Interpretable Prediction of Lymph Node Metastasis in Rectal Cancer MRI Using Variational Autoencoders

2025-07-15

AI Summary

Accurate preoperative prediction of lymph node metastasis (LNM) in rectal cancer via MRI is critical for personalized treatment, yet conventional morphology-based diagnostic criteria suffer from limited accuracy. To address this, we propose a lightweight, interpretable end-to-end VAE-MLP framework that takes T2-weighted MRI as input. Instead of relying on opaque, pretrained CNNs, our method employs a variational autoencoder (VAE) to learn disentangled and structurally organized latent representations; its reconstruction objective enables intuitive visualization of the latent space, substantially enhancing model transparency. Evaluated on an internal cohort of 168 patients, the framework achieves an AUC of 0.86 ± 0.05, sensitivity of 0.79 ± 0.06, and specificity of 0.85 ± 0.05—surpassing current state-of-the-art performance. This work establishes a novel, noninvasive clinical paradigm for LNM assessment that simultaneously delivers high diagnostic accuracy and strong interpretability.

Technology Category

• Machine Learning: Transparent, Interpretable, Explainable ML
• Computer Vision: Interpretability, Explainability, and Transparency
• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Effective treatment for rectal cancer relies on accurate lymph node metastasis (LNM) staging. However, radiological criteria based on lymph node (LN) size, shape and texture morphology have limited diagnostic accuracy. In this work, we investigate applying a Variational Autoencoder (VAE) as a feature encoder model to replace the large pre-trained Convolutional Neural Network (CNN) used in existing approaches. The motivation for using a VAE is that the generative model aims to reconstruct the images, so it directly encodes visual features and meaningful patterns across the data. This leads to a disentangled and structured latent space which can be more interpretable than a CNN. Models are deployed on an in-house MRI dataset with 168 patients who did not undergo neo-adjuvant treatment. The post-operative pathological N stage was used as the ground truth to evaluate model predictions. Our proposed model 'VAE-MLP' achieved state-of-the-art performance on the MRI dataset, with cross-validated metrics of AUC 0.86 +/- 0.05, Sensitivity 0.79 +/- 0.06, and Specificity 0.85 +/- 0.05. Code is available at: https://github.com/benkeel/Lymph_Node_Classification_MIUA.

Problem

• Improving lymph node metastasis staging accuracy in rectal cancer MRI
• Replacing large CNNs with interpretable Variational Autoencoder feature encoding
• Achieving high diagnostic performance using disentangled latent space features

Innovation

• Uses Variational Autoencoder for feature encoding
• Replaces large CNN with interpretable VAE model
• Achieves high accuracy in MRI metastasis prediction