Computer assisted minimally invasive surgery: is medical computer vision the answer to improving laparosurgery?

Minimally Invasive Surgery (MIS) is one of the most effective methods of modern surgical intervention that has considerable advantages compared with open surgery, including reduced trauma, pain, and post-operative recovery time. MIS has improved substantially over the years, chiefly due to new hardware innovations, including HD cameras and flexible head endoscopes. However, MIS continues to be hindered by several problems. In addition to hardware innovation, Computer Vision (CV) has been proposed as a way to overcome some of its current limitations. However, the research literature lacks a coherent picture of how the limitations can be best overcome by hardware, CV or a combination of the two. In this paper we focus on laparoscopic MIS, and list these limitations into 5 clear categories. We detail the effectiveness of hardware and CV solutions with respect to each limitation, from which we base the following hypothesis: CV is both complementary and necessary to hardware development, to overcome all 5 limitations in laparoscopy. Our paper is of value to laparoscopy surgeons, by conveying what is expected to be achieved in computer-aided laparoscopy over the next decade. It is also of value to medical CV researchers, by clarifying which problems are best solved with CV, in light of the hardware developments likely to occur over the next decade.