Modality agnostic intracranial aneurysm detection through supervised vascular surface classification

TL;DR

This paper introduces a modality-agnostic deep learning method for intracranial aneurysm detection that achieves high sensitivity across different angiographic modalities by classifying vascular surface patches.

Contribution

The novel approach uses surface parcellation and a DNN classifier to detect aneurysms, outperforming existing intensity-based methods and being effective across multiple imaging modalities.

Findings

Achieved 98.6% sensitivity with low false positives

Significantly outperformed state-of-the-art intensity-based methods

Demonstrated modality independence across DSA, CTA, and MRA

Abstract

Intracranial aneurysms (IAs) are generally asymptomatic and thus often discovered incidentally on angiographic scans like 3D DSA, CTA and MRA. Skilled radiologists achieved a sensitivity of 88% by means of visual detection, which seems inadequate considering that prevalence of IAs in general population is 3-5%. Deep learning models trained and executed on angiographic scans seem best-suited for IA detection, however, reported performances across different modalities is currently insufficient for clinical application. This paper presents a novel modality agnostic method for detection of IAs. First the triangulated surfaces of vascular structures were roughly extracted from the angiograms. For IA detection purpose, the extracted surfaces were randomly parcellated into local patches and then a translation, rotation and scale invariant classifier based on deep neural network (DNN) was trained. Test stage proceeded by mimicking the surface extraction and parcellation at several random locations, then the trained DNN model was applied for classification, and the results aggregated into IA detection heatmaps across entire vascular surface. For training and validation the extracted contours were presented to skilled neurosurgeon, who marked the locations of IAs. The DNN was trained and tested using three-fold cross-validation based on 57 DSAs, 5 CTAs and 5 MRAs and showed a 98.6% sensitivity at 0.2 false positive detections per image. Experimental results show that proposed approach not only significantly improved detection sensitivity and specificity compared to state-of-the-art intensity based methods, but is also modality agnostic and thus better suited for clinical application.