Search

Surgery is a major source of errors in patient care. Preventing complications from surgical errors in the operating room is estimated to lead to reduction of up to 41,846 readmissions and save $620.3 million per year. It is now established that poor technical skill is associated with an increased risk of severe adverse events after surgery and traditional models to train surgeons are being challenged by rapid advances in technology, an intensified patient-safety culture, and a need for value-driven health systems. This review discusses the current methods available for evaluating technical skills in cataract surgery as well as the recent technological advancements that have enabled capture and analysis of large amounts of complex surgical data for more automated objective skills assessment.

Purpose: Surgery involves modifying anatomy to achieve a goal. Reconstructing anatomy can facilitate surgical care through surgical planning, real-time decision support, or anticipating outcomes. Tool motion is a rich source of data that can be used to quantify anatomy. Our work develops and validates a method for reconstructing the nasal septum from unstructured motion of the Cottle elevator during the elevation phase of septoplasty surgery, without need to explicitly delineate the surface of the septum. Approach: The proposed method uses iterative closest point registration to initially register a template septum to the tool motion. Subsequently, statistical shape modeling with iterative most likely oriented point registration is used to fit the reconstructed septum to Cottle tip position and orientation during flap elevation. Regularization of the shape model and transformation is incorporated. The proposed methods were validated on 10 septoplasty surgeries performed on cadavers by operators of varying experience level. Preoperative CT images of the cadaver septums were segmented as ground truth. Results: We estimated reconstruction error as the difference between the projections of the Cottle tip onto the surface of the reconstructed septum and the ground-truth septum segmented from the CT image. We found translational differences of 2.74(2.06-2.81) mm and a rotational differences of 8.95(7.11-10.55) deg between the reconstructed septum and the ground-truth septum [median (interquartile range)], given the optimal regularization parameters. Conclusions: Accurate reconstruction of the nasal septum can be achieved from tool tracking data during septoplasty surgery on cadavers. This enables understanding of the septal anatomy without need for traditional medical imaging. This result may be used to facilitate surgical planning, intraoperative care, or skills assessment.

Objective assessment of intraoperative technical skill is necessary for technology to improve patient care through surgical training. Our objective in this study was to develop and validate deep learning techniques for technical skill assessment using videos of the surgical field. We used a data set of 99 videos of capsulorhexis, a critical step in cataract surgery. One expert surgeon annotated each video for technical skill using a standard structured rating scale, the International Council of Ophthalmology’s Ophthalmology Surgical Competency Assessment Rubric:phacoemulsification (ICO-OSCAR:phaco). Using two capsulorhexis indices in this scale (commencement of flap and follow-through, formation and completion), we specified an expert performance when at least one of the indices was 5 and the other index was at least 4, and novice otherwise. In addition, we used scores for capsulorhexis commencement and capsulorhexis formation as separate ground truths (Likert scale of 2 to 5; analyzed as 2/3, 4 and 5). We crowdsourced annotations of instrument tips. We separately modeled instrument trajectories and optical flow using temporal convolutional neural networks to predict a skill class (expert/novice) and score on each item for capsulorhexis in ICO-OSCAR:phaco. We evaluated the algorithms in a five-fold cross-validation and computed accuracy and area under the receiver operating characteristics curve (AUC). The accuracy and AUC were 0.848 and 0.863 for instrument tip velocities, and 0.634 and 0.803 for optical flow fields, respectively. Deep neural networks effectively model surgical technical skill in capsulorhexis given structured representation of intraoperative data such as optical flow fields extracted from video or crowdsourced tool localization information.

This study was designed to determine the effect of a novel simulation-based training curriculum for scleral tunnel construction in manual small incision cataract surgery (MSICS) compared with traditional training. In this multicenter, investigator-masked, randomized clinical trial, resident surgeons within 3 months of matriculation with minimal or no prior experience with MSICS were assigned either to simulation-based training, the Experimental Group (EG), or to conventional training, the Control Group (CG). EG residents were trained to perform scleral tunnel construction using a simulation-based curriculum (HelpMeSee Eye Surgery Simulator), while residents in the CG followed institution-specific curriculum before progressing to live surgery. Surgical videos of the first 20 attempts at tunnel construction were reviewed by masked video raters. The primary outcome was the total number of any of 9 pre-specified errors. On average, the total number of errors was 9.25 (95% CI 0–18.95) in the EG and 17.56 (95% CI 6.63–28.49) in the CG (P = 0.05); the number of major errors was 4.86 (95% CI 0.13–9.59) in the EG and 10.09 (95% CI 4.76–15.41) in the CG (P = 0.02); and the number of minor errors was 4.39 (95% CI 0–9.75) in the EG and 7.47 (95% CI 1.43–13.51) in the CG (P = 0.16). These results support that novice surgeons trained using the novel simulation-based curriculum performed fewer errors in their first 20 attempts at tunnel construction compared to those trained with a conventional curriculum.

