Surgical data science - from concepts toward clinical translation

Using Graduating Surgical Resident Milestone Ratings to Predict Patient Outcomes: A Blunt Instrument for a Complex Problem

In 2013, U.S. general surgery residency programs implemented a milestones assessment framework in an effort to incorporate more competency-focused evaluation methods. Developed by a group of surgical education leaders and other stakeholders working with the Accreditation Council for Graduate Medical Education and recently updated in a version 2.0, the surgery milestones framework is centered around 6 "core competencies": patient care, medical knowledge, practice-based learning and improvement, interpersonal and communication skills, professionalism, and systems-based practice. While prior work has focused on the validity of milestones as a measure of resident performance, associations between general surgery resident milestone ratings and their post-training patient outcomes have only recently been explored in an analysis in this issue of Academic Medicine by Kendrick et al. Despite their well-designed efforts to tackle this complex problem, no relationships were identified. This accompanying commentary discusses the broader implications for the use of milestone ratings beyond their intended application, alternative assessment methods, and the challenges of developing predictive assessments in the complex setting of surgical care. Although milestone ratings have not been shown to provide the specificity needed to predict clinical outcomes in the complex settings studied by Kendrick et al, hope remains that utilization of other outcomes, assessment frameworks, and data analytic tools could augment these models and further our progress toward a predictive assessment in surgical education. Evaluation of residents in general surgery residency programs has grown both more sophisticated and complicated in the setting of increasing patient and case complexity, constraints on time, and regulation of resident supervision in the operating room. Over the last decade, surgical education research efforts related to resident assessment have focused on measuring performance through accurate and reproducible methods with evidence for their validity, as well as on attempting to refine decision making about resident preparedness for unsupervised practice.

Classifying Vocal Folds Fixation from Endoscopic Videos with Machine Learning

Vocal folds motility evaluation is paramount in both the assessment of functional deficits and in the accurate staging of neoplastic disease of the glottis. Diagnostic endoscopy, and in particular videoendoscopy, is nowadays the method through which the motility is estimated. The clinical diagnosis, however, relies on the examination of the videoendoscopic frames, which is a subjective and professional-dependent task. Hence, a more rigorous, objective, reliable, and repeatable method is needed. To support clinicians, this paper proposes a machine learning (ML) approach for vocal cords motility classification. From the endoscopic videos of 186 patients with both vocal cords preserved motility and fixation, a dataset of 558 images relative to the two classes was extracted. Successively, a number of features was retrieved from the images and used to train and test four well-grounded ML classifiers. From test results, the best performance was achieved using XGBoost, with precision = 0.82, recall = 0.82, F1 score = 0.82, and accuracy = 0.82. After comparing the most relevant ML models, we believe that this approach could provide precise and reliable support to clinical evaluation.Clinical Relevance- This research represents an important advancement in the state-of-the-art of computer-assisted otolaryngology, to develop an effective tool for motility assessment in the clinical practice.

Acquisition and usage of robotic surgical data for machine learning analysis

BACKGROUND

The increasing use of robot-assisted surgery (RAS) has led to the need for new methods of assessing whether new surgeons are qualified to perform RAS, without the resource-demanding process of having expert surgeons do the assessment. Computer-based automation and artificial intelligence (AI) are seen as promising alternatives to expert-based surgical assessment. However, no standard protocols or methods for preparing data and implementing AI are available for clinicians. This may be among the reasons for the impediment to the use of AI in the clinical setting.

METHOD

We tested our method on porcine models with both the da Vinci Si and the da Vinci Xi. We sought to capture raw video data from the surgical robots and 3D movement data from the surgeons and prepared the data for the use in AI by a structured guide to acquire and prepare video data using the following steps: 'Capturing image data from the surgical robot', 'Extracting event data', 'Capturing movement data of the surgeon', 'Annotation of image data'.

RESULTS

15 participant (11 novices and 4 experienced) performed 10 different intraabdominal RAS procedures. Using this method we captured 188 videos (94 from the surgical robot, and 94 corresponding movement videos of the surgeons' arms and hands). Event data, movement data, and labels were extracted from the raw material and prepared for use in AI.

CONCLUSION

With our described methods, we could collect, prepare, and annotate images, events, and motion data from surgical robotic systems in preparation for its use in AI.

Development and Validation of an Electronic Health Record-Based Risk Assessment Tool for Hypoglycemia in Patients With Type 2 Diabetes Mellitus

BACKGROUND

The recent availability of high-quality data from clinical trials, together with machine learning (ML) techniques, presents exciting opportunities for developing prediction models for clinical outcomes.

METHODS

As a proof-of-concept, we translated a hypoglycemia risk model derived from the Action to Control Cardiovascular Risk in Diabetes (ACCORD) study into the HypoHazardScore, a risk assessment tool applicable to electronic health record (EHR) data. To assess its performance, we conducted a 16-week clinical study at the University of Minnesota where participants (N = 40) with type 2 diabetes mellitus (T2DM) had hypoglycemia assessed prospectively by continuous glucose monitoring (CGM).

