Search

Colorectal cancer has the second highest incidence of malignant tumors and is the fourth leading cause of cancer deaths in China. Early diagnosis and treatment of colorectal cancer will lead to an improvement in the 5-year survival rate, which will reduce medical costs. The current diagnostic methods for early colorectal cancer include excreta, blood, endoscopy, and computer-aided endoscopy. In this paper, research on image analysis and prediction of colorectal cancer lesions based on deep learning is reviewed with the goal of providing a reference for the early diagnosis of colorectal cancer lesions by combining computer technology, 3D modeling, 5G remote technology, endoscopic robot technology, and surgical navigation technology. The findings will supplement the research and provide insights to improve the cure rate and reduce the mortality of colorectal cancer.

Background Compared to traditional open surgery, laparoscopic surgery has become a standard approach for colorectal cancer due to its great superiorities including less postoperative pain, a shorter hospital stay, and better quality of life. In 2007, Whiteford et al reported the first natural orifice trans-anal endoscopic surgery (NOTES) sigmoidectomy using transanal endoscopic microsurgery. To date, all cases of NOTES colorectal resection have included a hybrid laparoscopic approach with the use of established rigid platforms. Aim To introduce a novel technique of peroral external traction-assisted transanal NOTES rectosigmoidectomy followed by intracorporeal colorectal end-to-end anastomosis by using only currently available and flexible endoscopic instrumentation in a live porcine model. Methods Three female pigs weighing 25-30 kg underwent NOTES rectosigmoid resection. After preoperative work-up and bowel preparation, general anesthesia combined with endotracheal intubation was achieved. One dual-channel therapeutic endoscope was used. Carbon dioxide insufflation was performed during the operation. The procedure of trans-anal NOTES rectosigmoidectomy included the following eight steps: (1) The rectosigmoid colon was tattooed with India ink by submucosal injection; (2) Creation of gastrostomy by directed submucosal tunneling; (3) Peroral external traction using endoloop ligation; (4) Creation of rectostomy on the anterior rectal wall by directed 3 cm submucosal tunneling; (5) Peroral external traction-assisted dissection of the left side of the colon; (6) Trans-anal rectosigmoid specimen transection, where an anvil was inserted into the proximal segment after purse-string suturing; (7) Intracorporeal colorectal end-to-end anastomosis using a circular stapler by a single stapling technique; and (8) Closure of gastrostomy using endoscopic clips. All animals were euthanized immediately after the procedure, abdominal exploration was performed, and the air-under-water leak test was carried out. Results The procedure was completed in all three animals, with the operation time ranging from 193 min to 259 min. Neither major intraoperative complications nor hemodynamic instability occurred during the operation. The length of the resected specimen ranged from 7 cm to 13 cm. With the assistance of a trans-umbilical rigid grasper, intracorporeal colorectal, tension-free, end-to-end anastomosis was achieved in the three animals. Conclusion Peroral traction-assisted transanal NOTES rectosigmoidectomy followed by intracorporeal colorectal end-to-end anastomosis is technically feasible and reproducible in an animal model and is worthy of further improvements.