Source: Deep Learning on Medium
New research from Harvard Medical Group shows AI-powered autonomous device navigation is possible in minimally invasive surgical procedures, and even in heart surgery.
Minimally invasive procedures involve navigating devices such as transcatheter valves, occlusion devices, and stent grafts from a small incision in the patient’s skin to the site of the intervention via catheter. Navigating a catheter involves tracking its location while controlling the forces it applies to surrounding tissue.
Harvard researchers employed two techniques: Wall following and Haptic vision.
In nature, wall following is a method used by creatures to track object boundaries under low-visibility conditions. Inspired by this approach, Harvard medical group designed positively thigmotactic algorithms to create low-force contact with heart tissue and follow the tissue walls to reach desired locations.
Another contribution of this work is applying haptic vision to tissue visualisation. In wall following, locally recapturing detailed visual features is necessary. Haptic vision is a sensing modality at the catheter tip that combines intracardiac endoscopy, machine learning and image processing algorithms to provide clear images of what the catheter tip is touching and identify how hard it is pressing. The researcher’s aim was to overcome the limitations of traditional tissue visualisation with noisy and low resolution images.
To reduce procedural trauma and risk in cardiac procedures it is better to avoid stopping the heart and placing the patient on cardiopulmonary bypass. Such heart surgeries however are complicated by both opaque blood and constantly moving cardiac tissue. Researchers conducted autonomous catheter navigation in beating hearts in in vivo animal experiments and compared the results to expert manual navigation in terms of time and efficiency.
Overall autonomous navigation results were more stable in terms of time (with less time variance). All experiments, including autonomous navigation, had high success rates.
It is hoped that autonomous navigation can relieve clinicians from manually controlling instruments and free them to focus on other critical components of a surgical procedure.
The paper Autonomous Robotic Intracardiac Catheter Navigation Using Haptic Vision is in Science Magazine.
Author: Hecate He | Editor: Michael Sarazen
2018 Fortune Global 500 Public Company AI Adaptivity Report is out!
Purchase a Kindle-formatted report on Amazon.
Apply for Insight Partner Program to get a complimentary full PDF report.
Follow us on Twitter @Synced_Global for daily AI news!
We know you don’t want to miss any stories. Subscribe to our popular Synced Global AI Weekly to get weekly AI updates.