Objective: To develop an artificial intelligence–based pipeline to assess and triage patient-submitted postoperative wound images. Background: The rise of outpatient surgeries, remote monitoring, and patient-submitted wound images via online portals has contributed to a growing administrative workload on clinicians. Early identification of surgical site infection (SSI) is essential for reducing postoperative morbidity. Methods: Patients ≥18 years old who underwent surgery at 9 affiliated Mayo Clinic hospitals (2019-2022) and were captured by the National Surgical Quality Improvement Program (NSQIP) were included. Eligibility required submission of one image via the patient portal within 30 days after surgery. Images were independently screened in duplicate to determine the presence of an incision. SSI outcomes were obtained from NSQIP. The developed model consisted of 2 stages: incision detection and SSI detection in images with incisions. Four pretrained architectures were evaluated using 10-fold cross-validation, with upsampling and data augmentation to mitigate class imbalance. An end-to-end pipeline, image quality assessment and sensitivity analysis stratified by race were also performed. Results: Among 6060 patients, the median age was 54 years (interquartile range: 40–65), 61.4% (n=3805) were female, and 92.5% (n=5731) identified as White. SSIs were confirmed in 6.2% (n=386) images. Vision Transformer outperformed all others, achieving an incision detection accuracy of 0.94 (area under the curve=0.98) and an SSI detection accuracy of 0.73 (area under the curve=0.81). In addition, it demonstrated strong performance in assessing image quality. Sensitivity analysis revealed comparable performance across racial subgroups. Conclusion: This artificial intelligence pipeline demonstrates promising performance in automating wound image assessment and SSI detection, reducing clinical workload and improving postoperative care.

Background: Robotic surgery was integrated into some healthcare systems despite there being few well designed, real-world studies on safety or benefit. This study compared the safety of robotic with laparoscopic, thoracoscopic, and open approaches in common robotic procedures. Methods: This was a population-based, retrospective study of all adults who underwent prostatectomy, hysterectomy, pulmonary lobectomy, or partial nephrectomy in Ontario, Canada, between 2008 and 2018. The primary outcome was 90-day total adverse events using propensity score overlap weights, and secondary outcomes were minor or major morbidity/adverse events. Results: Data on 24 741 prostatectomy, 75 473 hysterectomy, 18 252 pulmonary lobectomy, and 6608 partial nephrectomy operations were included. Relative risks for total adverse events in robotic compared with open surgery were 0.80 (95 per cent c.i. 0.74 to 0.87) for radical prostatectomy, 0.44 (0.37 to 0.52) for hysterectomy, 0.53 (0.44 to 0.65) for pulmonary lobectomy, and 0.72 (0.54 to 0.97) for partial nephrectomy. Relative risks for total adverse events in robotic surgery compared with a laparoscopic/thoracoscopic approach were 0.94 (0.77 to 1.15), 1.00 (0.82 to 1.23), 1.01 (0.84 to 1.21), and 1.23 (0.82 to 1.84) respectively. Conclusion: The robotic approach is associated with fewer adverse events than an open approach but similar to a laparoscopic/thoracoscopic approach. The benefit of the robotic approach is related to the minimally-invasive approach rather than the platform itself.