Scholarlink

<p>To enhance thrombolysis eligibility in acute ischemic stroke, we developed a deep learning model to estimate stroke onset within 4.5&#xa0;h using diffusion-weighted imaging (DWI) and fluid-attenuated inversion recovery (FLAIR) images. Given the variability in human interpretation, our multimodal Res-U-Net (mRUNet) model integrates a modified U-Net and ResNet-34 to classify stroke onset as &lt; 4.5 or ≥ 4.5&#xa0;h. Using DWI and FLAIR images from patients scanned within 24&#xa0;h of symptom onset, the modified U-Net generated a DWI–FLAIR mismatch image, while ResNet-34 performed the final classification. mRUNet was evaluated against ResNet-34 and DenseNet-121 on an internal test set (n = 123) and two external test sets: a single-center (n = 468) and a multi-center (n = 1151). mRUNet achieved an area under the receiver operating characteristic curve (AUC-ROC) of 0.903 on the internal set and 0.910 and 0.868 on external sets, significantly outperforming ResNet-34 and DenseNet-121. Our mRUNet model demonstrated robust and consistent classification of the 4.5-h onset-time window across datasets. By leveraging DWI and FLAIR images as a tissue clock, this model may support timely and individualized thrombolysis in patients with unclear stroke onset, such as those with wake-up stroke, in clinical settings.</p>

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

Deep learning to identify stroke within 4.5 h using DWI and FLAIR in a prospective multicenter study

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年7月19日

查看全文
  • Eun Namgung,
  • Young Sun Kim,
  • Eun-Jae Lee,
  • Dae-Il Chang,
  • Han Jin Cho,
  • Jun Lee,
  • Jae-Kwan Cha,
  • Man-Seok Park,
  • Kyung Ho Yu,
  • Jin-Man Jung,
  • Seong Hwan Ahn,
  • Dong-Eog Kim,
  • Ju Hun Lee,
  • Keun-Sik Hong,
  • Sung-Il Sohn,
  • Kyung-Pil Park,
  • Jun Young Chang,
  • Bum Joon Kim,
  • Sun U. Kwon,
  • Gayoung Park,
  • Hye-Soo Jung,
  • Jihoun Hong,
  • Dong-Wha Kang

摘要

To enhance thrombolysis eligibility in acute ischemic stroke, we developed a deep learning model to estimate stroke onset within 4.5 h using diffusion-weighted imaging (DWI) and fluid-attenuated inversion recovery (FLAIR) images. Given the variability in human interpretation, our multimodal Res-U-Net (mRUNet) model integrates a modified U-Net and ResNet-34 to classify stroke onset as < 4.5 or ≥ 4.5 h. Using DWI and FLAIR images from patients scanned within 24 h of symptom onset, the modified U-Net generated a DWI–FLAIR mismatch image, while ResNet-34 performed the final classification. mRUNet was evaluated against ResNet-34 and DenseNet-121 on an internal test set (n = 123) and two external test sets: a single-center (n = 468) and a multi-center (n = 1151). mRUNet achieved an area under the receiver operating characteristic curve (AUC-ROC) of 0.903 on the internal set and 0.910 and 0.868 on external sets, significantly outperforming ResNet-34 and DenseNet-121. Our mRUNet model demonstrated robust and consistent classification of the 4.5-h onset-time window across datasets. By leveraging DWI and FLAIR images as a tissue clock, this model may support timely and individualized thrombolysis in patients with unclear stroke onset, such as those with wake-up stroke, in clinical settings.