YOLOv7-RepFPN:改善嵌入式系统上腹腔镜工具检测的实时性能
Yuzhang Liu1, Yuichiro Hayashi1, Masahiro Oda1,2
1Graduate School of Informatics Nagoya University Aichi, Nagoya Japan.
Healthcare technology letters
|April 19, 2024
概括
这项研究使用YOLOv7-RepFPN模型提高了嵌入式设备上的腹腔镜工具检测速度. 修改后的模型实现更快的推断 (62.9 FPS),同时保持高精度 (88.2% mAP),这对于外科导航至关重要.
科学领域:
- 医学技术 医学技术 医学技术
- 计算机视觉 计算机视觉 计算机视觉
- 人工智能的人工智能是人工智能.
背景情况:
- laparoscopy 提供了减少患者康复和更少的并发症.
- 实时工具检测有助于外科导航,但面临嵌入式设备的局限性.
- 嵌入式设备越来越多地用于手术中的便携性和可扩展性.
研究的目的:
- 在资源受限的嵌入式设备上增强腹腔镜工具检测的推断速度.
- 为了保持高的检测准确度,同时显著提高处理速度.
- 适应现有的深度学习模型,以实现高效的实时手术应用.
主要方法:
- 一个经过修改的YOLOv7模型,称为YOLOv7-RepFPN,通过将功能通道减半并集成RepBlock. developed开发出来.
- 计算复杂性通过架构修改而减少.
- 焦点EIoU (联盟的有效交叉点) 损失函数被用于界限框回归.
主要成果:
- 在一个定制数据集上,YOLOv7-RepFPN模型实现了88.2%的mAP (IoU=0.5).
- 在嵌入式设备上,推断速度达到62.9 FPS,比原来的YOLOv7.1 FPS提高了21.1 FPS.
- 检测准确度保持在与原始YOLOv7 (89.3% mAP) 相比的水平.
结论:
- YOLOv7-RepFPN模型有效地提高了嵌入式设备上的腹腔镜工具检测的推断速度.
- 拟议的修改表明了实时手术辅助系统的可行方法.
- 这项工作突出了优化深度学习模型在推进微创手术方面的潜力.
相关概念视频
02:07Improving Translational Accuracy
Base complementarity between the three base pairs of mRNA codon and the tRNA anticodon is not a failsafe mechanism. Inaccuracies can range from a single mismatch to no correct base pairing at all. The free energy difference between the correct and nearly correct base pairs can be as small as 3 kcal/ mol. With complementarity being the only proofreading step, the estimated error frequency would be one wrong amino acid in every 100 amino acids incorporated. However, error frequencies observed in...
02:07Improving Translational Accuracy
Base complementarity between the three base pairs of mRNA codon and the tRNA anticodon is not a failsafe mechanism. Inaccuracies can range from a single mismatch to no correct base pairing at all. The free energy difference between the correct and nearly correct base pairs can be as small as 3 kcal/ mol. With complementarity being the only proofreading step, the estimated error frequency would be one wrong amino acid in every 100 amino acids incorporated. However, error frequencies observed in...