Empowered MRI Analysis for Consumer Electronic Health

Abstract

The intelligent use of artificial intelligence-generated content (AIGC) in magnetic resonance imaging (MRI) analysis is a significant step towards the rapidly advancing field of consumer electronics (CE) used in healthcare. It greatly improves the accuracy and usefulness of diagnostic processes. This paper introduces a novel MRI analysis approach through the lens of AIGC, leveraging physics-informed deep learning (PIDL) models. This integration pioneers a new paradigm in consumer healthcare diagnostics and embeds physics principles into deep learning (DL) models, thus improving interpretability and adherence to physical constraints. Additionally, this method improves the visibility of the healthcare community by integrating explainable AI (XAI) techniques, including gradient-weighted class activation mapping (Grad-CAM) and Local Interpretable Model-Agnostic Explanations (LIME). This approach demonstrates a reasonable precision rate of 96% when applied to brain tumor MRI images. Therefore, this research introduces a new para diagram of applying AIGC in medical imaging analysis within Consumer Healthcare Electronics (CHE).