RESULTS

The HypoHazardScore combines 16 risk factors commonly found within the EHR. The HypoHazardScore successfully predicted (area under the curve [AUC] = 0.723) whether participants experienced least one CGM-assessed hypoglycemic event (glucose <54 mg/dL for ≥15 minutes from two CGMs) while significantly correlating with frequency of CGM-assessed hypoglycemic events (r = 0.38) and percent time experiencing CGM-assessed hypoglycemia (r = 0.39). Compared to participants with a low HypoHazardScore (N = 19, score <4, median score of 4), participants with a high HypoHazardScore (N = 21, score ≥4) had more frequent CGM-assessed hypoglycemic events (high: 1.6 ± 2.2 events/week; low: 0.3 ± 0.5 events/week) and experienced more CGM-assessed hypoglycemia (high: 1.4% ± 2.0%; low: 0.2% ± 0.4% time) during the 16-week follow-up.

CONCLUSIONS

We demonstrated that a hypoglycemia risk model can be successfully adapted from the ACCORD data to the EHR, with validation by a prospective study using CGM-assessed hypoglycemia. The HypoHazardScore represents a significant advancement toward implementing an EHR-based decision support system that can help reduce hypoglycemia in patients with T2DM.

Dissecting self-supervised learning methods for surgical computer vision

The field of surgical computer vision has undergone considerable breakthroughs in recent years with the rising popularity of deep neural network-based methods. However, standard fully-supervised approaches for training such models require vast amounts of annotated data, imposing a prohibitively high cost; especially in the clinical domain. Self-Supervised Learning (SSL) methods, which have begun to gain traction in the general computer vision community, represent a potential solution to these annotation costs, allowing to learn useful representations from only unlabeled data. Still, the effectiveness of SSL methods in more complex and impactful domains, such as medicine and surgery, remains limited and unexplored. In this work, we address this critical need by investigating four state-of-the-art SSL methods (MoCo v2, SimCLR, DINO, SwAV) in the context of surgical computer vision. We present an extensive analysis of the performance of these methods on the Cholec80 dataset for two fundamental and popular tasks in surgical context understanding, phase recognition and tool presence detection. We examine their parameterization, then their behavior with respect to training data quantities in semi-supervised settings. Correct transfer of these methods to surgery, as described and conducted in this work, leads to substantial performance gains over generic uses of SSL - up to 7.4% on phase recognition and 20% on tool presence detection - as well as state-of-the-art semi-supervised phase recognition approaches by up to 14%. Further results obtained on a highly diverse selection of surgical datasets exhibit strong generalization properties. The code is available at https://github.com/CAMMA-public/SelfSupSurg.

A systematic review of annotation for surgical process model analysis in minimally invasive surgery based on video

BACKGROUND

Annotated data are foundational to applications of supervised machine learning. However, there seems to be a lack of common language used in the field of surgical data science. The aim of this study is to review the process of annotation and semantics used in the creation of SPM for minimally invasive surgery videos.

METHODS

For this systematic review, we reviewed articles indexed in the MEDLINE database from January 2000 until March 2022. We selected articles using surgical video annotations to describe a surgical process model in the field of minimally invasive surgery. We excluded studies focusing on instrument detection or recognition of anatomical areas only. The risk of bias was evaluated with the Newcastle Ottawa Quality assessment tool. Data from the studies were visually presented in table using the SPIDER tool.

RESULTS

Of the 2806 articles identified, 34 were selected for review. Twenty-two were in the field of digestive surgery, six in ophthalmologic surgery only, one in neurosurgery, three in gynecologic surgery, and two in mixed fields. Thirty-one studies (88.2%) were dedicated to phase, step, or action recognition and mainly relied on a very simple formalization (29, 85.2%). Clinical information in the datasets was lacking for studies using available public datasets. The process of annotation for surgical process model was lacking and poorly described, and description of the surgical procedures was highly variable between studies.

CONCLUSION

Surgical video annotation lacks a rigorous and reproducible framework. This leads to difficulties in sharing videos between institutions and hospitals because of the different languages used. There is a need to develop and use common ontology to improve libraries of annotated surgical videos.

Human visual explanations mitigate bias in AI-based assessment of surgeon skills

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.

Labelling instructions matter in biomedical image analysis

Biomedical image analysis algorithm validation depends on high-quality annotation of reference datasets, for which labelling instructions are key. Despite their importance, their optimization remains largely unexplored. Here we present a systematic study of labelling instructions and their impact on annotation quality in the field. Through comprehensive examination of professional practice and international competitions registered at the Medical Image Computing and Computer Assisted Intervention Society, the largest international society in the biomedical imaging field, we uncovered a discrepancy between annotators’ needs for labelling instructions and their current quality and availability. On the basis of an analysis of 14,040 images annotated by 156 annotators from four professional annotation companies and 708 Amazon Mechanical Turk crowdworkers using instructions with different information density levels, we further found that including exemplary images substantially boosts annotation performance compared with text-only descriptions, while solely extending text descriptions does not. Finally, professional annotators constantly outperform Amazon Mechanical Turk crowdworkers. Our study raises awareness for the need of quality standards in biomedical image analysis labelling instructions.

Artificial intelligence and automation in endoscopy and surgery

Modern endoscopy relies on digital technology, from high-resolution imaging sensors and displays to electronics connecting configurable illumination and actuation systems for robotic articulation. In addition to enabling more effective diagnostic and therapeutic interventions, the digitization of the procedural toolset enables video data capture of the internal human anatomy at unprecedented levels. Interventional video data encapsulate functional and structural information about a patient's anatomy as well as events, activity and action logs about the surgical process. This detailed but difficult-to-interpret record from endoscopic procedures can be linked to preoperative and postoperative records or patient imaging information. Rapid advances in artificial intelligence, especially in supervised deep learning, can utilize data from endoscopic procedures to develop systems for assisting procedures leading to computer-assisted interventions that can enable better navigation during procedures, automation of image interpretation and robotically assisted tool manipulation. In this Perspective, we summarize state-of-the-art artificial intelligence for computer-assisted interventions in gastroenterology and surgery.

