Automated anatomy-based post-processing reduces false positives and improved interpretability of deep learning intracranial aneurysm detection

2025-07-01

AI Summary

Deep learning models for intracranial aneurysm detection in CT angiography (CTA) suffer from high false-positive rates, severely hindering clinical deployment. To address this, we propose an anatomy-guided heuristic hybrid post-processing method that leverages multi-organ semantic segmentation masks—specifically of the brain parenchyma, arteries, veins, and cavernous sinus—to perform cross-modal false-positive suppression on detection outputs from both CPM-Net and a deformable 3D CNN-Transformer. This approach requires no model retraining and relies solely on anatomical priors to achieve automated, interpretable, and clinically grounded post-processing. Experimental results demonstrate substantial improvements: false-positive rates are reduced by 70.6% (CPM-Net) and 51.6% (3D CNN-Transformer), while maintaining constant true-positive rates. Consequently, the average number of false positives per case drops to 0.26 and 0.62, respectively—significantly enhancing discriminative performance and clinical reliability.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Calibration & Uncertainty Quantification
• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Large pretrained models with web data
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Introduction: Deep learning (DL) models can help detect intracranial aneurysms on CTA, but high false positive (FP) rates remain a barrier to clinical translation, despite improvement in model architectures and strategies like detection threshold tuning. We employed an automated, anatomy-based, heuristic-learning hybrid artery-vein segmentation post-processing method to further reduce FPs. Methods: Two DL models, CPM-Net and a deformable 3D convolutional neural network-transformer hybrid (3D-CNN-TR), were trained with 1,186 open-source CTAs (1,373 annotated aneurysms), and evaluated with 143 held-out private CTAs (218 annotated aneurysms). Brain, artery, vein, and cavernous venous sinus (CVS) segmentation masks were applied to remove possible FPs in the DL outputs that overlapped with: (1) brain mask; (2) vein mask; (3) vein more than artery masks; (4) brain plus vein mask; (5) brain plus vein more than artery masks. Results: CPM-Net yielded 139 true-positives (TP); 79 false-negative (FN); 126 FP. 3D-CNN-TR yielded 179 TP; 39 FN; 182 FP. FPs were commonly extracranial (CPM-Net 27.3%; 3D-CNN-TR 42.3%), venous (CPM-Net 56.3%; 3D-CNN-TR 29.1%), arterial (CPM-Net 11.9%; 3D-CNN-TR 53.3%), and non-vascular (CPM-Net 25.4%; 3D-CNN-TR 9.3%) structures. Method 5 performed best, reducing CPM-Net FP by 70.6% (89/126) and 3D-CNN-TR FP by 51.6% (94/182), without reducing TP, lowering the FP/case rate from 0.88 to 0.26 for CPM-NET, and from 1.27 to 0.62 for the 3D-CNN-TR. Conclusion: Anatomy-based, interpretable post-processing can improve DL-based aneurysm detection model performance. More broadly, automated, domain-informed, hybrid heuristic-learning processing holds promise for improving the performance and clinical acceptance of aneurysm detection models.

Problem

• Reducing false positives in deep learning aneurysm detection
• Improving interpretability of aneurysm detection models
• Enhancing clinical acceptance of automated detection systems

Innovation

• Automated anatomy-based post-processing reduces false positives
• Hybrid artery-vein segmentation improves interpretability
• 3D-CNN-TR and CPM-Net models enhance detection accuracy