Abstract: Techniques for automated surgeon performance evaluation include dressing an operator with colored surgical gloves. A first glove for a dominant hand has a first color and a second glove for the other hand has a different second color. Video data that views the operator's hands is captured during a surgical procedure on a subject. For each of multiple frames of the video data, a minimum rectangle of pixels, called a first rectangle, which encloses pixels having the first color, is determined automatically on a processor. A first time series for a representative property of the first rectangle at the multiple frames, and a first measure of entropy based on the first time series, are also automatically determined on a processor. A metric of operator performance based at least in part on the first measure of entropy is stored, e.g., for subsequent display.

Abstract: Techniques for automated surgeon performance evaluation include dressing an operator with colored surgical gloves. A first glove for a dominant hand has a first color and a second glove for the other hand has a different second color. Video data that views the operator's hands is captured during a surgical procedure on a subject. For each of multiple frames of the video data, a minimum rectangle of pixels, called a first rectangle, which encloses pixels having the first color, is determined automatically on a processor. A first time series for a representative property of the first rectangle at the multiple frames, and a first measure of entropy based on the first time series, are also automatically determined on a processor. A metric of operator performance based at least in part on the first measure of entropy is stored, e.g., for subsequent display.