Real-time automatic surgical phase recognition in laparoscopic sigmoidectomy using the convolutional neural network-based deep learning approach. Surg Endosc. 2020;34:4924-4931. [PMID: 31797047 DOI: 10.1007/s00464-019-07281-0]

Artificial intelligence-based automated surgical workflow recognition in esophageal endoscopic submucosal dissection: an international multicenter study (with video)

BACKGROUND

Endoscopic submucosal dissection (ESD) is a crucial yet challenging multi-phase procedure for treating early gastrointestinal cancers. This study developed an artificial intelligence (AI)-based automated surgical workflow recognition model for esophageal ESD and proposed an innovative training program based on esophageal ESD videos with or without AI labels to evaluate its effectiveness for trainees.

METHODS

We retrospectively analyzed complete ESD videos collected from seven hospitals worldwide between 2016 and 2024. The ESD surgical workflow was divided into 6 phases and these videos were divided into five datasets for AI model. Trainees were invited to participate in this multimedia training program and were assigned to the AI or control group randomly. The performance of the AI model and label testing were evaluated using the accuracy.

RESULTS

A total of 195 ESD videos (782,488 s, 9268 phases) were included. The AI model achieved accuracy of 92.08% (95% confidence interval (CI), 91.40-92.76%), 91.71% (95% CI 90.11-93.31%), and 89.84% (95% CI 87.42-92.25%) in the training, internal, and external test dataset (esophagus), respectively. It also achieved acceptable results in the external test dataset (stomach, colorectum). For the training program, the overall label testing accuracy of the AI group learning ESD videos with AI labels was 88.73 ± 2.97%, significantly higher than the control group without AI labels (81.51 ± 4.63%, P < 0.001).

CONCLUSION

The AI model achieved high accuracy in the large ESD video datasets. The training program improves understanding of the complexity of ESD workflow and demonstrates the program's effectiveness for trainees.

Automated Operative Phase and Step Recognition in Vestibular Schwannoma Surgery: Development and Preclinical Evaluation of a Deep Learning Neural Network (IDEAL Stage 0)

BACKGROUND AND OBJECTIVES

Machine learning (ML) in surgical video analysis offers promising prospects for training and decision support in surgery. The past decade has seen key advances in ML-based operative workflow analysis, though existing applications mostly feature shorter surgeries (<2 hours) with limited scene changes. The aim of this study was to develop and evaluate a ML model capable of automated operative workflow recognition for retrosigmoid vestibular schwannoma (VS) resection. In doing so, this project furthers previous research by applying workflow prediction platforms to lengthy (median >5 hours duration), data-heavy surgeries, using VS resection as an exemplar.

METHODS

A video dataset of 21 microscopic retrosigmoid VS resections was collected at a single institution over 3 years and underwent workflow annotation according to a previously agreed expert consensus (Approach, Excision, and Closure phases; and Debulking or Dissection steps within the Excision phase). Annotations were used to train a ML model consisting of a convolutional neural network and a recurrent neural network. 5-fold cross-validation was used, and performance metrics (accuracy, precision, recall, F1 score) were assessed for phase and step prediction.

RESULTS

Median operative video time was 5 hours 18 minutes (IQR 3 hours 21 minutes-6 hours 1 minute). The "Tumor Excision" phase accounted for the majority of each case (median 4 hours 23 minutes), whereas "Approach and Exposure" (28 minutes) and "Closure" (17 minutes) comprised shorter phases. The ML model accurately predicted operative phases (accuracy 81%, weighted F1 0.83) and dichotomized steps (accuracy 86%, weighted F1 0.86).

CONCLUSION

This study demonstrates that our ML model can accurately predict the surgical phases and intraphase steps in retrosigmoid VS resection. This demonstrates the successful application of ML in operative workflow recognition on low-volume, lengthy, data-heavy surgical videos. Despite this, there remains room for improvement in individual step classification. Future applications of ML in low-volume high-complexity operations should prioritize collaborative video sharing to overcome barriers to clinical translation.

