Precise Image-level Localization of Intracranial Hemorrhage on Head CT Scans with Deep Learning Models Trained on Study-level Labels

2024/08/28 by Yunan Wu, Michael Iorga, Suvarna Badhe +15 · 1 voice · 1 citation
Medicine · Neuroscience · Computer Science · #Radiomics and Machine Learning in Medical Imaging #Brain Tumor Detection and Classification #Medical Image Segmentation Techniques

paper · doi:10.1148/ryai.230296

openalex publication_date 2024/08/28 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/30

Abstract

To prioritize the reading of noncontrast head CT scans with intracranial hemorrhage, this weakly supervised detection workflow was highly generalizable, with good interpretability, high positive predictive values at the study level, and performance similar to two neuroradiologists.

Cited by

• Benchmarking Class Activation Map Methods for Explainable Brain Hemorrhage Classification on Hemorica Dataset

Discussions

• #ThrowbackThursday – Weakly supervised model for prioritizing the reading of noncontrast head CTs with intracranial hemorrhage (Nov 2024) https://doi.org/10.1148/ryai.230296 #AI #ML #Radiomics [bsky, 2 points, 0 comments]

Related

• Weakly supervised deep learning-based intracranial hemorrhage localization
• Improved ICH classification using task-dependent learning
• Development and Validation of Deep Learning Algorithms for Detection of Critical Findings in Head CT Scans
• Identification of Hemorrhage and Infarct Lesions on Brain CT Images using Deep Learning
• Weakly Supervised Learning Significantly Reduces the Number of Labels Required for Intracranial Hemorrhage Detection on Head CT