Explore the vast range of titles available.

The intraoperative activity of a surgeon has substantial impact on postoperative outcomes. However, for most surgical procedures, the details of intraoperative surgical actions, which can vary widely, are not well understood. Here we report a machine learning system leveraging a vision transformer and supervised contrastive learning for the decoding of elements of intraoperative surgical activity from videos commonly collected during robotic surgeries. The system accurately identified surgical steps, actions performed by the surgeon, the quality of these actions and the relative contribution of individual video frames to the decoding of the actions. Through extensive testing on data from three different hospitals located in two different continents, we show that the system generalizes across videos, surgeons, hospitals and surgical procedures, and that it can provide information on surgical gestures and skills from unannotated videos. Decoding intraoperative activity via accurate machine learning systems could be used to provide surgeons with feedback on their operating skills, and may allow for the identification of optimal surgical behaviour and for the study of relationships between intraoperative factors and postoperative outcomes. A machine learning system leveraging a vision transformer and supervised contrastive learning accurately decodes elements of intraoperative surgical activity from videos commonly collected during robotic surgeries.

Aging may detrimentally affect cognitive and motor function. However, age is also associated with experience, and how these factors interplay and affect outcomes following surgery is unclear. We sought to evaluate the effect of surgeon age on postoperative outcomes in patients undergoing common surgical procedures. We performed a retrospective cohort study of patients undergoing 1 of 25 common surgical procedures in Ontario, Canada, from 2007 to 2015. We evaluated the association between surgeon age and a composite outcome of death, readmission and complications. We used generalized estimating equations for analysis, accounting for relevant patient-, procedure-, surgeon- and hospital-level factors. We found 1 159 676 eligible patients who were treated by 3314 surgeons and ranged in age from 27 to 81 years. Modelled as a continuous variable, a 10-year increase in surgeon age was associated with a 5% relative decreased odds of the composite outcome (adjusted odds ratio [OR] 0.95, 95% confidence interval [CI] 0.92 to 0.98, p = 0.002). Considered dichotomously, patients receiving treatment from surgeons who were older than 65 years of age had a 7% lower odds of adverse outcomes (adjusted OR 0.93, 95% CI 0.88–0.97, p = 0.03; crude absolute difference = 3.1%). We found that increasing surgeon age was associated with decreasing rates of postoperative death, readmission and complications in a nearly linear fashion after accounting for patient-, procedure-, surgeon- and hospital-level factors. Further evaluation of the mechanisms underlying these findings may help to improve patient safety and outcomes, and inform policy about maintenance of certification and retirement age for surgeons.

Background Surgeons who receive reliable feedback on their performance quickly master the skills necessary for surgery. Such performance-based feedback can be provided by a recently-developed artificial intelligence (AI) system that assesses a surgeon’s skills based on a surgical video while simultaneously highlighting aspects of the video most pertinent to the assessment. However, it remains an open question whether these highlights, or explanations, are equally reliable for all surgeons. Methods Here, we systematically quantify the reliability of AI-based explanations on surgical videos from three hospitals across two continents by comparing them to explanations generated by humans experts. To improve the reliability of AI-based explanations, we propose the strategy of training with explanations –TWIX –which uses human explanations as supervision to explicitly teach an AI system to highlight important video frames. Results We show that while AI-based explanations often align with human explanations, they are not equally reliable for different sub-cohorts of surgeons (e.g., novices vs. experts), a phenomenon we refer to as an explanation bias. We also show that TWIX enhances the reliability of AI-based explanations, mitigates the explanation bias, and improves the performance of AI systems across hospitals. These findings extend to a training environment where medical students can be provided with feedback today. Conclusions Our study informs the impending implementation of AI-augmented surgical training and surgeon credentialing programs, and contributes to the safe and fair democratization of surgery. Plain language summary Surgeons aim to master skills necessary for surgery. One such skill is suturing which involves connecting objects together through a series of stitches. Mastering these surgical skills can be improved by providing surgeons with feedback on the quality of their performance. However, such feedback is often absent from surgical practice. Although performance-based feedback can be provided, in theory, by recently-developed artificial intelligence (AI) systems that use a computational model to assess a surgeon’s skill, the reliability of this feedback remains unknown. Here, we compare AI-based feedback to that provided by human experts and demonstrate that they often overlap with one another. We also show that explicitly teaching an AI system to align with human feedback further improves the reliability of AI-based feedback on new videos of surgery. Our findings outline the potential of AI systems to support the training of surgeons by providing feedback that is reliable and focused on a particular skill, and guide programs that give surgeons qualifications by complementing skill assessments with explanations that increase the trustworthiness of such assessments. Kiyasseh et al. compare the quality of feedback provided to surgeons by artificial intelligence (AI) to that provided by human experts. Teaching an AI system to explicitly follow human explanations improves the reliability and reduces the bias of AI-based feedback.