BACKGROUND: Ophthalmia neonatorum is an infection of the eyes in newborns that can lead to blindness, particularly if the infection is caused by Neisseria gonorrhoeae. Antiseptic or antibiotic medication is dispensed into the eyes of newborns, or dispensed systemically, soon after delivery to prevent neonatal conjunctivitis and potential vision impairment. OBJECTIVES: 1. To determine if any type of systemic or topical eye medication is better than placebo or no prophylaxis in preventing ophthalmia neonatorum. 2. To determine if any one systemic or topical eye medication is better than any other medication in preventing ophthalmia neonatorum. SEARCH METHODS: We searched CENTRAL, MEDLINE, Embase, LILACS, and three trials registers, date of last search 4 October 2019. We also searched references of included studies and contacted pharmaceutical companies. SELECTION CRITERIA: We included randomised and quasi‐randomised controlled trials of any topical, systemic, or combination medical interventions used to prevent ophthalmia neonatorum in newborns compared with placebo, no prophylaxis, or with each other. DATA COLLECTION AND ANALYSIS: We used standard methods expected by Cochrane. Outcomes were: blindness or any adverse visual outcome at 12 months, conjunctivitis at 1 month (gonococcal (GC), chlamydial (CC), bacterial (BC), any aetiology (ACAE), or unknown aetiology (CUE)), and adverse effects. MAIN RESULTS: We included 30 trials with a total of 79,198 neonates. Eighteen studies were conducted in high‐income settings (the USA, Europe, Israel, Canada), and 12 were conducted in low‐ and middle‐income settings (Africa, Iran, China, Indonesia, Mexico). Fifteen of the 30 studies were quasi‐randomised. We judged every study to be at high risk of bias in at least one domain. Ten studies included a comparison arm with no prophylaxis. There were 14 different prophylactic regimens and 12 different medications in the 30 included studies. Any prophylaxis compared to no prophylaxis Unless otherwise indicated, the following evidence comes from studies assessing one or more of the following interventions: tetracycline 1%, erythromycin 0.5%, povidone‐iodine 2.5%, silver nitrate 1%. None of the studies reported data on the primary outcomes: blindness or any adverse visual outcome at any time point. There was only very low‐certainty evidence on the risk of GC with prophylaxis (4/5340 newborns) compared to no prophylaxis (5/2889) at one month (risk ratio (RR) 0.79, 95% confidence interval (CI) 0.24 to 2.65, 3 studies). Low‐certainty evidence suggested there may be little or no difference in effect on CC (RR 0.96, 95% CI 0.57 to 1.61, 4874 newborns, 2 studies) and BC (RR 0.84, 95% CI 0.37 to 1.93, 3685 newborns, 2 studies). Moderate‐certainty evidence suggested a probable reduction in risk of ACAE at one month (RR 0.65, 95% 0.54 to 0.78, 9666 newborns, 8 studies assessing tetracycline 1%, erythromycin 0.5%, povidone‐iodine 2.5%, silver nitrate 1%, colostrum, bacitracin‐phenacaine ointment). There was only very low‐certainty evidence on CUE (RR 1.75, 95% CI 0.37 to 8.28, 330 newborns, 1 study). Very low‐certainty evidence on adverse effects suggested no increased nasolacrimal duct obstruction (RR 0.93, 95% CI 0.68 to 1.28, 404 newborns, 1 study of erythromycin 0.5% and silver nitrate 1%) and no increased keratitis (single study of 40 newborns assessing silver nitrate 1% with no events). Any prophylaxis compared to another prophylaxis Overall, evidence comparing different interventions did not suggest any consistently superior intervention. However, most of this evidence was of low‐certainty and was extremely limited. AUTHORS' CONCLUSIONS: There are no data on whether prophylaxis for ophthalmia neonatorum prevents serious outcomes such as blindness or any adverse visual outcome. Moderate‐certainty evidence suggests that the use of prophylaxis may lead to a reduction in the incidence of ACAE in newborns but the evidence for effect on GC, CC or BC was less certain. Comparison of individual interventions did not suggest any consistently superior intervention, but data were limited. A trial comparing tetracycline, povidone‐iodine (single administration), and chloramphenicol for GC and CC could potentially provide the community with an effective, universally applicable prophylaxis against ophthalmia neonatorum.

Lumbar spine surgery suffers from high variability in patient outcomes,1–9 heterogeneous treatment patterns, and a high frequency of undesirable outcomes, such as revision surgery,8 failed back surgery,7 and recurrent lumbar disk herniations.10 This heterogeneity could be attributed to variable patient characteristics, surgeon preferences, hospital and surgeon volume, and complexity of the intervention.11–13 However, current understanding of the contributing factors that determine the outcome variability remains ambiguous and underexplored.14–16 In this context, models to predict patient outcomes provide a valuable resource for patient selection, treatment optimization, and rehabilitative pathway design.11,17–19 Previous studies that investigated variability in pain and functional outcomes were limited to analyzing patient demographic and preoperative clinical characteristics as the underlying factors.11,12,19 While such factors [e.g., age, body mass index (BMI), and smoking status] partially contribute to the outcome variability, additional surgery-specific features could further illuminate and improve the utility of prediction models. Some studies have tended toward development of increasingly complex modeling techniques (e.g., deep neural networks14,20) to improve accuracy, particularly for scenarios in which the predictor variables arise from a limited number of available preoperative characteristics (i.e., demographic data). In spite of devising complex machine learning algorithms, model generalizability is challenging to achieve with a limited set of predictor variables that may not fully explain the variability in clinical outcomes. Even though advanced learning methods perform better under conditions of sparse data and/or limited input variables, increasingly complex models also introduce challenges to explainability in the predictions and the ability to identify actionable features. A large number of images are routinely acquired for perioperative diagnostic and therapeutic evaluation purposes before, during, and after spine surgery. These images play a vital role in diagnosing spinal pathology, determining surgery indications, ensuring safe intervention via intraoperative guidance, and assessing the surgical product in postoperative and follow-up visits. Such images capture distinctive features related to anatomy, pathology, and the changes effected during surgery. In fact, qualitative radiological assessment has been an important basis for outcome measurement prior to (or complementary to) patient-reported outcomes (PROs). Thus, incorporating automatically derived image features in addition to previously investigated patient characteristics could improve the accuracy and utility of prediction models. Consequently, several studies have evaluated the relationship between individual measures manually obtained from images and patient outcomes in lumbar, thoracic, and cervical spine surgery. Example measures include Modic changes,21,22 signal intensity-based metrics,23–25 and the lowest instrumented vertebra take-off angle.26,27 In contrast to manually obtained features, recent advances facilitate the automatic extraction of several anatomical features from spine imaging.28–31 Example image-based features include automatic labeling of vertebral levels, global spinal alignment, endplate angles, intervertebral space, number of levels treated, and measurements of surgical instrumentation. To our knowledge, there are no previous studies evaluating statistical models to predict patient outcomes in lumbar spine surgery that use automatically derived quantitative measurements from radiological images. The objective of this study was to develop and validate a data-analytic framework referred to as “SpineCloud,” which includes quantitative features automatically derived from spine imaging in addition to patient demographic and clinical characteristics to predict function and pain improvement after lumbar spine surgery. A learning algorithm was trained and validated to make predictions by combining disparate features based on demographics and image analytics. In this work, we used a boosted decision tree algorithm that inherently handles ordinal and continuous variables in a common predictor variable space. The prediction task was modeled as a binary classification to estimate functional improvement or nonimprovement at 3- and 12-month intervals after surgery.

