Scholarlink

This study presents a comprehensive framework that integrates a deep learning model with advanced image preprocessing techniques to improve the multilabel classification of five types of intracranial hemorrhage—epidural, intraparenchymal, intraventricular, subarachnoid, and subdural—using non-contrast computed tomography (CT) images. The framework includes strategies to mitigate overfitting, data augmentation, and a custom loss function. It was rigorously evaluated on a dataset of over 25,000 non-contrast CT scans, each labeled by expert radiologists across six classes. The proposed model achieved 99% accuracy and 92.1% sensitivity in detecting intracranial hemorrhage, outperforming previously reported methods.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Multilabel Classification of Intracranial Hemorrhages Using Deep Learning and Preprocessing Techniques on Non-contrast CT Images

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Rodrigo Salas,
  • Juan Sebastian Castro,
  • Marvin Querales,
  • Carolina Saavedra,
  • Claudia Prieto,
  • Steren Chabert

摘要

This study presents a comprehensive framework that integrates a deep learning model with advanced image preprocessing techniques to improve the multilabel classification of five types of intracranial hemorrhage—epidural, intraparenchymal, intraventricular, subarachnoid, and subdural—using non-contrast computed tomography (CT) images. The framework includes strategies to mitigate overfitting, data augmentation, and a custom loss function. It was rigorously evaluated on a dataset of over 25,000 non-contrast CT scans, each labeled by expert radiologists across six classes. The proposed model achieved 99% accuracy and 92.1% sensitivity in detecting intracranial hemorrhage, outperforming previously reported methods.