# Cerebrovascular diagnosis using CTA-based intracranial aneurysm classification via transfer learning and Grad-CAM visualization

## TL;DR

This study improves the diagnosis of brain aneurysms using AI with transfer learning and visualization tools, achieving better accuracy and interpretability than traditional methods.

## Contribution

A novel hybrid deep learning framework with transfer learning and Grad-CAM for interpretable intracranial aneurysm classification in limited-data settings.

## Key findings

- The DL+TL model achieved a mean AUC of 0.853 and accuracy of 84.0%, outperforming baseline AI and radiologists.
- Grad-CAM analysis showed higher attention precision (IoU: 0.68) and greater clinical relevance ratings (4.2/5) for DL+TL compared to DL.
- The framework offers improved accuracy and transparency for intracranial aneurysm classification in limited-data settings.

## Abstract

Intracranial aneurysm (IA) is a focal cerebral artery dilatation affecting 2–5% of the population, with rupture leading to high mortality and disability. Early, accurate classification from computed tomography angiography (CTA) is crucial for management but is challenged by small datasets and limited interpretability. We evaluate a hybrid deep transfer learning framework with integrated Grad-CAM to improve both discrimination and explainability in CTA-based IA classification.

In this retrospective study, 83 eligible patients from two centers underwent CTA. We employed stratified 5-fold cross-validation to compare: a baseline deep learning model (DL), a transfer learning-enhanced model (DL + TL), and radiologist assessment. Both AI models used a hybrid ResNet-18 architecture with LASSO feature selection and logistic regression. Performance was assessed using AUC, accuracy, calibration, decision curve analysis, NRI, and IDI. Interpretability was quantified via Grad-CAM using Intersection-over-Union (IoU) and Dice similarity coefficient.

The DL + TL model achieved superior performance with a mean AUC of 0.853 (95% CI: 0.789–0.912) and accuracy of 84.0%, outperforming both DL (AUC: 0.744, p = 0.012) and radiologists (AUC: 0.731, p = 0.008). Grad-CAM analysis showed DL + TL had significantly higher attention precision (IoU: 0.68 vs. 0.45 for DL, p < 0.001) and was rated more clinically relevant by blinded radiologists (4.2/5 vs. 2.8/5).

Integrating transfer learning with quantitative interpretability assessment improves both accuracy and transparency of IA classification in limited-data settings. This framework offers a validated, interpretable approach for neurovascular imaging, pending further multi-center validation.