Healthcare in general, and surgery/interventional care in particular, is evolving through rapid advances in technology and increasing complexity of care, with the goal of maximizing the quality and value of care. Whereas innovations in diagnostic and therapeutic technologies have driven past improvements in the quality of surgical care, future transformation in care will be enabled by data. Conventional methodologies, such as registry studies, are limited in their scope for discovery and research, extent and complexity of data, breadth of analytical techniques, and translation or integration of research findings into patient care. We foresee the emergence of surgical/interventional data science (SDS) as a key element to addressing these limitations and creating a sustainable path toward evidence-based improvement of interventional healthcare pathways. SDS will create tools to measure, model, and quantify the pathways or processes within the context of patient health states or outcomes and use information gained to inform healthcare decisions, guidelines, best practices, policy, and training, thereby improving the safety and quality of healthcare and its value. Data are pervasive throughout the surgical care pathway; thus, SDS can impact various aspects of care, including prevention, diagnosis, intervention, or postoperative recovery. The existing literature already provides preliminary results, suggesting how a data science approach to surgical decision-making could more accurately predict severe complications using complex data from preoperative, intraoperative, and postoperative contexts, how it could support intraoperative decision-making using both existing knowledge and continuous data streams throughout the surgical care pathway, and how it could enable effective collaboration between human care providers and intelligent technologies. In addition, SDS is poised to play a central role in surgical education, for example, through objective assessments, automated virtual coaching, and robot-assisted active learning of surgical skill. However, the potential for transforming surgical care and training through SDS may only be realized through a cultural shift that not only institutionalizes technology to seamlessly capture data but also assimilates individuals with expertise in data science into clinical research teams. Furthermore, collaboration with industry partners from the inception of the discovery process promotes optimal design of data products as well as their efficient translation and commercialization. As surgery continues to evolve through advances in technology that enhance delivery of care, SDS represents a new knowledge domain to engineer surgical care of the future.

Easy acquisition of surgical data opens many opportunities to automate skill evaluation and teaching. Current technology to search tool motion data for surgical activity segments of interest is limited by the need for manual pre-processing, which can be prohibitive at scale. We developed a content-based information retrieval method, query-by-example (QBE), to automatically detect activity segments within surgical data recordings of long duration that match a query. The example segment of interest (query) and the surgical data recording (target trial) are time series of kinematics. Our approach includes an unsupervised feature learning module using a stacked denoising autoencoder (SDAE), two scoring modules based on asymmetric subsequence dynamic time warping (AS-DTW) and template matching, respectively, and a detection module. A distance matrix of the query against the trial is computed using the SDAE features, followed by AS-DTW combined with template scoring, to generate a ranked list of candidate subsequences (substrings). To evaluate the quality of the ranked list against the ground-truth, thresholding conventional DTW distances and bipartite matching are applied. We computed the recall, precision, F1-score, and a Jaccard index-based score on three experimental setups. We evaluated our QBE method using a suture throw maneuver as the query, on two tool motion datasets (JIGSAWS and MISTIC-SL) captured in a training laboratory. We observed a recall of 93, 90 and 87 % and a precision of 93, 91, and 88 % with same surgeon same trial (SSST), same surgeon different trial (SSDT) and different surgeon (DS) experiment setups on JIGSAWS, and a recall of 87, 81 and 75 % and a precision of 72, 61, and 53 % with SSST, SSDT and DS experiment setups on MISTIC-SL, respectively. We developed a novel, content-based information retrieval method to automatically detect multiple instances of an activity within long surgical recordings. Our method demonstrated adequate recall across different complexity datasets and experimental conditions.

Automatically segmenting and classifying surgical activities is an important prerequisite to providing automated, targeted assessment and feedback during surgical training. Prior work has focused almost exclusively on recognizing gestures, or short, atomic units of activity such as pushing needle through tissue, whereas we also focus on recognizing higher-level maneuvers, such as suture throw. Maneuvers exhibit more complexity and variability than the gestures from which they are composed, however working at this granularity has the benefit of being consistent with existing training curricula.Prior work has focused on hidden Markov model and conditional-random-field-based methods, which typically leverage unary terms that are local in time and linear in model parameters. Because maneuvers are governed by long-term, nonlinear dynamics, we argue that the more expressive unary terms offered by recurrent neural networks (RNNs) are better suited for this task. Four RNN architectures are compared for recognizing activities from kinematics: simple RNNs, long short-term memory, gated recurrent units, and mixed history RNNs. We report performance in terms of error rate and edit distance, and we use a functional analysis-of-variance framework to assess hyperparameter sensitivity for each architecture.We obtain state-of-the-art performance for both maneuver recognition from kinematics (4 maneuvers; error rate of [Formula: see text]; normalized edit distance of [Formula: see text]) and gesture recognition from kinematics (10 gestures; error rate of [Formula: see text]; normalized edit distance of [Formula: see text]).Automated maneuver recognition is feasible with RNNs, an exciting result which offers the opportunity to provide targeted assessment and feedback at a higher level of granularity. In addition, we show that multiple hyperparameters are important for achieving good performance, and our hyperparameter analysis serves to aid future work in RNN-based activity recognition.

