Machine learning for technical skill assessment in surgery: a systematic review.

Accurate and objective performance assessment is essential for both trainees and certified surgeons. However, existing methods can be time consuming, labor intensive, and subject to bias. Machine learning (ML) has the potential to provide rapid, automated, and reproducible feedback without the need for expert reviewers. We aimed to systematically review the literature and determine the ML techniques used for technical surgical skill assessment and identify challenges and barriers in the field. A systematic literature search, in accordance with the PRISMA statement, was performed to identify studies detailing the use of ML for technical skill assessment in surgery. Of the 1896 studies that were retrieved, 66 studies were included. The most common ML methods used were Hidden Markov Models (HMM, 14/66), Support Vector Machines (SVM, 17/66), and Artificial Neural Networks (ANN, 17/66). 40/66 studies used kinematic data, 19/66 used video or image data, and 7/66 used both. Studies assessed the performance of benchtop tasks (48/66), simulator tasks (10/66), and real-life surgery (8/66). Accuracy rates of over 80% were achieved, although tasks and participants varied between studies. Barriers to progress in the field included a focus on basic tasks, lack of standardization between studies, and lack of datasets. ML has the potential to produce accurate and objective surgical skill assessment through the use of methods including HMM, SVM, and ANN. Future ML-based assessment tools should move beyond the assessment of basic tasks and towards real-life surgery and provide interpretable feedback with clinical value for the surgeon.PROSPERO: CRD42020226071.

An Observational Analysis of a Novel Digital Rectoscope.

BACKGROUND: This is an analysis of the first 50 in-human uses of a novel digital rigid sigmoidoscope. The technology provides digital image capture, telemedicine capabilities, improved ergonomics, and the ability to biopsy under pneumorectum while maintaining the low cost of conventional rigid sigmoidoscopy. The primary outcome was adverse events, and the secondary outcome was diagnostic view. PRELIMINARY RESULTS: Fifty patients underwent outpatient (n = 25) and surgical rectal assessment (n = 25), with a mean age of 60 years. This included 31 men and 19 women with 12 different clinical use indications. No adverse events were reported, and no defects were reported with the instrumentation. Satisfactory diagnoses were obtained in 48 (96%) of 50 uses, images were captured in 48 (96%) of 50 uses, and biopsies were successfully taken in 13 uses (26%). No adverse events were recorded. Independent reviewers of recorded videos agreed on the quality and diagnostic value of the images with a κ of 0.225 (95% CI, 0.144-0.305) when assessing whether the target pathology was adequately visualized. IMPACT OF INNOVATION: The improved views afforded by digital rectoscopy facilitated a satisfactory clinical diagnosis in 96% of uses. The device was successfully deployed in the operating room and outpatients irrespective of bowel preparation method, where it has the potential to replace flexible sigmoidoscopy for specific use cases. The technology provides a high-quality image and video that can be securely recorded for documentation and medicolegal purposes with agreement between blinded users despite a lack of standardized training and heterogenous pathology. We perceive significant impact of this technology for the assessment of colorectal anastomoses, the office management of colitis, "watch and wait," and for diagnostic support in rectal cancer diagnosis. The technology has significant potential to facilitate proctoring and training, and it now requires prospective trials to validate its diagnostic accuracy against more costly flexible sigmoidoscopy systems.

Insights from a global snapshot of the change in elective colorectal practice due to the COVID-19 pandemic.

BACKGROUND: There is a need to understand the impact of COVID-19 on colorectal cancer care globally and determine drivers of variation. OBJECTIVE: To evaluate COVID-19 impact on colorectal cancer services globally and identify predictors for behaviour change. DESIGN: An online survey of colorectal cancer service change globally in May and June 2020. PARTICIPANTS: Attending or consultant surgeons involved in the care of patients with colorectal cancer. MAIN OUTCOME MEASURES: Changes in the delivery of diagnostics (diagnostic endoscopy), imaging for staging, therapeutics and surgical technique in the management of colorectal cancer. Predictors of change included increased hospital bed stress, critical care bed stress, mortality and world region. RESULTS: 191 responses were included from surgeons in 159 centers across 46 countries, demonstrating widespread service reduction with global variation. Diagnostic endoscopy was reduced in 93% of responses, even with low hospital stress and mortality; whilst rising critical care bed stress triggered complete cessation (p = 0.02). Availability of CT and MRI fell by 40-41%, with MRI significantly reduced with high hospital stress. Neoadjuvant therapy use in rectal cancer changed in 48% of responses, where centers which had ceased surgery increased its use (62 vs 30%, p = 0.04) as did those with extended delays to surgery (p<0.001). High hospital and critical care bed stresses were associated with surgeons forming more stomas (p<0.04), using more experienced operators (p<0.003) and decreased laparoscopy use (critical care bed stress only, p<0.001). Patients were also more actively prioritized for resection, with increased importance of co-morbidities and ICU need. CONCLUSIONS: The COVID-19 pandemic was associated with severe restrictions in the availability of colorectal cancer services on a global scale, with significant variation in behaviours which cannot be fully accounted for by hospital burden or mortality.

