Exploiting the potential of unlabeled endoscopic video data with self-supervised learning

Surgical data science is a new research field that aims to observe all aspects and factors of the patient treatment process in order to provide the right assistance to the right person at the right time. Due to the breakthrough successes of deep learning-based solutions for automatic image annotation, the availability of reference annotations for algorithm training is becoming a major bottleneck in the field. The purpose of this paper was to investigate the concept of self-supervised learning to address this issue. Our approach is guided by the hypothesis that unlabeled video data can be used to learn a representation of the target domain that boosts the performance of state-of-the-art machine learning algorithms when used for pre-training. Essentially, this method involves an auxiliary task that requires training with unlabeled endoscopic video data from the target domain to initialize a convolutional neural network (CNN) for the target task. In this paper, we propose to undertake a re-colorization of medical images with generative adversarial network (GAN)-based architecture as an auxiliary task. A variant of the method involves a second pre-training step based on labeled data for the target task from a related domain. We have validated both variants using medical instrument segmentation as the target task. The proposed approach can be used to radically reduce the manual annotation effort involved in training CNNs. Compared to the baseline approach of generating annotated data from scratch, our method decreases exploratively the number of labeled images by up to 60% without sacrificing performance. Our method also outperforms alternative methods for CNN pre-training, such as pre-training on publicly available non-medical (COCO) or medical data (MICCAI endoscopic vision challenge 2017) using the target task (in this instance: segmentation).