Artificial intelligence (AI) systems can now reliably assess surgeon skills through videos of intraoperative surgical activity. With such systems informing future high-stakes decisions such as whether to credential surgeons and grant them the privilege to operate on patients, it is critical that they treat all surgeons fairly. However, it remains an open question whether surgical AI systems exhibit bias against surgeon sub-cohorts, and, if so, whether such bias can be mitigated. Here, we examine and mitigate the bias exhibited by a family of surgical AI systems—SAIS—deployed on videos of robotic surgeries from three geographically-diverse hospitals (USA and EU). We show that SAIS exhibits an underskilling bias, erroneously downgrading surgical performance, and an overskilling bias, erroneously upgrading surgical performance, at different rates across surgeon sub-cohorts. To mitigate such bias, we leverage a strategy —TWIX—which teaches an AI system to provide a visual explanation for its skill assessment that otherwise would have been provided by human experts. We show that whereas baseline strategies inconsistently mitigate algorithmic bias, TWIX can effectively mitigate the underskilling and overskilling bias while simultaneously improving the performance of these AI systems across hospitals. We discovered that these findings carry over to the training environment where we assess medical students’ skills today. Our study is a critical prerequisite to the eventual implementation of AI-augmented global surgeon credentialing programs, ensuring that all surgeons are treated fairly.

Background Chronic kidney disease (CKD) is associated with higher incidence of major surgery. No studies have evaluated the association between preoperative kidney function and postoperative outcomes across a wide spectrum of procedures. We aimed to evaluate the association between CKD and 30-day postoperative outcomes across surgical specialties. Methods We selected adult patients undergoing surgery across eight specialties. The primary study endpoint was major complications, defined as death, unplanned reoperation, cardiac complication, or stroke within 30 days following surgery. Secondary outcomes included Clavien-Dindo high-grade complications, as well as cardiac, pulmonary, infectious, and thromboembolic complications. Multivariable regression was performed to evaluate the association between CKD and 30-day postoperative complications, adjusted for baseline characteristics, surgical specialty, and operative time. Results In total, 1,912,682 patients were included. The odds of major complications (adjusted odds ratio [aOR] 2.14 [95% confidence interval (CI): 2.07, 2.21]), death (aOR 3.03 [95% CI: 2.88, 3.19]), unplanned reoperation (aOR 1.57 [95% CI: 1.51, 1.64]), cardiac complication (aOR 3.51 [95% CI: 3.25, 3.80]), and stroke (aOR 1.89 [95% CI: 1.64, 2.17]) were greater for patients with CKD stage 5 vs. stage 1. A similar pattern was observed for the secondary endpoints. Conclusion This population-based study demonstrates the negative impact of CKD on operative outcomes across a diverse range of procedures and patients.

Suturing skill scores have demonstrated strong predictive capabilities for patient functional recovery. The suturing can be broken down into several substep components, including needle repositioning , needle entry angle , etc. Artificial intelligence (AI) systems have been explored to automate suturing skill scoring. Traditional approaches to skill assessment typically focus on evaluating individual sub-skills required for particular substeps in isolation. However, surgical procedures require the integration and coordination of multiple sub-skills to achieve successful outcomes. Significant associations among the technical sub-skill have been established by existing studies. In this paper, we propose a framework for joint skill assessment that takes into account the interconnected nature of sub-skills required in surgery. The prior known relationships among sub-skills are firstly identified. Our proposed AI system is then empowered by the prior known relationships to perform the suturing skill scoring for each sub-skill domain simultaneously. Our approach can effectively improve skill assessment performance through the prior known relationships among sub-skills. Through the proposed approach to joint skill assessment, we aspire to enhance the evaluation of surgical proficiency and ultimately improve patient outcomes in surgery.