Optical Technologies for Endoscopic Real-Time Histologic Assessment of Colorectal Polyps: A Meta-Analysis.

OBJECTIVES: Accurate, real-time, endoscopic risk stratification of colorectal polyps would improve decision-making and optimize clinical efficiency. Technologies to manipulate endoscopic optical outputs can be used to predict polyp histology in vivo; however, it remains unclear how accuracy has progressed and whether it is sufficient for routine clinical implementation. METHODS: A meta-analysis was conducted by searching MEDLINE, Embase, and the Cochrane Library. Studies were included if they prospectively deployed an endoscopic optical technology for real-time in vivo prediction of adenomatous colorectal polyps. Polyposis and inflammatory bowel diseases were excluded. Bayesian bivariate meta-analysis was performed, presenting 95% confidence intervals (CI). RESULTS: One hundred two studies using optical technologies on 33,123 colorectal polyps were included. Digital chromoendoscopy differentiated neoplasia (adenoma and adenocarcinoma) from benign polyps with sensitivity of 92.2% (90.6%-93.9% CI) and specificity of 84.0% (81.5%-86.3% CI), with no difference between constituent technologies (narrow-band imaging, Fuji intelligent Chromo Endoscopy, iSCAN) or with only diminutive polyps. Dye chromoendoscopy had sensitivity of 92.7% (90.1%-94.9% CI) and specificity of 86.6% (82.9%-89.9% CI), similarly unchanged for diminutive polyps. Spectral analysis of autofluorescence had sensitivity of 94.4% (84.0%-99.1% CI) and specificity of 50.9% (13.2%-88.8% CI). Endomicroscopy had sensitivity of 93.6% (85.3%-98.3% CI) and specificity of 92.5% (81.8%-98.1% CI). Computer-aided diagnosis had sensitivity of 88.9% (74.2%-96.7% CI) and specificity of 80.4% (52.6%-95.7% CI). Prediction confidence and endoscopist experience alone did not significantly improve any technology. The only subgroup to demonstrate a negative predictive value for adenoma above 90% was digital chromoendoscopy, making high confidence predictions of diminutive recto-sigmoid polyps. Chronologic meta-analyses show a falling negative predictive value over time. A significant publication bias exists. DISCUSSION: This novel approach to meta-analysis demonstrates that existing optical technologies are increasingly unlikely to allow safe "resect and discard" strategies and that step-change innovation may be required. A "diagnose and leave" strategy may be supported for diminutive recto-sigmoid polyps diagnosed with high confidence; however, limitations exist in the evidence base for this cohort.

Postoperative hypothermia and surgical site infection following peritoneal insufflation with warm, humidified carbon dioxide during laparoscopic colorectal surgery: a cohort study with cost-effectiveness analysis.

BACKGROUND: Surgical Site Infection (SSI) occurs in 9 % of laparoscopic colorectal surgery. Warming and humidifying carbon dioxide (CO2) used for peritoneal insufflation may protect against SSI by avoiding postoperative hypothermia (itself a risk factor for SSI). This study aimed to assess the impact of CO2 conditioning on postoperative hypothermia and SSI and to perform a cost-effectiveness analysis. METHODS: A retrospective cohort study of patients undergoing elective laparoscopic colorectal resection was performed at a single UK specialist centre. The control group (n = 123) received peritoneal insufflation with room temperature, dry CO2, whereas the intervention group (n = 123) received warm, humidified CO2 (using HumiGard™, Fisher & Paykel Healthcare). The outcomes were postoperative hypothermia, SSI and costs. Multivariate analysis was performed. RESULTS: A total of 246 patients were included in the study. The mean age was 68 (20-87) and mean BMI 28 (15-51). The primary diagnosis was cancer (n = 173), and there were no baseline differences between the groups. CO2 conditioning significantly decreased the incidence of postoperative hypothermia (odds ratio 0.10, 95 % CI 0.04-0.23), with hypothermic patients found to be at increased risk of SSI (odds ratio 4.0, 95 % CI 1.25-12.9). Use of conditioned CO2 significantly decreased the incidence of SSI by 66 % (p = 0.04). The intervention group incurred costs of £155 less per patient. The incremental cost-effectiveness ratio was negative. CONCLUSION: CO2 conditioning during laparoscopic colorectal surgery is a safe, feasible and a cost-effective intervention. It improves the quality of surgical care relating to SSI and postoperative hypothermia.