Nogueira-Rodríguez Alba, Glez-Peña Daniel, Reboiro-Jato Miguel, López-Fernández Hugo
CINBIO, Department of Computer Science, ESEI-Escuela Superior de Ingeniería Informática, Universidade de Vigo, 32004 Ourense, Spain.
SING Research Group, Galicia Sur Health Research Institute (IIS Galicia Sur), SERGAS-UVIGO, 36213 Vigo, Spain.
Diagnostics (Basel). 2023 Mar 3;13(5):966. doi: 10.3390/diagnostics13050966.
Deep learning object-detection models are being successfully applied to develop computer-aided diagnosis systems for aiding polyp detection during colonoscopies. Here, we evidence the need to include negative samples for both (i) reducing false positives during the polyp-finding phase, by including images with artifacts that may confuse the detection models (e.g., medical instruments, water jets, feces, blood, excessive proximity of the camera to the colon wall, blurred images, etc.) that are usually not included in model development datasets, and (ii) correctly estimating a more realistic performance of the models. By retraining our previously developed YOLOv3-based detection model with a dataset that includes 15% of additional not-polyp images with a variety of artifacts, we were able to generally improve its F1 performance in our internal test datasets (from an average F1 of 0.869 to 0.893), which now include such type of images, as well as in four public datasets that include not-polyp images (from an average F1 of 0.695 to 0.722).
深度学习目标检测模型正成功应用于开发计算机辅助诊断系统，以辅助在结肠镜检查期间检测息肉。在此，我们证明有必要纳入阴性样本，这是为了：（i）在息肉检测阶段减少误报，通过纳入可能会混淆检测模型的带有伪影的图像（例如医疗器械、水喷流、粪便、血液、摄像头与结肠壁过度靠近、图像模糊等），而这些图像通常不包含在模型开发数据集中；以及（ii）正确估计模型更实际的性能。通过使用包含15%带有各种伪影的额外非息肉图像的数据集对我们之前开发的基于YOLOv3的检测模型进行重新训练，我们能够在我们的内部测试数据集中总体上提高其F1性能（从平均F1值0.869提高到0.893），现在这些内部测试数据集包含了此类图像，并且在四个包含非息肉图像的公共数据集中也提高了性能（从平均F1值0.695提高到0.722）。
