NeoNet: An End-to-End 3D MRI-Based Deep Learning Framework for Non-Invasive Prediction of Perineural Invasion via Generation-Driven Classification

2026-03-31

AI Summary

This study addresses the lack of reliable non-invasive imaging criteria for diagnosing perineural invasion (PNI) in cholangiocarcinoma. To this end, the authors propose NeoNet, an end-to-end 3D deep learning framework that first localizes and crops tumor regions to focus on lesions of interest. It then employs a ControlNet-guided 3D latent diffusion model to generate synthetic images, mitigating data imbalance, and introduces a novel 3D dual-attention mechanism to automatically capture subtle intensity and spatial features associated with PNI—without requiring manual feature predefinition. This work represents the first integration of a generation-driven strategy with 3D dual attention for PNI detection, achieving an AUC of 0.7903 in five-fold cross-validation, significantly outperforming existing 3D baseline models.

Abstract

Minimizing invasive diagnostic procedures to reduce the risk of patient injury and infection is a central goal in medical imaging. And yet, noninvasive diagnosis of perineural invasion (PNI), a critical prognostic factor involving infiltration of tumor cells along the surrounding nerve, still remains challenging, due to the lack of clear and consistent imaging criteria criteria for identifying PNI. To address this challenge, we present NeoNet, an integrated end-to-end 3D deep learning framework for PNI prediction in cholangiocarcinoma that does not rely on predefined image features. NeoNet integrates three modules: (1) NeoSeg, utilizing a Tumor-Localized ROI Crop (TLCR) algorithm; (2) NeoGen, a 3D Latent Diffusion Model (LDM) with ControlNet, conditioned on anatomical masks to generate synthetic image patches, specifically balancing the dataset to a 1:1 ratio; and (3) NeoCls, the final prediction module. For NeoCls, we developed the PNI-Attention Network (PattenNet), which uses the frozen LDM encoder and specialized 3D Dual Attention Blocks (DAB) designed to detect subtle intensity variations and spatial patterns indicative of PNI. In 5-fold cross-validation, NeoNet outperformed baseline 3D models and achieved the highest performance with a maximum AUC of 0.7903.

Problem

• Perineural Invasion
• Non-invasive diagnosis
• Medical imaging
• Cholangiocarcinoma
• 3D MRI

Innovation

• 3D Latent Diffusion Model
• ControlNet
• Dual Attention Blocks
• End-to-End Framework
• Perineural Invasion Prediction