Causal-Adversarial Probing of Clinical Covariates for Prostate MRI Grading

2026-07-16

AI Summary

Prostate MRI-based cancer grading models may implicitly rely on clinical covariates that contain both disease-relevant signals and shortcut features detrimental to generalization, yet the underlying mechanisms remain unclear. This work proposes the first framework integrating causal modeling with adversarial probing, treating MRI findings and ISUP grade as observed proxies of an unmeasured pathological latent variable. By adversarially suppressing the decodability of individual covariates while preserving grading performance, the approach disentangles harmful dependencies from useful clinical signals. Evaluated on 2,903 training and 576 external validation cases, the analysis reveals that suppressing age or BMI slightly improves AUC (up to +1.42%), whereas suppressing PSA or prostate volume substantially degrades AUC (up to −7.61%), elucidating the distinct roles of clinical variables in shaping imaging representations and their differential impact on model performance.

Abstract

Deep learning models for prostate MRI-based cancer grading may encode clinical covariates that either reflect useful disease-related signal or non-generalising shortcut information, but their role is usually assumed. We propose a causal-reasoning framework for probing covariate dependence in MRI-based International Society of Urological Pathology (ISUP) Grade Group prediction. Rather than treating mpMRI as a direct cause of grade, we model MRI appearance and ISUP grade as observations of latent tumour pathology, and test whether candidate clinical variables act as nuisance correlates, disease-related proxies, or irrelevant covariates in the learned representation. We implement this using an adversarial framework that suppresses the decodability of individual clinical covariate at a time while preserving MRI-based grade prediction. The approach is developed and evaluated on 2,903 prostate MRI examinations, with external validation on 576 patients. We report a set of interesting and previously under-explored imaging-to-clinical-variable interactions in the context of deep learning generalisation. For examples, in binary ISUP Grade Group $\geq2$ classification, suppressing age, BMI, and alcohol use improved AUC by 1.23%, 0.84%, and 1.42%, respectively (all p < 0.05), suggesting reduced non-generalising covariate information; In contrast, suppressing PSA and prostate volume degraded AUC by 1.91% and 7.61% (all p < 0.001), indicating that these variables carried task-relevant signal. These findings show that adversarial covariate suppression can provide a practical representation-level analysis for distinguishing potentially harmful dependence from informative signal in prostate MRI grading models.

Problem

• prostate MRI grading
• clinical covariates
• causal probing
• representation analysis
• generalisation

Innovation

• causal reasoning
• adversarial probing
• clinical covariates
• prostate MRI grading
• representation analysis