Auditing Sybil: Explaining Deep Lung Cancer Risk Prediction Through Generative Interventional Attributions

• URL: http://arxiv.org/abs/2602.02560v1
• Date: Fri, 30 Jan 2026 15:21:52 GMT
• Title: Auditing Sybil: Explaining Deep Lung Cancer Risk Prediction Through Generative Interventional Attributions
• Authors: Bartlomiej Sobieski, Jakub Grzywaczewski, Karol Dobiczek, Mateusz Wójcik, Tomasz Bartczak, Patryk Szatkowski, Przemysław Bombiński, Matthew Tivnan, Przemyslaw Biecek,
• Abstract summary: Lung cancer remains the leading cause of cancer mortality.<n>Current assessments rely purely on observational metrics.<n>We propose S(H)NAP, a model-agnostic auditing framework.
• Score: 13.541880000403323
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Lung cancer remains the leading cause of cancer mortality, driving the development of automated screening tools to alleviate radiologist workload. Standing at the frontier of this effort is Sybil, a deep learning model capable of predicting future risk solely from computed tomography (CT) with high precision. However, despite extensive clinical validation, current assessments rely purely on observational metrics. This correlation-based approach overlooks the model's actual reasoning mechanism, necessitating a shift to causal verification to ensure robust decision-making before clinical deployment. We propose S(H)NAP, a model-agnostic auditing framework that constructs generative interventional attributions validated by expert radiologists. By leveraging realistic 3D diffusion bridge modeling to systematically modify anatomical features, our approach isolates object-specific causal contributions to the risk score. Providing the first interventional audit of Sybil, we demonstrate that while the model often exhibits behavior akin to an expert radiologist, differentiating malignant pulmonary nodules from benign ones, it suffers from critical failure modes, including dangerous sensitivity to clinically unjustified artifacts and a distinct radial bias.

