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

PURPOSE: Surgical data science is a new research field that aims to observe all aspects of the patient treatment process in order to provide the right assistance 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. METHODS: 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. Core of the method is an auxiliary task based on raw endoscopic video data of the target domain that is used to initialize the convolutional neural network (CNN) for the target task. In this paper, we propose the re-colorization of medical images with a conditional generative adversarial network (cGAN)-based architecture as 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 validate both variants using medical instrument segmentation as target task. RESULTS: 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 75% 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 EndoVis2017 challenge) using the target task (in this instance: segmentation). CONCLUSION: As it makes efficient use of available (non-)public and (un-)labeled data, the approach has the potential to become a valuable tool for CNN (pre-)training.

The effect of application time and soil factors on the occurrence of Beauveria brongniartii applied as a biological control agent in soil.

The effects of abiotic and biotic soil factors on occurrence of the entomopathogenic fungus Beauveria brongniartii after application at different times of the year were examined in Switzerland. Applications made from May to August generally resulted in an increase of 1-5 x 10(3) CFU g(-1) dry soil compared to untreated control plots. Conversely, soils treated in October and November yielded no increase. Soil temperatures between 20 and 25 degrees C, and high clay content of the soil had a positive effect on the occurrence and density of B. brongniartii whereas increased catalase activity and temperatures above 27 degrees C had a negative influence. Laboratory experiments revealed that a higher number of CFUs developed after one month of incubation at 22 degrees C than at 12 degrees C. Differences were not detected after three months of incubation, indicating that growth rate was simply slower at sub-optimal temperatures. The increase was different in three native soils, but was not correlated with different clay contents of the soil. In sterilized soil, though, the differences were not detected, suggesting that biotic factors have a greater influence rather than soil texture.