Deep probabilistic logic programming for diagnostic reasoning from incomplete information: A case study in stroke detection

2026-08-09

AI Summary

This work addresses the challenge in medical diagnosis where access to raw patient data is often restricted due to privacy concerns, leaving only aggregated statistics from published literature—typically insufficient for comprehensive analysis. To overcome this limitation, the authors propose a neuro-symbolic approach that leverages the principle of maximum entropy to reconstruct missing probabilistic information and constructs a causal model grounded in ProbLog 2. This model is then reformulated into a discriminative architecture compatible with DeepProbLog, enabling the development of an interpretable stroke diagnosis system from limited statistical evidence. Experimental results demonstrate the model’s effectiveness under severe data constraints and, for the first time, validate the potential of ProbFOIL 2 in compressing large-scale discriminative models while preserving performance and interpretability.

Abstract

In medical applications, raw data is frequently associated with significant privacy concerns, lending particular importance to the encoding of summary statistics from the literature. On the other hand, deep learning has become an invaluable tool for assessing symptoms based on visual or auditory sensor data. DeepProbLog allows for an extensible neuro-symbolic approach that accommodates connectionist components to analyse patient images within a transparent and rigorous probabilistic framework, namely probabilistic logic programming under the distribution semantics. Framed as a case study in stroke detection from multimodal data, this contribution explores the pathway from summary statistics available in the literature to a DeepProbLog-based diagnostic system. It suggests a workflow using established maximum entropy techniques to complete available probabilistic information and the probabilistic logic programming system ProbLog 2 to move from the entropy-maximising causal model to a discriminative neuro-symbolic model expressible within DeepProbLog. The relative performance of models derived from less complete data is analysed alongside the potential of the probabilistic inductive logic programming system ProbFOIL 2 for compressing large discriminative models, and the perspectives and implications of using DeepProbLog for diagnostic reasoning are discussed.

Problem

• diagnostic reasoning
• incomplete information
• stroke detection
• privacy concerns
• multimodal data

Innovation

• DeepProbLog
• neuro-symbolic reasoning
• probabilistic logic programming
• maximum entropy
• incomplete diagnostic data