Comparative validation of machine learning algorithms for surgical workflow and skill analysis with the HeiChole benchmark

PURPOSE

Surgical workflow and skill analysis are key technologies for the next generation of cognitive surgical assistance systems. These systems could increase the safety of the operation through context-sensitive warnings and semi-autonomous robotic assistance or improve training of surgeons via data-driven feedback. In surgical workflow analysis up to 91% average precision has been reported for phase recognition on an open data single-center video dataset. In this work we investigated the generalizability of phase recognition algorithms in a multicenter setting including more difficult recognition tasks such as surgical action and surgical skill.

METHODS

To achieve this goal, a dataset with 33 laparoscopic cholecystectomy videos from three surgical centers with a total operation time of 22 h was created. Labels included framewise annotation of seven surgical phases with 250 phase transitions, 5514 occurences of four surgical actions, 6980 occurences of 21 surgical instruments from seven instrument categories and 495 skill classifications in five skill dimensions. The dataset was used in the 2019 international Endoscopic Vision challenge, sub-challenge for surgical workflow and skill analysis. Here, 12 research teams trained and submitted their machine learning algorithms for recognition of phase, action, instrument and/or skill assessment.

RESULTS

F1-scores were achieved for phase recognition between 23.9% and 67.7% (n = 9 teams), for instrument presence detection between 38.5% and 63.8% (n = 8 teams), but for action recognition only between 21.8% and 23.3% (n = 5 teams). The average absolute error for skill assessment was 0.78 (n = 1 team).

CONCLUSION

Surgical workflow and skill analysis are promising technologies to support the surgical team, but there is still room for improvement, as shown by our comparison of machine learning algorithms. This novel HeiChole benchmark can be used for comparable evaluation and validation of future work. In future studies, it is of utmost importance to create more open, high-quality datasets in order to allow the development of artificial intelligence and cognitive robotics in surgery.

Laparoscopic Video Analysis Using Temporal, Attention, and Multi-Feature Fusion Based-Approaches

Adapting intelligent context-aware systems (CAS) to future operating rooms (OR) aims to improve situational awareness and provide surgical decision support systems to medical teams. CAS analyzes data streams from available devices during surgery and communicates real-time knowledge to clinicians. Indeed, recent advances in computer vision and machine learning, particularly deep learning, paved the way for extensive research to develop CAS. In this work, a deep learning approach for analyzing laparoscopic videos for surgical phase recognition, tool classification, and weakly-supervised tool localization in laparoscopic videos was proposed. The ResNet-50 convolutional neural network (CNN) architecture was adapted by adding attention modules and fusing features from multiple stages to generate better-focused, generalized, and well-representative features. Then, a multi-map convolutional layer followed by tool-wise and spatial pooling operations was utilized to perform tool localization and generate tool presence confidences. Finally, the long short-term memory (LSTM) network was employed to model temporal information and perform tool classification and phase recognition. The proposed approach was evaluated on the Cholec80 dataset. The experimental results (i.e., 88.5% and 89.0% mean precision and recall for phase recognition, respectively, 95.6% mean average precision for tool presence detection, and a 70.1% F1-score for tool localization) demonstrated the ability of the model to learn discriminative features for all tasks. The performances revealed the importance of integrating attention modules and multi-stage feature fusion for more robust and precise detection of surgical phases and tools.

Investigating the attentional focus to workplace-related soundscapes in a complex audio-visual-motor task using EEG

Introduction

In demanding work situations (e.g., during a surgery), the processing of complex soundscapes varies over time and can be a burden for medical personnel. Here we study, using mobile electroencephalography (EEG), how humans process workplace-related soundscapes while performing a complex audio-visual-motor task (3D Tetris). Specifically, we wanted to know how the attentional focus changes the processing of the soundscape as a whole.

Method

Participants played a game of 3D Tetris in which they had to use both hands to control falling blocks. At the same time, participants listened to a complex soundscape, similar to what is found in an operating room (i.e., the sound of machinery, people talking in the background, alarm sounds, and instructions). In this within-subject design, participants had to react to instructions (e.g., "place the next block in the upper left corner") and to sounds depending on the experimental condition, either to a specific alarm sound originating from a fixed location or to a beep sound that originated from varying locations. Attention to the alarm reflected a narrow attentional focus, as it was easy to detect and most of the soundscape could be ignored. Attention to the beep reflected a wide attentional focus, as it required the participants to monitor multiple different sound streams.

Results and discussion

Results show the robustness of the N1 and P3 event related potential response during this dynamic task with a complex auditory soundscape. Furthermore, we used temporal response functions to study auditory processing to the whole soundscape. This work is a step toward studying workplace-related sound processing in the operating room using mobile EEG.

