Surgical Hand Gesture Recognition Utilizing Electroencephalogram as Input to the Machine Learning and Network Neuroscience Algorithms.

Surgical gestures detection can provide targeted, automated surgical skill assessment and feedback during surgical training for robot-assisted surgery (RAS). Several sources including surgical videos, robot tool kinematics, and an electromyogram (EMG) have been proposed to reach this goal. We aimed to extract features from electroencephalogram (EEG) data and use them in machine learning algorithms to classify robot-assisted surgical gestures. EEG was collected from five RAS surgeons with varying experience while performing 34 robot-assisted radical prostatectomies over the course of three years. Eight dominant hand and six non-dominant hand gesture types were extracted and synchronized with associated EEG data. Network neuroscience algorithms were utilized to extract functional brain network and power spectral density features. Sixty extracted features were used as input to machine learning algorithms to classify gesture types. The analysis of variance (ANOVA) F-value statistical method was used for feature selection and 10-fold cross-validation was used to validate the proposed method. The proposed feature set used in the extra trees (ET) algorithm classified eight gesture types performed by the dominant hand of five RAS surgeons with an accuracy of 90%, precision: 90%, sensitivity: 88%, and also classified six gesture types performed by the non-dominant hand with an accuracy of 93%, precision: 94%, sensitivity: 94%.

Prevalence and Predictors of Venous Thromboembolism After Robot-Assisted Radical Cystectomy.

OBJECTIVE: To describe incidence and variables associated with venous thromboembolism (VTE) after robot-assisted radical cystectomy (RARC). METHODS: A retrospective review of the prospectively maintained departmental database was performed. Extended thromboprophylaxis (for 4 weeks postoperatively) was implemented November 2017. Patients were divided into VTE (deep venous thrombosis [DVT] and/or pulmonary embolism [PE]) and non-VTE groups. Baseline demographics, disease characteristics and perioperative outcomes were compared. Cochran-Armitage trend test was used to assess trends of VTE. Multivariate logistic regression was used to identify variables associated with VTE. The Kaplan-Meier method was used to depict recurrence free survival (RFS), disease specific survival (DSS), and overall survival (OS). RESULTS: Twenty nine patients (5%) developed VTE (14 developed DVT and 15 developed PE). Median time to DVT was 28 days and to PE was 23 days after RARC. The rate of VTE remained stable between 2005 and 2020 (P= .99). Patients who developed VTE had significantly higher BMI (31 vs 29, P = .04), had COPD more often (34% vs 14%, P < .01) and had longer median hospital stay (8 vs 7 days, P = .01). Multivariate analysis showed that BMI (odds ratio [OR] 1.05; 95% confidence interval [CI] 1.005-1.10; P = .03), COPD (OR 3.24; 95% CI 1.43-7.30; P < .01),and non-organ confined disease (OR 2.73; 95% CI 1.22-6.11; P = 0.01) were associated with VTE. Kaplan-Meier curves showed that patients who developed VTE exhibited similar RFS (79% vs 64%, P = .28), DSS (90% vs 76%, P = .17), and OS (54% vs 52%, P = .76) at 5 years compared to those who did not develop VTE. CONCLUSION: VTE remains a significant complication after RARC. Higher BMI, COPD, and non-organ confined disease were significantly associated with VTE.

Development and Cross-Validation of a Nomogram for Chronic Kidney Disease Following Robot-Assisted Radical Cystectomy.

Introduction: We sought to identify the factors associated with deterioration of renal functions after robot-assisted radical cystectomy, and to develop a nomogram to detect the probability of progression to chronic kidney disease (CKD). Materials and Methods: A retrospective review of our prospectively maintained database. Estimated glomerular filtration rate (eGFR) was calculated using the CKD-Epidemiology Collaboration creatinine formula utilizing all follow-up creatinine values. CKD was defined as stage 3b (eGFR <45 mL/minute/1.73 m2) based on the National Kidney Foundation classification. Kaplan-Meier curves were used to depict CKD-free survival. A multivariate Cox regression model was used to determine predictors for CKD and to build the perioperative nomogram. Results: The data set comprised 442 patients with a median follow-up of 25 months (12-59). Thirty-seven percent developed CKD at a median of 9 months (4-18). CKD-free survival rates at 1, 3, and 5 years were 75%, 58%, and 50%, respectively. CKD was significantly associated with preoperative eGFR (hazards ratio [HR]: 0.96, 95% confidence interval [CI]: 0.95-0.97, p < 0.01), body mass index (HR: 1.03, 95% CI: 1.01-1.05, p = 0.03), Charlson Comorbidity Index ≥3 (HR: 2.20, 95% CI: 1.35-3.58, p < 0.01), diabetes (HR: 1.59, 95% CI: 1.09-2.31, p = 0.02), 90 days postoperative strictures (HR: 4.04, 95% CI: 1.76-9.30, p < 0.01), 90 days postoperative hydronephrosis (HR: 2.26, 95% CI: 1.34-3.79, p < 0.01), 90 days recurrent urinary tract infection (HR: 1.84, 95% CI: 1.08-3.14, p = 0.02), 90 days acute kidney injury (HR: 1.70, 95% CI: 1.19-2.43, p < 0.01), and node positive disease (HR: 1.94, 95% CI: 1.31-2.86, p < 0.01). A 5-year CKD-free survival nomogram was developed. Conclusion: We have developed and cross-validated a nomogram for detecting CKD-free survival. This nomogram may have a role in counseling and follow up of patients. This study was done after the approval of the IRB committee (I-79606).