Machine learning-based Automatic Evaluation of Tissue Handling Skills in Laparoscopic Colorectal Surgery: A Retrospective Experimental Study.

OBJECTIVE: To develop a machine learning model that automatically quantifies the spread of blood in the surgical field using intraoperative videos of laparoscopic colorectal surgery and evaluate whether the index measured with the developed model can be used to assess tissue handling skill. BACKGROUND: Although skill evaluation is crucial in laparoscopic surgery, existing evaluation systems suffer from evaluator subjectivity and are labor-intensive. Therefore, automatic evaluation using machine learning is potentially useful. MATERIALS AND METHODS: In this retrospective experimental study, we used training data with annotated labels of blood or non-blood pixels on intraoperative images to develop a machine learning model to classify pixel RGB values into blood and non-blood. The blood pixel count per frame (the total number of blood pixels throughout a surgery divided by the number of frames) was compared among groups of surgeons with different tissue handling skills. RESULTS: The overall accuracy of the machine learning model for the blood classification task was 85.7%. The high tissue handling skill group had the lowest blood pixel count per frame, and the novice surgeon group had the highest count (mean [SD]: high tissue handling skill group 20972.23 [19287.05] vs. low tissue handling skill group 34473.42 [28144.29] vs. novice surgeon group 50630.04 [42427.76], P <0.01). The difference between any 2 groups was significant. CONCLUSIONS: We developed a machine learning model to measure blood pixels in laparoscopic colorectal surgery images using RGB information. The blood pixel count per frame measured with this model significantly correlated with surgeons' tissue handling skills.

Cytoreductive surgery plus hyperthermic intraperitoneal chemotherapy versus R0 resection for resectable colorectal cancer with peritoneal metastases and low peritoneal cancer index scores: a collaborative observational study from Korea and Japan.

BACKGROUND: The benefits of hyperthermic intraperitoneal chemotherapy (HIPEC) after cytoreductive surgery (CRS) for colorectal cancer with peritoneal metastasis (CPM) remain controversial. R0 resection without peritoneal stripping might be as effective as CRS plus HIPEC. The authors aimed to compare the long-term oncological outcomes of patients with CPM and peritoneal cancer index (PCI) scores less than or equal to 6 who underwent R0 resection in Japan with those who underwent CRS plus HIPEC in Korea. MATERIALS AND METHODS: This international, retrospective cohort study was conducted in Korea and Japan using a prospectively collected clinical database. Patients who underwent surgery from July 2014 to December 2021 for CPM with a PCI score of less than or equal to 6 and completeness of the cytoreduction score-0 were included. The primary outcome was relapse-free survival (RFS), and the secondary outcomes were overall survival, peritoneal RFS (PRFS), and postoperative outcomes. RESULTS: The 3-year RFS was significantly longer in the CRS+HIPEC group than in the R0 resection group: 35.9% versus 6.9% ( P <0.001); 31.0% versus 6.7% ( P =0.040) after propensity score matching. The median PRFS was significantly longer in the CRS+HIPEC group than in the R0 resection group: 24.5 months versus 17.2 months ( P =0.017). The 3-year overall survival and postoperative complications did not significantly differ between the two groups. CONCLUSIONS: RFS and PRFS rates were significantly prolonged after CRS plus HIPEC, whereas postoperative complications and length of hospital stay were not increased. Therefore, curative CRS plus HIPEC may be considered a treatment strategy for selected patients with resectable CPM and low PCI scores.