Automated Image Clarity Detection for the Improvement of Colposcopy Imaging with Multiple Devices

The proportion of women dying from cervical cancer in middle- and low-income countries is over 60%, twice that of their high-income counterparts. A primary screening strategy to eliminate this burden is cervix visualization and application of 3-5% acetic acid, inducing contrast in potential lesions. Recently, machine learning tools have emerged to aid visual diagnosis. As low-cost visualization tools expand, it is important to maximize image quality at the time of the exam or of images that are used in algorithms. Objective: In this study, we present the use of an object detection algorithm, the YOLOv5 model, to identify cervix regions of interest and describe blur within a multi-device image database. Methods: We took advantage of the Fourier domain to provide pseudo-labeling of training and testing images. A YOLOv5 model was trained using Pocket Colposcope, Mobile ODT EVA, and standard of care digital colposcope images. The former two are mobile devices. Results: When tested on all devices, this model achieved a mean average precision score, sensitivity, and specificity of 0.9, 0.89, and 0.89, respectively, reflecting the generalizability of the algorithm. Compared to physician annotation, it yielded an accuracy of 0.72. Conclusion: This method provides an informed quantitative analysis of captured images, can be generalized across different imaging devices, and is highly concordant with expert annotation. Significance: This quality control framework can assist in the standardization of colposcopy workflow, data acquisition, and image analysis and in doing so increase the availability of usable positive images for the development of deep learning algorithms.