Scholarlink

<p>The precise identification of Retinal Nerve Fiber Layer Defects (RNFLD) plays a pivotal role in diagnosing ocular conditions like glaucoma. This research applies a new paradigm in automatic detection and localization of the RNFLD using object detection techniques. The methodology employed a comprehensive public database, robust data augmentation techniques, and the incorporation of diverse color spaces to train the model. The results obtained from the YOLOv8 model demonstrate a sensitivity of 0.888, precision of 0.882, and an F1 score of 0.885 in RNFLD detection/localization, marking a significant improvement over methods such as RNNs, CNNs, and Random Forest. In comparison, experiments conducted with the YOLOv10 model yield a sensitivity of 0.798, a precision of 0.847, and an F1 score of 0.822. Although the YOLOv10 model’s results do not surpass those of YOLOv8, they still reflect commendable performance and applicability in real-world scenarios. These findings underscore the potential of object detection models as groundbreaking tools in ophthalmic diagnostics, paving the way for future innovations in the field.</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

Advanced AI strategies for accurate detection and localization of retinal nerve fiber layer defects: employing multifaceted analysis across various color spaces

  • 期刊论文

  • 发表日期：

    2025年2月15日

查看全文
  • Eduardo Hernández-Barrera,
  • Gendry Alfonso-Francia,
  • Mariana Badillo-Fernández,
  • Saul Tovar-Arriaga

摘要

The precise identification of Retinal Nerve Fiber Layer Defects (RNFLD) plays a pivotal role in diagnosing ocular conditions like glaucoma. This research applies a new paradigm in automatic detection and localization of the RNFLD using object detection techniques. The methodology employed a comprehensive public database, robust data augmentation techniques, and the incorporation of diverse color spaces to train the model. The results obtained from the YOLOv8 model demonstrate a sensitivity of 0.888, precision of 0.882, and an F1 score of 0.885 in RNFLD detection/localization, marking a significant improvement over methods such as RNNs, CNNs, and Random Forest. In comparison, experiments conducted with the YOLOv10 model yield a sensitivity of 0.798, a precision of 0.847, and an F1 score of 0.822. Although the YOLOv10 model’s results do not surpass those of YOLOv8, they still reflect commendable performance and applicability in real-world scenarios. These findings underscore the potential of object detection models as groundbreaking tools in ophthalmic diagnostics, paving the way for future innovations in the field.