Deep learning for DNase I hypersensitive sites identification.

BACKGROUND: The DNase I hypersensitive sites (DHSs) are associated with the cis-regulatory DNA elements. An efficient method of identifying DHSs can enhance the understanding on the accessibility of chromatin. Despite a multitude of resources available on line including experimental datasets and computational tools, the complex language of DHSs remains incompletely understood. METHODS: Here, we address this challenge using an approach based on a state-of-the-art machine learning method. We present a novel convolutional neural network (CNN) which combined Inception like networks with a gating mechanism for the response of multiple patterns and longterm association in DNA sequences to predict multi-scale DHSs in Arabidopsis, rice and Homo sapiens. RESULTS: Our method obtains 0.961 area under curve (AUC) on Arabidopsis, 0.969 AUC on rice and 0.918 AUC on Homo sapiens. CONCLUSIONS: Our method provides an efficient and accurate way to identify multi-scale DHSs sequences by deep learning.

A Deep-Learning-Based Guidewire Compliant Control Method for the Endovascular Surgery Robot.

Endovascular surgery is a high-risk operation with limited vision and intractable guidewires. At present, endovascular surgery robot (ESR) systems based on force feedback liberates surgeons' operation skills, but it lacks the ability to combine force perception with vision. In this study, a deep learning-based guidewire-compliant control method (GCCM) is proposed, which guides the robot to avoid surgical risks and improve the efficiency of guidewire operation. First, a deep learning-based model called GCCM-net is built to identify whether the guidewire tip collides with the vascular wall in real time. The experimental results in a vascular phantom show that the best accuracy of GCCM-net is 94.86 ± 0.31%. Second, a real-time operational risk classification method named GCCM-strategy is proposed. When the surgical risks occur, the GCCM-strategy uses the result of GCCM-net as damping and decreases the robot's running speed through virtual resistance. Compared with force sensors, the robot with GCCM-strategy can alleviate the problem of force position asynchrony caused by the long and soft guidewires in real-time. Experiments run by five guidewire operators show that the GCCM-strategy can reduce the average operating force by 44.0% and shorten the average operating time by 24.6%; therefore the combination of vision and force based on deep learning plays a positive role in improving the operation efficiency in ESR.