# Physics-Embedded Feature Learning for AI in Medical Imaging

arXiv: https://arxiv.org/abs/2603.28057 · HTML: https://mlsanity.com/paper/2603.28057/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 7 of its 20 references that are in the map.
- First posted on arXiv 30 Mar 2026.

## Builds on (most cited first, 7 of 7)

- [Scientific Machine Learning through Physics-Informed Neural Networks: Where we are and What's next](https://mlsanity.com/paper/2201.05624/) (2022)
- [Inf-Net: Automatic COVID-19 Lung Infection Segmentation from CT Images](https://mlsanity.com/paper/2004.14133/) (2020)
- [Brain Tumor Segmentation with Deep Neural Networks](https://mlsanity.com/paper/1505.03540/) (2015)
- [Physics-informed Score-based Diffusion Model for Limited-angle Reconstruction of Cardiac Computed Tomography](https://mlsanity.com/paper/2405.14770/) (2024)
- [An Explainable AI Framework for Artificial Intelligence of Medical Things](https://mlsanity.com/paper/2403.04130/) (2024)
- [Bayesian Inference of Tissue Heterogeneity for Individualized Prediction of Glioma Growth](https://mlsanity.com/paper/2209.12089/) (2022)
- [Empowering Healthcare through Privacy-Preserving MRI Analysis](https://mlsanity.com/paper/2403.09836/) (2024)

## Abstract (arXiv)

Deep learning (DL) models have achieved strong performance in an intelligence healthcare setting, yet most existing approaches operate as black boxes and ignore the physical processes that govern tumor growth, limiting interpretability, robustness, and clinical trust. To address this limitation, we propose PhysNet, a physics-embedded DL framework that integrates tumor growth dynamics directly into the feature learning process of a convolutional neural network (CNN). Unlike conventional physics-informed methods that impose physical constraints only at the output level, PhysNet embeds a reaction diffusion model of tumor growth within intermediate feature representations of a ResNet backbone. The architecture jointly performs multi-class tumor classification while learning a latent tumor density field, its temporal evolution, and biologically meaningful physical parameters, including tumor diffusion and growth rates, through end-to-end training. This design is necessary because purely data-driven models, even when highly accurate or ensemble-based, cannot guarantee physically consistent predictions or provide insight into tumor behavior. Experimental results on a large brain MRI dataset demonstrate that PhysNet outperforms multiple state-of-the-art DL baselines, including MobileNetV2, VGG16, VGG19, and ensemble models, achieving superior classification accuracy and F1-score. In addition to improved performance, PhysNet produces interpretable latent representations and learned bio-physical parameters that align with established medical knowledge, highlighting physics-embedded representation learning as a practical pathway toward more trustworthy and clinically meaningful medical AI systems.