Analysing multi-perspective patient-related data during laparoscopic gynaecology procedures

Fusing data from different medical perspectives inside the operating room (OR) sets the stage for developing intelligent context-aware systems. These systems aim to promote better awareness inside the OR by keeping every medical team well informed about the work of other teams and thus mitigate conflicts resulting from different targets. In this research, a descriptive analysis of data collected from anaesthesiology and surgery was performed to investigate the relationships between the intra-abdominal pressure (IAP) and lung mechanics for patients during laparoscopic procedures. Data of nineteen patients who underwent laparoscopic gynaecology were included. Statistical analysis of all subjects showed a strong relationship between the IAP and dynamic lung compliance (r = 0.91). Additionally, the peak airway pressure was also strongly correlated to the IAP in volume-controlled ventilated patients (r = 0.928). Statistical results obtained by this study demonstrate the importance of analysing the relationship between surgical actions and physiological responses. Moreover, these results form the basis for developing medical decision support models, e.g., automatic compensation of IAP effects on lung function.

The Dresden Surgical Anatomy Dataset for Abdominal Organ Segmentation in Surgical Data Science

Laparoscopy is an imaging technique that enables minimally-invasive procedures in various medical disciplines including abdominal surgery, gynaecology and urology. To date, publicly available laparoscopic image datasets are mostly limited to general classifications of data, semantic segmentations of surgical instruments and low-volume weak annotations of specific abdominal organs. The Dresden Surgical Anatomy Dataset provides semantic segmentations of eight abdominal organs (colon, liver, pancreas, small intestine, spleen, stomach, ureter, vesicular glands), the abdominal wall and two vessel structures (inferior mesenteric artery, intestinal veins) in laparoscopic view. In total, this dataset comprises 13195 laparoscopic images. For each anatomical structure, we provide over a thousand images with pixel-wise segmentations. Annotations comprise semantic segmentations of single organs and one multi-organ-segmentation dataset including segments for all eleven anatomical structures. Moreover, we provide weak annotations of organ presence for every single image. This dataset markedly expands the horizon for surgical data science applications of computer vision in laparoscopic surgery and could thereby contribute to a reduction of risks and faster translation of Artificial Intelligence into surgical practice.

Automatic generation of operation notes in endoscopic pituitary surgery videos using workflow recognition

Operation notes are a crucial component of patient care. However, writing them manually is prone to human error, particularly in high pressured clinical environments. Automatic generation of operation notes from video recordings can alleviate some of the administrative burdens, improve accuracy, and provide additional information. To achieve this for endoscopic pituitary surgery, 27-steps were identified via expert consensus. Then, for the 97-videos recorded for this study, a timestamp of each step was annotated by an expert surgeon. To automatically determine whether a step is present in a video, a three-stage architecture was created. Firstly, for each step, a convolution neural network was used for binary image classification on each frame of a video. Secondly, for each step, the binary frame classifications were passed to a discriminator for binary video classification. Thirdly, for each video, the binary video classifications were passed to an accumulator for multi-label step classification. The architecture was trained on 77-videos, and tested on 20-videos, where a 0.80 weighted-F1 score was achieved. The classifications were inputted into a clinically based predefined template, and further enriched with additional video analytics. This work therefore demonstrates automatic generation of operative notes from surgical videos is feasible, and can assist surgeons during documentation.

Developing artificial intelligence models for medical student suturing and knot-tying video-based assessment and coaching

BACKGROUND

Early introduction and distributed learning have been shown to improve student comfort with basic requisite suturing skills. The need for more frequent and directed feedback, however, remains an enduring concern for both remote and in-person training. A previous in-person curriculum for our second-year medical students transitioning to clerkships was adapted to an at-home video-based assessment model due to the social distancing implications of COVID-19. We aimed to develop an Artificial Intelligence (AI) model to perform video-based assessment.

METHODS

Second-year medical students were asked to submit a video of a simple interrupted knot on a penrose drain with instrument tying technique after self-training to proficiency. Proficiency was defined as performing the task under two minutes with no critical errors. All the videos were first manually rated with a pass-fail rating and then subsequently underwent task segmentation. We developed and trained two AI models based on convolutional neural networks to identify errors (instrument holding and knot-tying) and provide automated ratings.

RESULTS

A total of 229 medical student videos were reviewed (150 pass, 79 fail). Of those who failed, the critical error distribution was 15 knot-tying, 47 instrument-holding, and 17 multiple. A total of 216 videos were used to train the models after excluding the low-quality videos. A k-fold cross-validation (k = 10) was used. The accuracy of the instrument holding model was 89% with an F-1 score of 74%. For the knot-tying model, the accuracy was 91% with an F-1 score of 54%.

CONCLUSIONS

Medical students require assessment and directed feedback to better acquire surgical skill, but this is often time-consuming and inadequately done. AI techniques can instead be employed to perform automated surgical video analysis. Future work will optimize the current model to identify discrete errors in order to supplement video-based rating with specific feedback.

Domain generalization improves end-to-end object detection for real-time surgical tool detection

PURPOSE

Computer assistance for endoscopic surgery depends on knowledge about the contents in an endoscopic scene. An important step of analysing the video contents is real-time surgical tool detection. Most methods for tool detection nevertheless depend on multi-step algorithms building upon prior knowledge like anchor boxes or non-maximum suppression which ultimately decrease performance. A real-world difficulty encountered by learning-based methods are limited datasets. Training a neural network on data matching a specific distribution (e.g. from a single hospital or showing a specific type of surgery) can result in a lack of generalization.

METHODS

In this paper, we propose the application of a transformer based architecture for end-to-end tool detection. This architecture promises state-of-the-art accuracy while decreasing the complexity resulting in improved run-time performance. To improve the lack of cross-domain generalization due to limited datasets, we enhance the architecture with a latent feature space via variational encoding to capture common intra-domain information. This feature space models the linear dependencies between domains by constraining their rank.

