| 作成者 |
|
|
|
|
|
| 本文言語 |
|
| 出版者 |
|
| 発行日 |
|
| 収録物名 |
|
| 巻 |
|
| 開始ページ |
|
| 終了ページ |
|
| 会議情報 |
|
| 出版タイプ |
|
| アクセス権 |
|
| Crossref DOI |
|
| 権利関係 |
|
| 権利関係 |
|
| 概要 |
Cerebrovascular Accident (CVA), commonly known as stroke, remains a leading cause of death and disability, particularly in low-resource settings like the Philippines. Manual diagnosis of stroke throug...h CT scans faces challenges such as delayed processing, limited access to specialists, and human error, often resulting in late interventions. This study introduces a Smart Brain CT Scan Screening System designed to automate the detection of ischemic and hemorrhagic strokes. The system employs computer vision techniques and a U-Net-based deep learning model trained on over 2,455 CT images. After 1,000 training epochs, the model achieved satisfactory performance metrics, including an Intersection over Union (IoU) score of 1.0000 and a minimal validation loss of 2.82 × 10⁻⁸. The system enhances early detection and supports timely medical intervention by minimizing diagnostic delays and reducing reliance on radiological expertise. This is especially valuable in underserved regions. Recommendations for further work include expanding the dataset, applying advanced augmentation techniques, and conducting broader clinical validations. Overall, this study showcases the potential of artificial intelligence to revolutionize medical imaging and improve stroke care in resource-constrained environments.続きを見る
|
Mendeley出力
2025_p1289