Related papers

• Suppressing Prior-Comparison Hallucinations in Radiology Report Generation via Semantically Decoupled Latent Steering [94.37535002230504]
  We develop a training-free, inference-time control framework termed Semantically Decoupled Latent Steering.<n>Our approach constructs a semantic-free intervention vector via large language model (LLM)-driven semantic decomposition.<n>We show that our approach significantly reduces the probability of historical hallucinations.
  arXiv  Detail & Related papers  (2026-02-27T04:49:01Z)
• X-ray Insights Unleashed: Pioneering the Enhancement of Multi-Label Long-Tail Data [86.52299247918637]
  Long-tailed pulmonary anomalies in chest radiography present formidable diagnostic challenges.<n>Despite the recent strides in diffusion-based methods for enhancing the representation of tailed lesions, the paucity of rare lesion exemplars curtails the generative capabilities of these approaches.<n>We propose a novel data synthesis pipeline designed to augment tail lesions utilizing a copious supply of conventional normal X-rays.
  arXiv  Detail & Related papers  (2025-12-24T06:14:55Z)
• A DeepSeek-Powered AI System for Automated Chest Radiograph Interpretation in Clinical Practice [83.11942224668127]
  Janus-Pro-CXR (1B) is a chest X-ray interpretation system based on DeepSeek Janus-Pro model.<n>Our system outperforms state-of-the-art X-ray report generation models in automated report generation.
  arXiv  Detail & Related papers  (2025-12-23T13:26:13Z)
• An Explainable Hybrid AI Framework for Enhanced Tuberculosis and Symptom Detection [55.35661671061754]
  Tuberculosis remains a critical global health issue, particularly in resource-limited and remote areas.<n>We propose a framework which enhances disease and symptom detection on chest X-rays by integrating two supervised heads and a self-supervised head.<n>Our model achieves an accuracy of 98.85% for distinguishing between COVID-19, tuberculosis, and normal cases, and a macro-F1 score of 90.09% for multilabel symptom detection.
  arXiv  Detail & Related papers  (2025-10-21T17:18:55Z)
• Unsupervised Deep Generative Models for Anomaly Detection in Neuroimaging: A Systematic Scoping Review [0.8373057326694192]
  Unsupervised deep generative models are emerging as a promising alternative to supervised methods for detecting and segmenting anomalies in brain imaging.<n>These models can be trained exclusively on healthy data and identify anomalies as deviations from learned normative brain structures.<n>This PRISMA-guided scoping review synthesises recent work on unsupervised deep generative models for anomaly detection in neuroimaging.
  arXiv  Detail & Related papers  (2025-10-16T09:02:52Z)
• RAD: Towards Trustworthy Retrieval-Augmented Multi-modal Clinical Diagnosis [56.373297358647655]
  Retrieval-Augmented Diagnosis (RAD) is a novel framework that injects external knowledge into multimodal models directly on downstream tasks.<n>RAD operates through three key mechanisms: retrieval and refinement of disease-centered knowledge from multiple medical sources, a guideline-enhanced contrastive loss transformer, and a dual decoder.
  arXiv  Detail & Related papers  (2025-09-24T10:36:14Z)
• Multi-pathology Chest X-ray Classification with Rejection Mechanisms [36.0596663889937]
  Overconfidence in deep learning models poses a significant risk in high-stakes medical imaging tasks.<n>This study introduces an uncertainty-aware framework for chest X-ray diagnosis based on a DenseNet-121 backbone.
  arXiv  Detail & Related papers  (2025-09-12T15:36:26Z)
• Explainability Through Human-Centric Design for XAI in Lung Cancer Detection [2.380494879018844]
  We present XpertXAI, a generalizable expert-driven model for interpretable lung cancer diagnosis.<n>XpertXAI preserves human-interpretable clinical concepts while scaling to detect multiple lung pathologies.<n>We find that existing techniques frequently fail to produce clinically meaningful explanations.
  arXiv  Detail & Related papers  (2025-05-14T19:40:12Z)
• Conformal Risk Control for Pulmonary Nodule Detection [0.6494572832138593]
  We present a case study on pulmonary nodule detection for lung cancer screening.<n>We enhance an advanced detection model with an uncertainty quantification technique called conformal risk control (CRC)<n>We demonstrate that prediction sets with conformal guarantees are attractive measures of predictive uncertainty in the safety-critical healthcare domain.
  arXiv  Detail & Related papers  (2024-12-28T14:41:57Z)
• Instrumental Variable Learning for Chest X-ray Classification [52.68170685918908]
  We propose an interpretable instrumental variable (IV) learning framework to eliminate the spurious association and obtain accurate causal representation. Our approach's performance is demonstrated using the MIMIC-CXR, NIH ChestX-ray 14, and CheXpert datasets.
  arXiv  Detail & Related papers  (2023-05-20T03:12:23Z)
• Penalized Deep Partially Linear Cox Models with Application to CT Scans of Lung Cancer Patients [42.09584755334577]
  Lung cancer is a leading cause of cancer mortality globally, highlighting the importance of understanding its mortality risks to design effective therapies. The National Lung Screening Trial (NLST) employed computed tomography texture analysis to quantify the mortality risks of lung cancer patients. We propose a novel Penalized Deep Partially Linear Cox Model (Penalized DPLC), which incorporates the SCAD penalty to select important texture features and employs a deep neural network to estimate the nonparametric component of the model.
  arXiv  Detail & Related papers  (2023-03-09T15:38:16Z)
• Towards Reliable and Explainable AI Model for Solid Pulmonary Nodule Diagnosis [20.510918720980467]
  Lung cancer has the highest mortality rate of deadly cancers in the world. Computer-aided diagnosis (CAD) systems have been developed to assist radiologists in nodule detection and diagnosis. Lack of model reliability and interpretability remains a major obstacle for its large-scale clinical application.
  arXiv  Detail & Related papers  (2022-04-08T08:21:00Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.