RESULTS

The trained neural networks show a distinct improvement on out-of-domain data indicating better generalization to unseen domains. Inference with the end-to-end architecture can be performed at up to 138 frames per second (FPS) achieving a speedup in comparison to older approaches.

CONCLUSIONS

Experimental results on three representative datasets demonstrate the performance of the method. We also show that our approach leads to better domain generalization.

The role of artificial intelligence in surgical simulation

Artificial Intelligence (AI) plays an integral role in enhancing the quality of surgical simulation, which is increasingly becoming a popular tool for enriching the training experience of a surgeon. This spans the spectrum from facilitating preoperative planning, to intraoperative visualisation and guidance, ultimately with the aim of improving patient safety. Although arguably still in its early stages of widespread clinical application, AI technology enables personal evaluation and provides personalised feedback in surgical training simulations. Several forms of surgical visualisation technologies currently in use for anatomical education and presurgical assessment rely on different AI algorithms. However, while it is promising to see clinical examples and technological reports attesting to the efficacy of AI-supported surgical simulators, barriers to wide-spread commercialisation of such devices and software remain complex and multifactorial. High implementation and production costs, scarcity of reports evidencing the superiority of such technology, and intrinsic technological limitations remain at the forefront. As AI technology is key to driving the future of surgical simulation, this paper will review the literature delineating its current state, challenges, and prospects. In addition, a consolidated list of FDA/CE approved AI-powered medical devices for surgical simulation is presented, in order to shed light on the existing gap between academic achievements and the universal commercialisation of AI-enabled simulators. We call for further clinical assessment of AI-supported surgical simulators to support novel regulatory body approved devices and usher surgery into a new era of surgical education.

Surgomics: personalized prediction of morbidity, mortality and long-term outcome in surgery using machine learning on multimodal data

BACKGROUND

Personalized medicine requires the integration and analysis of vast amounts of patient data to realize individualized care. With Surgomics, we aim to facilitate personalized therapy recommendations in surgery by integration of intraoperative surgical data and their analysis with machine learning methods to leverage the potential of this data in analogy to Radiomics and Genomics.

METHODS

We defined Surgomics as the entirety of surgomic features that are process characteristics of a surgical procedure automatically derived from multimodal intraoperative data to quantify processes in the operating room. In a multidisciplinary team we discussed potential data sources like endoscopic videos, vital sign monitoring, medical devices and instruments and respective surgomic features. Subsequently, an online questionnaire was sent to experts from surgery and (computer) science at multiple centers for rating the features' clinical relevance and technical feasibility.

RESULTS

In total, 52 surgomic features were identified and assigned to eight feature categories. Based on the expert survey (n = 66 participants) the feature category with the highest clinical relevance as rated by surgeons was "surgical skill and quality of performance" for morbidity and mortality (9.0 ± 1.3 on a numerical rating scale from 1 to 10) as well as for long-term (oncological) outcome (8.2 ± 1.8). The feature category with the highest feasibility to be automatically extracted as rated by (computer) scientists was "Instrument" (8.5 ± 1.7). Among the surgomic features ranked as most relevant in their respective category were "intraoperative adverse events", "action performed with instruments", "vital sign monitoring", and "difficulty of surgery".

CONCLUSION

Surgomics is a promising concept for the analysis of intraoperative data. Surgomics may be used together with preoperative features from clinical data and Radiomics to predict postoperative morbidity, mortality and long-term outcome, as well as to provide tailored feedback for surgeons.

Breaking down the silos of artificial intelligence in surgery: glossary of terms

BACKGROUND

The literature on artificial intelligence (AI) in surgery has advanced rapidly during the past few years. However, the published studies on AI are mostly reported by computer scientists using their own jargon which is unfamiliar to surgeons.

METHODS

A literature search was conducted in using PubMed following the preferred reporting items for systematic reviews and meta-analyses (PRISMA) statement. The primary outcome of this review is to provide a glossary with definitions of the commonly used AI terms in surgery to improve their understanding by surgeons.

RESULTS

One hundred ninety-five studies were included in this review, and 38 AI terms related to surgery were retrieved. Convolutional neural networks were the most frequently culled term by the search, accounting for 74 studies on AI in surgery, followed by classification task (n = 62), artificial neural networks (n = 53), and regression (n = 49). Then, the most frequent expressions were supervised learning (reported in 24 articles), support vector machine (SVM) in 21, and logistic regression in 16. The rest of the 38 terms was seldom mentioned.

CONCLUSIONS

The proposed glossary can be used by several stakeholders. First and foremost, by residents and attending consultant surgeons, both having to understand the fundamentals of AI when reading such articles. Secondly, junior researchers at the start of their career in Surgical Data Science and thirdly experts working in the regulatory sections of companies involved in the AI Business Software as a Medical Device (SaMD) preparing documents for submission to the Food and Drug Administration (FDA) or other agencies for approval.

Multicentric validation of EndoDigest: a computer vision platform for video documentation of the critical view of safety in laparoscopic cholecystectomy

BACKGROUND

A computer vision (CV) platform named EndoDigest was recently developed to facilitate the use of surgical videos. Specifically, EndoDigest automatically provides short video clips to effectively document the critical view of safety (CVS) in laparoscopic cholecystectomy (LC). The aim of the present study is to validate EndoDigest on a multicentric dataset of LC videos.