IMPORTANCE: Daytime sleepiness in surgical trainees can impair intraoperative technical skill and thus affect their learning and pose a risk to patient safety. OBJECTIVE: To determine the association between daytime sleepiness of surgeons in residency and fellowship training and their intraoperative technical skill during septoplasty. DESIGN, SETTING, AND PARTICIPANTS: This prospective cohort study included 19 surgical trainees in otolaryngology–head and neck surgery programs at 2 academic institutions (Johns Hopkins University School of Medicine and MedStar Georgetown University Hospital). The physicians were recruited from June 13, 2016, to April 20, 2018. The analysis includes data that were captured between June 27, 2016, and April 20, 2018. MAIN OUTCOMES AND MEASURES: Attending physician and surgical trainee self-rated intraoperative technical skill using the Septoplasty Global Assessment Tool (SGAT) and visual analog scales. Daytime sleepiness reported by surgical trainees was measured using the Epworth Sleepiness Scale (ESS). RESULTS: Of 19 surgical trainees, 17 resident physicians (9 female [53%]) and 2 facial plastic surgery fellowship physicians (1 female and 1 male) performed a median of 3.00 septoplasty procedures (range, 1-9 procedures) under supervision by an attending physician. Of the 19 surgical trainees, 10 (53%) were aged 25 to 30 years and 9 (47%) were 31 years or older. The mean ESS score overall was 6.74 (95% CI, 5.96-7.52), and this score did not differ between female and male trainees. The mean ESS score was 7.57 (95% CI, 6.58-8.56) in trainees aged 25 to 30 years and 5.44 (95% CI, 4.32-6.57) in trainees aged 31 years or older. In regression models adjusted for sex, age, postgraduate year, and technical complexity of the procedure, there was a statistically significant inverse association between ESS scores and attending physician–rated technical skill for both SGAT (−0.41; 95% CI, −0.55 to −0.27; P

Automating objective assessment of surgical technical skill is necessary to support training and professional certification at scale, even in settings with limited access to an expert surgeon. Likewise, automated surgical activity recognition can improve operating room workflow efficiency, teaching and self-review, and aid clinical decision support systems. However, current supervised learning methods to do so, rely on large training datasets. Crowdsourcing has become a standard in curating such large training datasets in a scalable manner. The use of crowdsourcing in surgical data annotation and its effectiveness has been studied only in a few settings. In this study, we evaluated reliability and validity of crowdsourced annotations for information on surgical instruments (name of instruments and pixel location of key points on instruments). For 200 images sampled from videos of two cataract surgery procedures, we collected 9 independent annotations per image. We observed an inter-rater agreement of 0.63 (Fleiss’ kappa), and an accuracy of 0.88 for identification of instruments compared against an expert annotation. We obtained a mean pixel error of 5.77 pixels for annotation of instrument tip key points. Our study shows that crowdsourcing is a reliable and accurate alternative to expert annotations to identify instruments and instrument tip key points in videos of cataract surgery.

Background Surgeons who perform laparotomy have a number of decisions to make regarding abdominal closure. Material and size of potential suture types varies widely. In addition, surgeons can choose to close the incision in anatomic layers or mass ('en masse'), as well as using either a continuous or interrupted suturing technique, of which there are different styles of each. There is ongoing debate as to which suturing techniques and suture materials are best for achieving definitive wound closure while minimising the risk of short- and long-term complications. Objectives The objectives of this review were to identify the best available suture techniques and suture materials for closure of the fascia following laparotomy incisions, by assessing the following comparisons: absorbable versus non-absorbable sutures; mass versus layered closure; continuous versus interrupted closure techniques; monofilament versus multifilament sutures; and slow absorbable versus fast absorbable sutures. Our objective was not to determine the single best combination of suture material and techniques, but to compare the individual components of abdominal closure. Search methods On 8 February 2017 we searched CENTRAL, MEDLINE, Embase, two trials registries, and Science Citation Index. There were no limitations based on language or date of publication. We searched the reference lists of all included studies to identify trials that our searches may have missed. Selection criteria We included randomised controlled trials (RCTs) that compared suture materials or closure techniques, or both, for fascial closure of laparotomy incisions. We excluded trials that compared only types of skin closures, peritoneal closures or use of retention sutures. Data collection and analysis We abstracted data and assessed the risk of bias for each trial. We calculated a summary risk ratio (RR) for the outcomes assessed in the review, all of which were dichotomous. We used random-effects modelling, based on the heterogeneity seen throughout the studies and analyses. We completed subgroup analysis planned a priori for each outcome, excluding studies where interventions being compared differed by more than one component, making it impossible to determine which variable impacted on the outcome, or the possibility of a synergistic effect. We completed sensitivity analysis, excluding trials with at least one trait with high risk of bias. We assessed the quality of evidence using the GRADEpro guidelines. Main results Fifty-five RCTs with a total of 19,174 participants met the inclusion criteria and were included in the meta-analysis. Included studies were heterogeneous in the type of sutures used, methods of closure and patient population. Many of the included studies reported multiple comparisons. For our primary outcome, the proportion of participants who developed incisional hernia at one year or more of follow-up, we did not find evidence that suture absorption (absorbable versus non-absorbable sutures, RR 1.07, 95% CI 0.86 to 1.32, moderate-quality evidence; or slow versus fast absorbable sutures, RR 0.81, 95% CI 0.63 to 1.06, moderate-quality evidence), closure method (mass versus layered, RR 1.92, 95% CI 0.58 to 6.35, very low-quality evidence) or closure technique (continuous versus interrupted, RR 1.01, 95% CI 0.76 to 1.35, moderate-quality evidence) resulted in a difference in the risk of incisional hernia. We did, however, find evidence to suggest that monofilament sutures reduced the risk of incisional hernia when compared with multifilament sutures (RR 0.76, 95% CI 0.59 to 0.98, I2 = 30%, moderate-quality evidence). For our secondary outcomes, we found that none of the interventions reduced the risk of wound infection, whether based on suture absorption (absorbable versus non-absorbable sutures, RR 0.99, 95% CI 0.84 to 1.17, moderate-quality evidence; or slow versus fast absorbable sutures, RR 1.16, 95% CI 0.85 to 1.57, moderate-quality evidence), closure method (mass versus layered, RR 0.93, 95% CI 0.67 to 1.30, low-quality evidence) or closure technique (continuous versus interrupted, RR 1.13, 95% CI 0.96 to 1.34, moderate-quality evidence). Similarily, none of the interventions reduced the risk of wound dehiscence whether based on suture absorption (absorbable versus non-absorbable sutures, RR 0.78, 95% CI 0.55 to 1.10, moderate-quality evidence; or slow versus fast absorbable sutures, RR 1.55, 95% CI 0.92 to 2.61, moderate-quality evidence), closure method (mass versus layered, RR 0.69, 95% CI 0.31 to 1.52, moderate-quality evidence) or closure technique (continuous versus interrupted, RR 1.21, 95% CI 0.90 to 1.64, moderate-quality evidence). Absorbable sutures, compared with non-absorbable sutures (RR 0.49, 95% CI 0.26 to 0.94, low-quality evidence) reduced the risk of sinus or fistula tract formation. None of the other comparisons showed a difference (slow versus fast absorbable sutures, RR 0.88, 95% CI 0.05 to 16.05, very low-quality evidence; mass versus layered, RR 0.49, 95% CI 0.15 to 1.62, low-quality evidence; continuous versus interrupted, RR 1.51, 95% CI 0.64 to 3.61, very low-quality evidence). Authors' conclusions Based on this moderate-quality body of evidence, monofilament sutures may reduce the risk of incisional hernia. Absorbable sutures may also reduce the risk of sinus or fistula tract formation, but this finding is based on low-quality evidence. We had serious concerns about the design or reporting of several of the 55 included trials. The comparator arms in many trials differed by more than one component, making it impossible to attribute differences between groups to any one component. In addition, the patient population included in many of the studies was very heterogeneous. Trials included both emergency and elective cases, different types of disease pathology (e.g. colon surgery, hepatobiliary surgery, etc.) or different types of incisions (e.g. midline, paramedian, subcostal). Consequently, larger, high-quality trials to further address this clinical challenge are warranted. Future studies should ensure that proper randomisation and allocation techniques are performed, wound assessors are blinded, and that the duration of follow-up is adequate. It is important that only one type of intervention is compared between groups. In addition, a homogeneous patient population would allow for a more accurate assessment of the interventions.

