Counterfactual Explanations in Medical Imaging: Exploring SPN-Guided Latent Space Manipulation

2025-07-25

AI Summary

Deep learning models for medical imaging face challenges in interpretability and trustworthiness due to their “black-box” nature. To address this, we propose a counterfactual generation method that synergistically integrates variational autoencoders (VAEs) with sum-product networks (SPNs): an SPN probabilistically models the latent space of a semi-supervised VAE, enabling both classification and faithful distribution characterization—thus supporting semantically plausible and distributionally realistic counterfactual reasoning. Our method jointly optimizes counterfactuals in latent space for similarity to the input, confidence in the target class, and generation plausibility, while balancing regularization strength against explanation quality. Experiments on the CheXpert dataset demonstrate that our generated counterfactuals exhibit superior clinical interpretability and statistical fidelity compared to neural network baselines, significantly enhancing decision transparency and model trustworthiness.

Technology Category

• Computer Vision: Interpretability, Explainability, and Transparency
• Machine Learning: Deep Generative Models & Autoencoders
• Reasoning under Uncertainty: Relational Probabilistic 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
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Artificial intelligence is increasingly leveraged across various domains to automate decision-making processes that significantly impact human lives. In medical image analysis, deep learning models have demonstrated remarkable performance. However, their inherent complexity makes them black box systems, raising concerns about reliability and interpretability. Counterfactual explanations provide comprehensible insights into decision processes by presenting hypothetical "what-if" scenarios that alter model classifications. By examining input alterations, counterfactual explanations provide patterns that influence the decision-making process. Despite their potential, generating plausible counterfactuals that adhere to similarity constraints providing human-interpretable explanations remains a challenge. In this paper, we investigate this challenge by a model-specific optimization approach. While deep generative models such as variational autoencoders (VAEs) exhibit significant generative power, probabilistic models like sum-product networks (SPNs) efficiently represent complex joint probability distributions. By modeling the likelihood of a semi-supervised VAE's latent space with an SPN, we leverage its dual role as both a latent space descriptor and a classifier for a given discrimination task. This formulation enables the optimization of latent space counterfactuals that are both close to the original data distribution and aligned with the target class distribution. We conduct experimental evaluation on the cheXpert dataset. To evaluate the effectiveness of the integration of SPNs, our SPN-guided latent space manipulation is compared against a neural network baseline. Additionally, the trade-off between latent variable regularization and counterfactual quality is analyzed.

Problem

• Generating plausible counterfactual explanations for medical imaging AI models
• Ensuring counterfactuals adhere to similarity and interpretability constraints
• Integrating SPNs to optimize latent space manipulation in VAEs

Innovation

• SPN-guided latent space manipulation
• VAE-SPN hybrid model
• Optimizing counterfactuals with SPNs