Interpretable Artificial Intelligence for Detecting Acute Heart Failure on Acute Chest CT Scans

2025-07-11

AI Summary

This study addresses the clinical challenge of early and objective identification of acute heart failure (AHF) in emergency department chest CT—where conventional interpretation suffers from low sensitivity and high inter-observer variability. We propose a highly interpretable AI-assisted diagnostic framework: anatomical structures (heart and lungs) are automatically segmented using TotalSegmentator to extract quantitative imaging biomarkers; a Boosted Trees classifier is trained for AHF prediction, with SHAP values integrated for case-wise decision attribution, explicitly quantifying each anatomical feature’s contribution. On an independent test set, the model achieves an AUC of 0.87—comparable to thoracic radiologists’ performance. Expert blinded review revealed that 38% of cases originally labeled as false positives or false negatives were correctly classified by the model, leading to correction of erroneous clinical reports. To our knowledge, this is the first work to tightly integrate anatomy-aware segmentation with inherently interpretable machine learning for AHF diagnosis, establishing a new paradigm for rapid, transparent, and clinically trustworthy AI assistance in emergency settings.

Technology Category

• Computer Vision: Interpretability, Explainability, and Transparency
• Machine Learning: Transparent, Interpretable, Explainable ML
• Humans and AI: Explainable AI (XAI) for Human Understanding

Application Category

• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Search and Retrieval-Augmented AI: Agentic search

Abstract

Introduction: Chest CT scans are increasingly used in dyspneic patients where acute heart failure (AHF) is a key differential diagnosis. Interpretation remains challenging and radiology reports are frequently delayed due to a radiologist shortage, although flagging such information for emergency physicians would have therapeutic implication. Artificial intelligence (AI) can be a complementary tool to enhance the diagnostic precision. We aim to develop an explainable AI model to detect radiological signs of AHF in chest CT with an accuracy comparable to thoracic radiologists. Methods: A single-center, retrospective study during 2016-2021 at Copenhagen University Hospital - Bispebjerg and Frederiksberg, Denmark. A Boosted Trees model was trained to predict AHF based on measurements of segmented cardiac and pulmonary structures from acute thoracic CT scans. Diagnostic labels for training and testing were extracted from radiology reports. Structures were segmented with TotalSegmentator. Shapley Additive explanations values were used to explain the impact of each measurement on the final prediction. Results: Of the 4,672 subjects, 49% were female. The final model incorporated twelve key features of AHF and achieved an area under the ROC of 0.87 on the independent test set. Expert radiologist review of model misclassifications found that 24 out of 64 (38%) false positives and 24 out of 61 (39%) false negatives were actually correct model predictions, with the errors originating from inaccuracies in the initial radiology reports. Conclusion: We developed an explainable AI model with strong discriminatory performance, comparable to thoracic radiologists. The AI model's stepwise, transparent predictions may support decision-making.

Problem

• Detecting acute heart failure in chest CT scans
• Improving diagnostic accuracy with explainable AI
• Addressing radiologist shortages and report delays

Innovation

• Boosted Trees model for AHF detection
• TotalSegmentator for cardiac structure segmentation
• Shapley values for explainable AI predictions