This paper introduces Surgical Data Science as an emerging scientific discipline. Key perspectives are based on discussions during an intensive two-day international interactive workshop that brought together leading researchers working in the related field of computer and robot assisted interventions. Our consensus opinion is that increasing access to large amounts of complex data, at scale, throughout the patient care process, complemented by advances in data science and machine learning techniques, has set the stage for a new generation of analytics that will support decision-making and quality improvement in interventional medicine. In this article, we provide a consensus definition for Surgical Data Science, identify associated challenges and opportunities and provide a roadmap for advancing the field.

Background Aqueous shunts are employed to control intraocular pressure (IOP) for people with primary or secondary glaucomas who fail or are not candidates for standard surgery. Objectives To assess the effectiveness and safety of aqueous shunts for reducing IOP in glaucoma compared with standard surgery, another type of aqueous shunt, or modification to the aqueous shunt procedure. Search methods We searched CENTRAL (which contains the Cochrane Eyes and Vision Trials Register) (2016, Issue 8), MEDLINE Ovid (1946 to August 2016), Embase.com (1947 to August 2016), PubMed (1948 to August 2016), LILACS (Latin American and Caribbean Health Sciences Literature Database) (1982 to August 2016), ClinicalTrials.gov (www.clinicaltrials.gov); searched 15 August 2016, and the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP) (www.who.int/ictrp/search/en); searched 15 August 2016. We did not use any date or language restrictions in the electronic search for trials. We last searched the electronic databases on 15 August 2016. We also searched the reference lists of identified trial reports and the Science Citation Index to find additional trials. Selection criteria We included randomized controlled trials that compared various types of aqueous shunts with standard surgery or to each other in eyes with glaucoma. Data collection and analysis Two review authors independently screened search results for eligibility, assessed the risk of bias, and extracted data from included trials. We contacted trial investigators when data were unclear or not reported. We graded the certainty of the evidence using the GRADE approach. We followed standard methods as recommended by Cochrane. Main results We included 27 trials with a total of 2099 participants with mixed diagnoses and comparisons of interventions. Seventeen studies reported adequate methods of randomization, and seven reported adequate allocation concealment. Data collection and follow-up times varied. Four trials compared an aqueous shunt (Ahmed or Baerveldt) with trabeculectomy, of which three reported one-year outcomes. At one-year, the difference in IOP between aqueous shunt groups and trabeculectomy groups was uncertain (mean difference (MD) 2.55 mmHg, 95% confidence interval (CI) -0.78 to 5.87; 380 participants; very low-certainty evidence). The difference in logMAR visual acuity was also uncertain (MD 0.12 units, 95% CI -0.07 to 0.31; 380 participants; very low-certainty evidence). In two trials, the difference in visual field score was uncertain (MD -0.25, 95% CI -1.91 to 1.40; 196 participants; very low-certainty evidence). The mean number of antiglaucoma medications was higher in the aqueous shunt group than the trabeculectomy group in one trial (MD 0.80, 95% CI 0.48 to 1.12; 184 participants; low-certainty evidence). The effect on needing additional glaucoma surgery was uncertain between groups in two trials (risk ratio (RR) 0.24, 95% CI 0.04 to 1.36; 329 participants; very low-certainty evidence). In one trial, fewer total adverse events were reported in the aqueous shunt group than the trabeculectomy group (RR 0.59, 95% CI 0.43 to 0.81; 212 participants; very low-certainty evidence). No trial reported quality-of-life outcomes at one-year follow-up. Two trials that compared the Ahmed implant with the Baerveldt implant for glaucoma found higher mean IOP in the Ahmed group at one-year follow-up (MD 2.60 mmHg, 95% CI 1.58 to 3.62; 464 participants; moderate-certainty evidence). The difference in logMAR visual acuity was uncertain between groups (MD -0.07 units, 95% CI -0.27 to 0.13; 501 participants; low-certainty evidence). The MD in number of antiglaucoma medications was within one between groups (MD 0.35, 95% CI 0.11 to 0.59; 464 participants; moderate-certainty evidence). More participants in the Ahmed group required additional glaucoma surgery than the Baerveldt group (RR 2.77, 95% CI 1.02 to 7.54; 514 participants; moderate-certainty evidence). The two trials reported specific adverse events but not overall number of adverse events. Neither trial reported visual field or quality-of-life outcomes at one-year follow-up. One trial compared the Ahmed implant with the Molteno implant for glaucoma over two-year follow-up. Mean IOP was higher in the Ahmed group than the Molteno group (MD 1.64 mmHg, 95% CI 0.85 to 2.43; 57 participants; low-certainty evidence). The differences in logMAR visual acuity (MD 0.08 units, 95% CI -0.24 to 0.40; 57 participants; very low-certainty evidence) and mean deviation in visual field (MD -0.18 dB, 95% CI -3.13 to 2.77; 57 participants; very low-certainty evidence) were uncertain between groups. The mean number of antiglaucoma medications was also uncertain between groups (MD -0.38, 95% CI -1.03 to 0.27; 57 participants; low-certainty evidence). The trial did not report the proportion needing additional glaucoma surgery, total adverse events, or quality-of-life outcomes. Two trials compared the double-plate Molteno implant with the Schocket shunt for glaucoma; one trial reported outcomes only at six-month follow-up, and the other did not specify the follow-up time. At six-months, mean IOP was lower in the Molteno group than the Schocket group (MD -2.50 mmHg, 95% CI -4.60 to -0.40; 115 participants; low-certainty evidence). Neither trial reported the proportion needing additional glaucoma surgery, total adverse events, or visual acuity, visual field, or quality-of-life outcomes. The remaining 18 trials evaluated modifications to aqueous shunts, including 14 trials of Ahmed implants (early aqueous suppression versus standard medication regimen, 2 trials; anti-vascular endothelial growth factor agent versus none, 4 trials; corticosteroids versus none, 2 trials; shunt augmentation versus none, 3 trials; partial tube ligation versus none, 1 trial; pars plana implantation versus conventional implantation, 1 trial; and model M4 versus model S2,1 trial); 1 trial of 500 mm2 Baerveldt versus 350 mm2 Baerveldt; and 3 trials of Molteno implants (single-plate with oral corticosteroids versus single-plate without oral corticosteroids, 1 trial; double-plate versus single-plate, 1 trial; and pressure-ridge versus double-plate with tube ligation, 1 trial). Authors' conclusions Information was insufficient to conclude whether there are differences between aqueous shunts and trabeculectomy for glaucoma treatment. While the Baerveldt implant may lower IOP more than the Ahmed implant, the evidence was of moderate-certainty and it is unclear whether the difference in IOP reduction is clinically significant. Overall, methodology and data quality among existing randomized controlled trials of aqueous shunts was heterogeneous across studies, and there are no well-justified or widely accepted generalizations about the superiority of one surgical procedure or device over another.

