Real-Time Polyp Detection, Localisation and Segmentation in Colonoscopy Using Deep Learning

Computer-aided detection, localisation, and segmentation methods can help improve colonoscopy procedures. Even though many methods have been built to tackle automatic detection and segmentation of polyps, benchmarking of state-of-the-art methods still remains an open problem.This is due to the increasing number of researched computer-vision methods that can be applied to polyp datasets. Benchmarking of novel methods can provide a direction to the development of automated polyp detection and segmentation tasks. Furthermore, it ensures that the produced results in the community are reproducible and provide a fair comparison of developed methods. In this paper, we benchmark several recent state-of-the-art methods using Kvasir-SEG, an open-access dataset of colonoscopy images, for polyp detection, localisation, and segmentation evaluating both method accuracy and speed. Whilst, most methods in literature have competitive performance over accuracy, we show that YOLOv4 with a Darknet53 backbone and cross-stage-partial connections achieved a better trade-off between an average precision of 0.8513 and mean IoU of 0.8025, and the fastest speed of 48 frames per second for the detection and localisation task. Likewise, UNet with a ResNet34 backbone achieved the highest dice coefficient of 0.8757 and the best average speed of 35 frames per second for the segmentation task. Our comprehensive comparison with various state-of-the-art methods reveal the importance of benchmarking the deep learning methods for automated real-time polyp identification and delineations that can potentially transform current clinical practices and minimise miss-detection rates.

使用深度学习在结肠镜检查中实时息肉检测，定位和分割

计算机辅助的检测，定位和分割方法可以帮助改善结肠镜检查程序。尽管已经建立了许多方法来解决息肉的自动检测和分割，但最新方法的基准测试仍然是一个悬而未决的问题。.. 这是由于可应用于息肉数据集的研究计算机视觉方法的数量不断增加。新方法的基准测试可以为自动息肉检测和分割任务的发展提供指导。此外，它确保社区中产生的结果可重复，并提供已开发方法的公平比较。在本文中，我们使用Kvasir-SEG（结肠镜检查图像的开放获取数据集）对几种最新技术进行了基准测试，用于息肉检测，定位和分割，评估了方法的准确性和速度。虽然文献中的大多数方法都具有优于准确性的性能，但我们证明，具有Darknet53主干和跨阶段的部分连接的YOLOv4在0.8513的平均精度和0.8025的平均IoU之间取得了更好的折衷，检测和定位任务的最快速度为每秒48帧。同样，具有ResNet34主干的UNet在分割任务中实现了最高的骰子系数0.8757和最佳的平均速度，每秒35帧。我们与各种最先进方法的全面比较表明，对深度学习方法进行基准测试以进行实时实时息肉识别和勾画非常重要，这可以潜在地改变当前的临床实践并最大程度地降低漏检率。 （阅读更多）