METHODS

LC videos from 4 centers were manually annotated with the time of the cystic duct division and an assessment of CVS criteria. Incomplete recordings, bailout procedures and procedures with an intraoperative cholangiogram were excluded. EndoDigest leveraged predictions of deep learning models for workflow analysis in a rule-based inference system designed to estimate the time of the cystic duct division. Performance was assessed by computing the error in estimating the manually annotated time of the cystic duct division. To provide concise video documentation of CVS, EndoDigest extracted video clips showing the 2 min preceding and the 30 s following the predicted cystic duct division. The relevance of the documentation was evaluated by assessing CVS in automatically extracted 2.5-min-long video clips.

RESULTS

144 of the 174 LC videos from 4 centers were analyzed. EndoDigest located the time of the cystic duct division with a mean error of 124.0 ± 270.6 s despite the use of fluorescent cholangiography in 27 procedures and great variations in surgical workflows across centers. The surgical evaluation found that 108 (75.0%) of the automatically extracted short video clips documented CVS effectively.

CONCLUSIONS

EndoDigest was robust enough to reliably locate the time of the cystic duct division and efficiently video document CVS despite the highly variable workflows. Training specifically on data from each center could improve results; however, this multicentric validation shows the potential for clinical translation of this surgical data science tool to efficiently document surgical safety.

Computer vision in surgery: from potential to clinical value

Hundreds of millions of operations are performed worldwide each year, and the rising uptake in minimally invasive surgery has enabled fiber optic cameras and robots to become both important tools to conduct surgery and sensors from which to capture information about surgery. Computer vision (CV), the application of algorithms to analyze and interpret visual data, has become a critical technology through which to study the intraoperative phase of care with the goals of augmenting surgeons' decision-making processes, supporting safer surgery, and expanding access to surgical care. While much work has been performed on potential use cases, there are currently no CV tools widely used for diagnostic or therapeutic applications in surgery. Using laparoscopic cholecystectomy as an example, we reviewed current CV techniques that have been applied to minimally invasive surgery and their clinical applications. Finally, we discuss the challenges and obstacles that remain to be overcome for broader implementation and adoption of CV in surgery.

Generic surgical process model for minimally invasive liver treatment methods

Surgical process modelling is an innovative approach that aims to simplify the challenges involved in improving surgeries through quantitative analysis of a well-established model of surgical activities. In this paper, surgical process model strategies are applied for the analysis of different Minimally Invasive Liver Treatments (MILTs), including ablation and surgical resection of the liver lesions. Moreover, a generic surgical process model for these differences in MILTs is introduced. The generic surgical process model was established at three different granularity levels. The generic process model, encompassing thirteen phases, was verified against videos of MILT procedures and interviews with surgeons. The established model covers all the surgical and interventional activities and the connections between them and provides a foundation for extensive quantitative analysis and simulations of MILT procedures for improving computer-assisted surgery systems, surgeon training and evaluation, surgeon guidance and planning systems and evaluation of new technologies.

Trans-SVNet: hybrid embedding aggregation Transformer for surgical workflow analysis

PURPOSE

Real-time surgical workflow analysis has been a key component for computer-assisted intervention system to improve cognitive assistance. Most existing methods solely rely on conventional temporal models and encode features with a successive spatial-temporal arrangement. Supportive benefits of intermediate features are partially lost from both visual and temporal aspects. In this paper, we rethink feature encoding to attend and preserve the critical information for accurate workflow recognition and anticipation.

METHODS

We introduce Transformer in surgical workflow analysis, to reconsider complementary effects of spatial and temporal representations. We propose a hybrid embedding aggregation Transformer, named Trans-SVNet, to effectively interact with the designed spatial and temporal embeddings, by employing spatial embedding to query temporal embedding sequence. We jointly optimized by loss objectives from both analysis tasks to leverage their high correlation.

RESULTS

We extensively evaluate our method on three large surgical video datasets. Our method consistently outperforms the state-of-the-arts across three datasets on workflow recognition task. Jointly learning with anticipation, recognition results can gain a large improvement. Our approach also shows its effectiveness on anticipation with promising performance achieved. Our model achieves a real-time inference speed of 0.0134 second per frame.

CONCLUSION

Experimental results demonstrate the efficacy of our hybrid embeddings integration by rediscovering the crucial cues from complementary spatial-temporal embeddings. The better performance by multi-task learning indicates that anticipation task brings the additional knowledge to recognition task. Promising effectiveness and efficiency of our method also show its promising potential to be used in operating room.

Artificial Intelligence in Colorectal Cancer Surgery: Present and Future Perspectives

Artificial intelligence (AI) and computer vision (CV) are beginning to impact medicine. While evidence on the clinical value of AI-based solutions for the screening and staging of colorectal cancer (CRC) is mounting, CV and AI applications to enhance the surgical treatment of CRC are still in their early stage. This manuscript introduces key AI concepts to a surgical audience, illustrates fundamental steps to develop CV for surgical applications, and provides a comprehensive overview on the state-of-the-art of AI applications for the treatment of CRC. Notably, studies show that AI can be trained to automatically recognize surgical phases and actions with high accuracy even in complex colorectal procedures such as transanal total mesorectal excision (TaTME). In addition, AI models were trained to interpret fluorescent signals and recognize correct dissection planes during total mesorectal excision (TME), suggesting CV as a potentially valuable tool for intraoperative decision-making and guidance. Finally, AI could have a role in surgical training, providing automatic surgical skills assessment in the operating room. While promising, these proofs of concept require further development, validation in multi-institutional data, and clinical studies to confirm AI as a valuable tool to enhance CRC treatment.

