Machine learning accurately measures tension in colorectal surgery

A novel machine learning algorithm based on long short-term memory neural networks successfully estimates mechanical tension in ex vivo porcine colons, achieving 88% accuracy and strong correlations in force measurement. This innovative approach provides a significant advance over traditional subjective tension assessment during robotic surgery, addressing the challenge of anastomotic leaks, which affect one in ten colorectal surgeries. The method’s adaptability to other tissue types could also enhance surgical outcomes.