Tornike Davitashvili
Papers
5
Total Citations
179
H-Index
5
About
Tornike Davitashvili is a leading researcher at the intersection of artificial intelligence and robotic surgery, with a primary focus on developing cognitive assistance systems for minimally invasive procedures. His work spans surgical workflow analysis, deep learning for semantic segmentation, and multi-agent reinforcement learning for cooperative robotic assistance. Davitashvili made a landmark contribution with the HeiChole benchmark, a comparative validation of machine learning algorithms for surgical workflow and skill analysis that has garnered 96 citations, establishing a standard for evaluating AI in the operating room. His deep learning approach for semantic segmentation of organs and tissues in laparoscopic surgery (40 citations) is foundational for context-aware surgical assistance. Notably, his pioneering application of multi-agent reinforcement learning to cooperative robotic surgery (21 citations) demonstrates how autonomous systems can reduce surgeon fatigue and mental exertion. Davitashvili also introduced the concept of "surgomics" through an active learning study for extracting surgical process characteristics from robot-assisted esophagectomy (16 citations), aiming to personalize surgical outcome predictions. His work is driving the next generation of cognitive surgical tools that promise safer, more intelligent, and semi-autonomous robotic assistance.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5