Robust hand tracking for surgical telestration

PURPOSE

As human failure has been shown to be one primary cause for post-operative death, surgical training is of the utmost socioeconomic importance. In this context, the concept of surgical telestration has been introduced to enable experienced surgeons to efficiently and effectively mentor trainees in an intuitive way. While previous approaches to telestration have concentrated on overlaying drawings on surgical videos, we explore the augmented reality (AR) visualization of surgical hands to imitate the direct interaction with the situs.

METHODS

We present a real-time hand tracking pipeline specifically designed for the application of surgical telestration. It comprises three modules, dedicated to (1) the coarse localization of the expert's hand and the subsequent (2) segmentation of the hand for AR visualization in the field of view of the trainee and (3) regression of keypoints making up the hand's skeleton. The semantic representation is obtained to offer the ability for structured reporting of the motions performed as part of the teaching.

RESULTS

According to a comprehensive validation based on a large data set comprising more than 14,000 annotated images with varying application-relevant conditions, our algorithm enables real-time hand tracking and is sufficiently accurate for the task of surgical telestration. In a retrospective validation study, a mean detection accuracy of 98%, a mean keypoint regression accuracy of 10.0 px and a mean Dice Similarity Coefficient of 0.95 were achieved. In a prospective validation study, it showed uncompromised performance when the sensor, operator or gesture varied.

CONCLUSION

Due to its high accuracy and fast inference time, our neural network-based approach to hand tracking is well suited for an AR approach to surgical telestration. Future work should be directed to evaluating the clinical value of the approach.

A Delphi consensus statement for digital surgery

The use of digital technology is increasing rapidly across surgical specialities, yet there is no consensus for the term ‘digital surgery’. This is critical as digital health technologies present technical, governance, and legal challenges which are unique to the surgeon and surgical patient. We aim to define the term digital surgery and the ethical issues surrounding its clinical application, and to identify barriers and research goals for future practice. 38 international experts, across the fields of surgery, AI, industry, law, ethics and policy, participated in a four-round Delphi exercise. Issues were generated by an expert panel and public panel through a scoping questionnaire around key themes identified from the literature and voted upon in two subsequent questionnaire rounds. Consensus was defined if >70% of the panel deemed the statement important and <30% unimportant. A final online meeting was held to discuss consensus statements. The definition of digital surgery as the use of technology for the enhancement of preoperative planning, surgical performance, therapeutic support, or training, to improve outcomes and reduce harm achieved 100% consensus agreement. We highlight key ethical issues concerning data, privacy, confidentiality and public trust, consent, law, litigation and liability, and commercial partnerships within digital surgery and identify barriers and research goals for future practice. Developers and users of digital surgery must not only have an awareness of the ethical issues surrounding digital applications in healthcare, but also the ethical considerations unique to digital surgery. Future research into these issues must involve all digital surgery stakeholders including patients.

Methods and Measures for Mental Stress Assessment in Surgery: A Systematic Review of 20 Years of Literature

Real-time mental stress monitoring from surgeons and surgical staff in operating rooms may reduce surgical injuries, improve performance and quality of medical care, and accelerate implementation of stress-management strategies. Motivated by the increase in usage of objective and subjective metrics for cognitive monitoring and by the gap in reviews of experimental design setups and data analytics, a systematic review of 71 studies on mental stress and workload measurement in surgical settings, published in 2001-2020, is presented. Almost 61% of selected papers used both objective and subjective measures, followed by 25% that only administered subjective tools - mostly consisting of validated instruments and customized surveys. An overall increase in the total number of publications on intraoperative stress assessment was observed from mid-2010 s along with a momentum in the use of both subjective and real-time objective measures. Cardiac activity, including heart-rate variability metrics, stress hormones, and eye-tracking metrics were the most frequently and electroencephalography (EEG) was the least frequently used objective measures. Around 40% of selected papers collected at least two objective measures, 41% used wearable devices, 23% performed synchronization and annotation, and 76% conducted baseline or multi-point data acquisition. Furthermore, 93% used a variety of statistical techniques, 14% applied regression models, and only one study released a public, anonymized dataset. This review of data modalities, experimental setups, and analysis techniques for intraoperative stress monitoring highlights the initiatives of surgical data science and motivates research on computational techniques for mental and surgical skills assessment and cognition-guided surgery.

Artificial Intelligence Methods and Artificial Intelligence-Enabled Metrics for Surgical Education: A Multidisciplinary Consensus

BACKGROUND

Artificial intelligence (AI) methods and AI-enabled metrics hold tremendous potential to advance surgical education. Our objective was to generate consensus guidance on specific needs for AI methods and AI-enabled metrics for surgical education.

STUDY DESIGN

The study included a systematic literature search, a virtual conference, and a 3-round Delphi survey of 40 representative multidisciplinary stakeholders with domain expertise selected through purposeful sampling. The accelerated Delphi process was completed within 10 days. The survey covered overall utility, anticipated future (10-year time horizon), and applications for surgical training, assessment, and feedback. Consensus was agreement among 80% or more respondents. We coded survey questions into 11 themes and descriptively analyzed the responses.