Training skillful and competent surgeons is critical to ensure high quality of care and to minimize disparities in access to effective care. Traditional models to train surgeons are being challenged by rapid advances in technology, an intensified patient-safety culture, and a need for value-driven health systems. Simultaneously, technological developments are enabling capture and analysis of large amounts of complex surgical data. These developments are motivating a “surgical data science” approach to objective computer-aided technical skill evaluation (OCASE-T) for scalable, accurate assessment; individualized feedback; and automated coaching. We define the problem space for OCASE-T and summarize 45 publications representing recent research in this domain. We find that most studies on OCASE-T are simulation based; very few are in the operating room. The algorithms and validation methodologies used for OCASE-T are highly varied; there is no uniform consensus. Future research should emphasize competency assessment in the operating room, validation against patient outcomes, and effectiveness for surgical training.

Ocular infestation with Toxoplasma gondii, a parasite, may result in inflammation in the retina, choroid, and uvea and consequently lead to complications such as glaucoma, cataract, and posterior synechiae.The objective of this systematic review was to assess the effects of adjunctive use of corticosteroids for ocular toxoplasmosis.We searched CENTRAL (which contains the Cochrane Eyes and Vision Group Trials Register) (The Cochrane Library 2012, Issue 9), Ovid MEDLINE, Ovid MEDLINE In-Process and Other Non-Indexed Citations, Ovid MEDLINE Daily, Ovid OLDMEDLINE, (January 1950 to October 2012), EMBASE (January 1980 to October 2012), Latin American and Caribbean Literature on Health Sciences (LILACS) (January 1982 to October 2012), the metaRegister of Controlled Trials (mRCT) (www.controlled-trials.com), ClinicalTrials.gov (www.clinicaltrials.gov) and the WHO International Clinical Trials Registry Platform (ICTRP) (www.who.int/ictrp/search/en). We searched the reference lists of included studies for any additional studies not identified by the electronic searches. We did not use any date or language restrictions in the electronic searches for trials. We last searched the electronic databases on 11 October 2012.We planned to include randomized and quasi-randomized controlled trials. Eligible trials would have enrolled participants of any age who were immunocompetent and were diagnosed with active ocular toxoplasmosis. Included trials would have compared anti-parasitic therapy plus corticosteroids versus anti-parasitic therapy alone, or different doses or times of initiation of corticosteroids.Two authors independently screened titles and abstracts retrieved from the electronic searches. We retrieved full-text articles of studies categorized as 'unsure' or 'include' after review of the abstracts. Two authors independently reviewed each full-text article. Discrepancies were resolved through discussion.The electronic searches retrieved 368 titles and abstracts. We reviewed 20 full-text articles. We identified no trials eligible for inclusion in this systematic review.Although research has identified wide variation in practices regarding use of corticosteroids, our systematic review did not identify evidence from randomized controlled trials for the role of corticosteroids in the management of ocular toxoplasmosis. Several questions remain unanswered by well-conducted randomized trials in this context, including whether use of corticosteroids is more effective than use of anti-parasitic therapy alone, when corticosteroids should be initiated in the treatment regimen (early versus late course of treatment), and which dosage and duration of steroid use is best. These questions are easily amenable to research using a randomized controlled design and they are ethical due to the absence of evidence to support or discourage use of corticosteroids for this condition. The question of foremost importance, however, is whether they should be used as adjunct therapy (that is, additional) to anti-parasitic agents.