The transformation of prostatic epithelial cells to prostate cancer (PCa) has been characterized as a transition from citrate secretion to citrate oxidation, from which one would anticipate enhanced mitochondrial complex I (CI) respiratory flux. Molecular mechanisms for this transformation are attributed to declining mitochondrial zinc concentrations. The unique metabolic properties of PCa cells have become a hot research area. Several publications have provided indirect evidence based on investigations using pre-clinical models, established cell lines, and fixed or frozen tissue bank samples. However, confirmatory respiratory analysis on fresh human tissue has been hampered by multiple difficulties. Thus, few mitochondrial respiratory assessments of freshly procured human PCa tissue have been published on this question. Our objective is to document relative mitochondrial CI and complex II (CII) convergent electron flow to the Q-junction and to identify electron transport system (ETS) alterations in fresh PCa tissue. The results document a CII succinate: quinone oxidoreductase (SQR) dominant succinate oxidative flux model in the fresh non-malignant prostate tissue, which is enhanced in malignant tissue. CI NADH: ubiquinone oxidoreductase activity is impaired rather than predominant in high-grade malignant fresh prostate tissue. Given these novel findings, succinate and CII are promising targets for treating and preventing PCa.

The limitations of current forms of prostate cancer therapy have driven researchers to search for new alternatives. Previously we showed cytopathic effect related to HSV-tk in prostate cancer. In this study we present initial results of a neoadjuvant HSV-tk gene therapy trial and address some of the potential mechanistic aspects of its effect in human tissues. We enrolled 23 men with clinically localized prostate cancer but high risk for recurrence in this Phase I-II trial. Intraprostatic viral injections (one to four) were followed by 2 weeks of ganciclovir and prostatectomy 2-4 weeks later. Toxicity was modest. Surgical specimens were embedded fully and whole-mount slides were imaged and analyzed for areas of cytopathic effect. The larger the tumor the greater the cytopathic effect. The effect also seems to be related to areas of high CAR expression. However, the number of injection sites did not influence effect. Local (CD8+ cells and macrophages) and systemic immune response (CD8+ and activated CD8+, IL-12) was increased in patients treated with HSV-tk. Increased apoptosis and decreased microvessel density were also noted in these patients. The results suggest a tumor-specific effect mediated by systemic and local immune response, antiangiogenic effect, and modulation of apoptosis.

Purpose To assess the clinical performance of ProsTAV ® , a blood-based test based on telomere associate variables (TAV) measurement, to support biopsy decision-making when diagnosing suspicious prostate cancer (PCa). Methods Preliminary data of a prospective observational pragmatic study of patients with prostate-specific antigen (PSA) levels 3–10 ng/ml and suspicious PCa. Results were combined with other clinical data, and all patients underwent prostate biopsies according to each center’s routine clinical practice, while magnetic resonance imaging (MRI) before the prostate biopsy was optional. Sensitivity, specificity, positive and negative predicted values, and subjects where biopsies could have been avoided using ProsTAV were determined. Results The mean age of the participants ( n  = 251) was 67.4 years, with a mean PSA of 5.90 ng/ml, a mean free PSA of 18.9%, and a PSA density of 0.14 ng/ml. Digital rectal examination was abnormal in 21.1% of the subjects, and according to biopsy, the prevalence of significant PCa was 47.8%. The area under the ROC curve of ProsTAV was 0.7, with a sensitivity of 0.90 (95% CI, 0.85–0.95) and specificity of 0.27 (95% CI, 0.19–0.34). The positive and negative predictive values were 0.53 (95% CI, 0.46–0.60) and 0.74 (95% CI, 0.62–0.87), respectively. ProsTAV could have reduced the biopsies performed by 27% and showed some initial evidence of a putative benefit in the diagnosis pathway combined with MRI. Conclusions ProsTAV increases the prediction capacity of significant PCa in patients with PSA between 3 and 10 ng/ml and could be considered a complementary tool to improve the patient diagnosis pathway.

Surgical patients can be discharged to a variety of facilities which vary widely in intensity of care. Postoperative readmissions have been found to be more strongly associated with post-discharge events than pre-discharge complications, indicating the importance of discharge destination. We sought to evaluate the association between discharge destination and 30-day outcomes. A retrospective cohort study was conducted using the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) database. Patients were dichotomized based on discharge destination: home versus non-home. The main outcome of interest was 30-day unplanned readmission. The secondary outcomes included post-discharge pulmonary, infectious, thromboembolic, and bleeding complications, as well as death. In this cohort study of over 1.5 million patients undergoing common surgical procedures across eight surgical specialties, we found non-home discharge to be associated with adverse 30-day post-operative outcomes, namely, unplanned readmissions, post-discharge pulmonary, infectious, thromboembolic, and bleeding complications, as well as death. Non-home discharge is associated with worse 30-day outcomes among patients undergoing common surgical procedures. Patients and caregivers should be counseled regarding discharge destination, as non-home discharge is associated with adverse post-operative outcomes.