RESULTS

The respondents included surgeons (40%), engineers (15%), affiliates of industry (27.5%), professional societies (7.5%), regulatory agencies (7.5%), and a lawyer (2.5%). The survey included 155 questions; consensus was achieved on 136 (87.7%). The panel listed 6 deliverables each for AI-enhanced learning curve analytics and surgical skill assessment. For feedback, the panel identified 10 priority deliverables spanning 2-year (n = 2), 5-year (n = 4), and 10-year (n = 4) timeframes. Within 2 years, the panel expects development of methods to recognize anatomy in images of the surgical field and to provide surgeons with performance feedback immediately after an operation. The panel also identified 5 essential that should be included in operative performance reports for surgeons.

CONCLUSIONS

The Delphi panel consensus provides a specific, bold, and forward-looking roadmap for AI methods and AI-enabled metrics for surgical education.

Endoscopic Imaging Technology Today

One of the most applied imaging methods in medicine is endoscopy. A highly specialized image modality has been developed since the first modern endoscope, the "Lichtleiter" of Bozzini was introduced in the early 19th century. Multiple medical disciplines use endoscopy for diagnostics or to visualize and support therapeutic procedures. Therefore, the shapes, functionalities, handling concepts, and the integrated and surrounding technology of endoscopic systems were adapted to meet these dedicated medical application requirements. This survey gives an overview of modern endoscopic technology's state of the art. Therefore, the portfolio of several manufacturers with commercially available products on the market was screened and summarized. Additionally, some trends for upcoming developments were collected.

Data-centric multi-task surgical phase estimation with sparse scene segmentation

Purpose

Surgical workflow estimation techniques aim to divide a surgical video into temporal segments based on predefined surgical actions or objectives, which can be of different granularity such as steps or phases. Potential applications range from real-time intra-operative feedback to automatic post-operative reports and analysis. A common approach in the literature for performing automatic surgical phase estimation is to decouple the problem into two stages: feature extraction from a single frame and temporal feature fusion. This approach is performed in two stages due to computational restrictions when processing large spatio-temporal sequences.

Methods

The majority of existing works focus on pushing the performance solely through temporal model development. Differently, we follow a data-centric approach and propose a training pipeline that enables models to maximise the usage of existing datasets, which are generally used in isolation. Specifically, we use dense phase annotations available in Cholec80, and sparse scene (i.e., instrument and anatomy) segmentation annotation available in CholecSeg8k in less than 5% of the overlapping frames. We propose a simple multi-task encoder that effectively fuses both streams, when available, based on their importance and jointly optimise them for performing accurate phase prediction.

Results and conclusion

We show that with a small fraction of scene segmentation annotations, a relatively simple model can obtain comparable results than previous state-of-the-art and more complex architectures when evaluated in similar settings. We hope that this data-centric approach can encourage new research directions where data, and how to use it, plays an important role along with model development.

Reducing Prediction volatility in the surgical workflow recognition of endoscopic pituitary surgery

Purpose:

Workflow recognition can aid surgeons before an operation when used as a training tool, during an operation by increasing operating room efficiency, and after an operation in the completion of operation notes. Although several methods have been applied to this task, they have been tested on few surgical datasets. Therefore, their generalisability is not well tested, particularly for surgical approaches utilising smaller working spaces which are susceptible to occlusion and necessitate frequent withdrawal of the endoscope. This leads to rapidly changing predictions, which reduces the clinical confidence of the methods, and hence limits their suitability for clinical translation.

Methods:

Firstly, the optimal neural network is found using established methods, using endoscopic pituitary surgery as an exemplar. Then, prediction volatility is formally defined as a new evaluation metric as a proxy for uncertainty, and two temporal smoothing functions are created. The first (modal, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_n$$\end{document}Mn) mode-averages over the previous n predictions, and the second (threshold, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_n$$\end{document}Tn) ensures a class is only changed after being continuously predicted for n predictions. Both functions are independently applied to the predictions of the optimal network.

Results:

The methods are evaluated on a 50-video dataset using fivefold cross-validation, and the optimised evaluation metric is weighted-\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$F_1$$\end{document}F1 score. The optimal model is ResNet-50+LSTM achieving 0.84 in 3-phase classification and 0.74 in 7-step classification. Applying threshold smoothing further improves these results, achieving 0.86 in 3-phase classification, and 0.75 in 7-step classification, while also drastically reducing the prediction volatility.

Conclusion:

The results confirm the established methods generalise to endoscopic pituitary surgery, and show simple temporal smoothing not only reduces prediction volatility, but actively improves performance.

[Oncological surgery in the interdisciplinary context-On the way to personalized medicine]

Oncological surgery is a discipline which closely interacts with other clinical partners and remains in many cases the cornerstone of a curative treatment of solid tumors. Due to the progress in the field of systemic tumor treatment as well as innovations in surgical techniques, the indications in oncological surgery are also changing, such as extended indications for patients with oligometastatic disease. Surgery of metastases has long been established for colorectal cancer and is being further tested for other entities, such as pancreatic and gastric cancer, within randomized controlled clinical trials (e.g. RENAISSANCE and METAPANC). A new challenge is the handling of a clinical complete remission after total neoadjuvant therapy, for example in locally advanced rectal cancer or in esophageal cancer. Here, organ and function preservation are increasingly propagated but should only be performed within clinical trials until stratification enables the identification of patients in whom this concept is oncologically safe. The personalized use of oncological surgery is dependent on the patient, the tumor and on the total multimodal concept.