This is a protocol for a Cochrane Review (Intervention). The objectives are as follows: PRIMARY OBJECTIVES: The aim of this review is to evaluate the effects of MMC on the cornea and answer the question: is there sufficient evidence for using intraoperative topical MMC on the cornea in surface ablation (PRK or LASEK or epi‐LASIK) as prophylaxis for haze formation and to reduce refractive regression? This review will address the use of MMC in the prophylaxis of myopic patients who underwent PRK or LASEK or epi‐LASIK. SECONDARY OBJECTIVES: If sufficient data are available, a secondary objective of this review is to determine the refractive predictability and complications with different concentrations and times of exposure to MMC as well as to find out if there are differences in proportion of patients with haze among different laser platforms used for PRK, LASEK, and epi‐LASIK. Refractive predictability is defined as the difference between the attempted and achieved refractive change, in other words, the difference between the mean postoperative spherical equivalent (SE) and the target SE using the manifest refraction (Joo 2005).

Suturing is a challenging and highly dexterous task in minimally invasive surgery, even with the assistance of robotic surgical systems. In this work, we propose a simple yet versatile impedance virtual fixture framework, which can be applied on the master manipulator in a tele-operated robotic surgical system. With this framework, we further develop two types of virtual fixtures that assist with the needle passing and knot tying sub-tasks in suturing. The paper also presents the results of a 14-participant user study for both needle passing and knot tying sub-tasks, showing that virtual fixture assistance for novice users increases the needle passing exit point accuracy, reduces the number of errors (suture slip) in knot tying, and simultaneously decreases the task completion time and overall operator workload.

Robotic surgery and other minimally-invasive surgical techniques are an integral part of patient care, and readily yield large amounts of data. Surgical tool motion (kinematic data) contains information that is useful for assessment and education. Typically, assessment and education tools that rely upon the kinematic data require substantial manual processing such as activity annotations. The goal of this paper was to develop an automated method to align surgical recordings and assign activity annotations. We developed an approach based on unsupervised alignment to efficient annotate kinematic data for its constituent activity segments. Our method includes extracting non-linear features from the kinematic data using a stacked de-noising autoencoder, and using modified dynamic time warping to align the kinematic data from different trials of the study task. We combined alignment between a test and one or a small set of template trials (with prior manual annotations) with voting based on kernel density estimation to transfer labels from the template to the test trial. Our experiments on performance of this method using two datasets captured in the training laboratory demonstrate an accuracy of 72% to 94% for annotating activity segments within a surgical training task. Our findings are robust to data captured from several surgeons, and to deviations in activity from a canonical activity sequence.

We apply recurrent neural networks to the task of recognizing surgical activities from robot kinematics. Prior work in this area focuses on recognizing short, low-level activities, or gestures, and has been based on variants of hidden Markov models and conditional random fields. In contrast, we work on recognizing both gestures and longer, higher-level activites, or maneuvers, and we model the mapping from kinematics to gestures/maneuvers with recurrent neural networks. To our knowledge, we are the first to apply recurrent neural networks to this task. Using a single model and a single set of hyperparameters, we match state-of-the-art performance for gesture recognition and advance state-of-the-art performance for maneuver recognition, in terms of both accuracy and edit distance. Code is available at https://github.com/rdipietro/miccai-2016-surgical-activity-rec.

Background Surgical tasks are performed in a sequence of steps, and technical skill evaluation includes assessing task flow efficiency. Our objective was to describe differences in task flow for expert and novice surgeons for a basic surgical task. Methods We used a hierarchical semantic vocabulary to decompose and annotate maneuvers and gestures for 135 instances of a surgeon’s knot performed by 18 surgeons. We compared counts of maneuvers and gestures, and analyzed task flow by skill level. Results Experts used fewer gestures to perform the task (26.29; 95% CI = 25.21 to 27.38 for experts vs. 31.30; 95% CI = 29.05 to 33.55 for novices) and made fewer errors in gestures than novices (1.00; 95% CI = 0.61 to 1.39 vs. 2.84; 95% CI = 2.3 to 3.37). Transitions among maneuvers, and among gestures within each maneuver for expert trials were more predictable than novice trials. Conclusions Activity segments and state flow transitions within a basic surgical task differ by surgical skill level, and can be used to provide targeted feedback to surgical trainees.

Objective Task-level metrics of time and motion efficiency are valid measures of surgical technical skill. Metrics may be computed for segments (maneuvers and gestures) within a task after hierarchical task decomposition. Our objective was to compare task-level and segment (maneuver and gesture)-level metrics for surgical technical skill assessment. Design Our analyses include predictive modeling using data from a prospective cohort study. We used a hierarchical semantic vocabulary to segment a simple surgical task of passing a needle across an incision and tying a surgeon’s knot into maneuvers and gestures. We computed time, path length, and movements for the task, maneuvers, and gestures using tool motion data. We fit logistic regression models to predict experience-based skill using the quantitative metrics. We compared the area under a receiver operating characteristic curve (AUC) for task-level, maneuver-level, and gesture-level models. Setting Robotic surgical skills training laboratory. Participants In total, 4 faculty surgeons with experience in robotic surgery and 14 trainee surgeons with no or minimal experience in robotic surgery. Results Experts performed the task in shorter time (49.74 s; 95% CI = 43.27-56.21 vs. 81.97; 95% CI = 69.71-94.22), with shorter path length (1.63 m; 95% CI = 1.49-1.76 vs. 2.23; 95% CI = 1.91-2.56), and with fewer movements (429.25; 95% CI = 383.80-474.70 vs. 728.69; 95% CI = 631.84-825.54) than novices. Experts differed from novices on metrics for individual maneuvers and gestures. The AUCs were 0.79; 95% CI = 0.62-0.97 for task-level models, 0.78; 95% CI = 0.6-0.96 for maneuver-level models, and 0.7; 95% CI = 0.44-0.97 for gesture-level models. There was no statistically significant difference in AUC between task-level and maneuver-level (p = 0.7) or gesture-level models (p = 0.17). Conclusions Maneuver-level and gesture-level metrics are discriminative of surgical skill and can be used to provide targeted feedback to surgical trainees.

