Search

Micromanipulation refers to the manipulation under a microscope in order to perform delicate procedures. It is difficult for humans to manipulate objects accurately under a microscope due to tremor and imperfect perception, limiting performance. This project seeks to understand factors affecting accuracy in micromanipulation, and to propose strategies for learning improving accuracy. Psychomotor experiments were conducted using computer-controlled setups to determine how various feedback modalities and learning methods can influence micromanipulation performance. In a first experiment, static and motion accuracy of surgeons, medical students and non-medical students under different magniification levels and grip force settings were compared. A second experiment investigated whether the non-dominant hand placed close to the target can contribute to accurate pointing of the dominant hand. A third experiment tested a training strategy for micromanipulation using unstable dynamics to magnify motion error, a strategy shown to be decreasing deviation in large arm movements. Two virtual reality (VR) modules were then developed to train needle grasping and needle insertion tasks, two primitive tasks in a microsurgery suturing procedure. The modules provided the trainee with a visual display in stereoscopic view and information on their grip, tool position and angles. Using the VR module, a study examining effects of visual cues was conducted to train tool orientation. Results from these studies suggested that it is possible to learn and improve accuracy in micromanipulation using appropriate sensorimotor feedback and training.