Background Cataract formation or acceleration can occur after intraocular surgery, especially following vitrectomy, a surgical technique for removing the vitreous that is used in the treatment of many disorders that affect the posterior segment of the eye. The underlying problem that led to vitrectomy may limit the benefit from removal of the cataractous lens. Objectives To evaluate the effectiveness and safety of surgery versus no surgery for postvitrectomy cataract with respect to visual acuity, quality of life, and other outcomes. Search methods We searched the Cochrane Central Register of Controlled Trials (CENTRAL) (which contains the Cochrane Eyes and Vision Trials Register) (2017, Issue 5), MEDLINE Ovid (1946 to 17 May 2017), Embase.com (1947 to 17 May 2017), PubMed (1946 to 17 May 2017), Latin American and Caribbean Health Sciences Literature database (LILACS) (January 1982 to 17 May 2017), the metaRegister of Controlled Trials (mRCT) (www.controlled-trials.com); last searched May 2013, ClinicalTrials.gov (www.clinicaltrials.gov); searched 17 May 2017, and the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP) (www.who.int/ictrp/search/en); searched 17 May 2017. We did not use any date or language restrictions in the electronic searches for trials. Selection criteria We planned to include randomized controlled trials (RCTs) and quasi-RCTs that had compared surgery versus no surgery to remove the lens from eyes of adults in which cataracts had developed following vitrectomy. Data collection and analysis Two review authors independently screened the search results according to the standard methodological procedures expected by Cochrane. Main results We found no RCTs or quasi-RCTs that had compared surgery versus no surgery to remove the lens from eyes of adults in which cataracts had developed following vitrectomy. Authors' conclusions There is no evidence from RCTs or quasi-RCTs on which to base clinical recommendations for surgery for postvitrectomy cataract. There is a clear need for RCTs to address this evidence gap. Such trials should stratify participants by their age, the retinal disorder leading to vitrectomy, and the status of the underlying disease process in the contralateral eye. Outcomes assessed in such trials may include changes (both gains and losses) of visual acuity, quality of life, and adverse events such as posterior capsular rupture and retinal detachment. Both short-term (six-month) and long-term (one- or two-year) outcomes should be examined.

Topic To summarize the relative effects of bevacizumab (Avastin; Genentech, Inc, South San Francisco, CA) and ranibizumab (Lucentis; Genentech, Inc.), using findings from a Cochrane Eyes and Vision Group systematic review. Clinical Relevance Neovascular age-related macular degeneration (NVAMD) is the most common cause of uncorrectable vision loss among the elderly in developed countries. Bevacizumab and ranibizumab are the most frequently used anti–vascular endothelial growth factor (VEGF) agents injected intravitreally to treat NVAMD. Methods For this systematic review, we included only randomized controlled trials in which the 2 anti-VEGF agents had been compared directly. The primary outcome was 1-year gain in best-corrected visual acuity (BCVA) of ≥15 letters. We followed Cochrane methods for trial selection, data extraction, and data analyses. Relative effects of bevacizumab versus ranibizumab are presented as estimated risk ratios (RRs) and mean differences (MDs) with 95% confidence intervals (CIs). Results We identified 6 eligible randomized controlled trials with 2809 participants. The proportion of eyes that gained ≥15 letters of BCVA by 1 year was similar for the 2 agents when the same regimens were compared (RR, 0.90; 95% CI, 0.73–1.11). The mean change in BCVA from baseline also was similar (MD, −0.5 letter; 95% CI, −1.6 to +0.6). Other BCVA and quality of life outcomes were similar for the 2 agents. One-year treatment cost with ranibizumab was 5.1 and 25.5 times the cost for bevacizumab in the 2 largest trials. Ocular adverse events were uncommon ( Conclusions We found no important difference in effectiveness or safety between bevacizumab and ranibizumab for NVAMD treatment, but there was a large cost difference.

Background Previous studies have documented strategies to promote off-label use of drugs using journal publications and other means. Few studies have presented internal company communications that discussed financial reasons for manipulating the scholarly record related to off-label indications. The objective of this study was to build on previous studies to illustrate implementation of a publication strategy by the drug manufacturer for four off-label uses of gabapentin (Neurontin®, Pfizer, Inc.): migraine prophylaxis, treatment of bipolar disorders, neuropathic pain, and nociceptive pain. Methods We included in this study internal company documents, email correspondence, memoranda, study protocols and reports that were made publicly available in 2008 as part of litigation brought by consumers and health insurers against Pfizer for fraudulent sales practices in its marketing of gabapentin (see http://pacer.mad.uscourts.gov/dc/cgi-bin/recentops.pl?filename=saris/pdf/ucl%20opinion.pdf for the Court’s findings). We reviewed documents pertaining to 20 clinical trials, 12 of which were published. We categorized our observations related to reporting biases and linked them with topics covered in internal documents, that is, deciding what should and should not be published and how to spin the study findings (re-framing study results to explain away unfavorable findings or to emphasize favorable findings); and where and when findings should be published and by whom. Results We present extracts from internal company marketing assessments recommending that Pfizer and Parke-Davis (Pfizer acquired Parke-Davis in 2000) adopt a publication strategy to conduct trials and disseminate trial findings for unapproved uses rather than an indication strategy to obtain regulatory approval. We show internal company email correspondence and documents revealing how publication content was influenced and spin was applied; how the company selected where trial findings would be presented or published; how publication of study results was delayed; and the role of ghost authorship. Conclusions Taken together, the extracts we present from internal company documents illustrate implementation of a strategy at odds with unbiased study conduct and dissemination. Our findings suggest that Pfizer and Parke-Davis’s publication strategy had the potential to distort the scientific literature, and thus misinform healthcare decision-makers.