Deep learning-based intraoperative visual guidance model for ureter identification in laparoscopic sigmoidectomy

BACKGROUND

Identifying the left ureter is a key step while performing laparoscopic sigmoid resection to prevent intraoperative injury and postoperative complications.

METHODS

This feasibility study aims to evaluate the real-time performance of a deep learning-based computer vision model in identifying the left ureter during laparoscopic sigmoid resection. A deep learning model for ureteral identification was developed using a semantic segmentation algorithm trained from intraoperative images of ureteral dissection in videos depicted from laparoscopic sigmoid resection. We used 86 laparoscopic sigmoid resection recordings performed at King Abdulaziz University Hospital (KAUH), which were further processed with manual annotation. A total of 1237 images were extracted and annotated by three colorectal surgeons. Deep learning You Only Look Once (YOLO) versions 8 and 11 models were applied to the video recording of ureteral identification. Per-frame five-fold cross-validation was used to evaluate model performance.

RESULTS

Experiments showed high results with a mean Average Precision (mAP50) of 0.92 for the Intersection over Union (IoU) threshold greater than or equal to 0.5. The precision, recall, and Dice Coefficient (DC) evaluation metrics are 0.94, 0.88, and 0.90, respectively. The highest DC result is 0.95, achieved through the fourth-fold cross-validation. The stricter IoU threshold between 0.5 and 0.95 is represented by mAP50-95, which is 0.53. The model operated at a speed of 32 Frames Per Second (FPS), indicating it can work in real-time.

CONCLUSION

Deep learning YOLO 8 and 10 for semantic segmentation demonstrates accurate real-time identification of the left ureter in selected videos. A deep learning model could be used to project high-accuracy identification of real-time left ureter during laparoscopic sigmoidectomy using surgeons' expertise in intraoperative image navigation. Limitations included the sample size, lack of diversity in the methods of surgery, incomplete surgical processes, and lack of external validation.

Automated assessment of simulated laparoscopic surgical skill performance using deep learning

Artificial intelligence (AI) has the potential to improve healthcare and patient safety and is currently being adopted across various fields of medicine and healthcare. AI and in particular computer vision (CV) are well suited to the analysis of minimally invasive surgical simulation videos for training and performance improvement. CV techniques have rapidly improved in recent years from accurately recognizing objects, instruments, and gestures to phases of surgery and more recently to remembering past surgical steps. Lack of labeled data is a particular problem in surgery considering its complexity, as human annotation and manual assessment are both expensive in time and cost, and in most cases rely on direct intervention of clinical expertise. In this study, we introduce a newly collected simulated Laparoscopic Surgical Performance Dataset (LSPD) specifically designed to address these challenges. Unlike existing datasets that focus on instrument tracking or anatomical structure recognition, the LSPD is tailored for evaluating simulated laparoscopic surgical skill performance at various expertise levels. We provide detailed statistical analyses to identify and compare poorly performed and well-executed operations across different skill levels (novice, trainee, expert) for three specific skills: stack, bands, and tower. We employ a 3-dimensional convolutional neural network (3DCNN) with a weakly-supervised approach to classify the experience levels of surgeons. Our results show that the 3DCNN effectively distinguishes between novices, trainees, and experts, achieving an F1 score of 0.91 and an AUC of 0.92. This study highlights the value of the LSPD dataset and demonstrates the potential of leveraging 3DCNN-based and weakly-supervised approaches to automate the evaluation of surgical performance, reducing reliance on manual expert annotation and assessments. These advancements contribute to improving surgical training and performance analysis.

Artificial intelligence-assisted phase recognition and skill assessment in laparoscopic surgery: a systematic review

With the widespread adoption of minimally invasive surgery, laparoscopic surgery has been an essential component of modern surgical procedures. As key technologies, laparoscopic phase recognition and skill evaluation aim to identify different stages of the surgical process and assess surgeons' operational skills using automated methods. This, in turn, can improve the quality of surgery and the skill of surgeons. This review summarizes the progress of research in laparoscopic surgery, phase recognition, and skill evaluation. At first, the importance of laparoscopic surgery is introduced, clarifying the relationship between phase recognition, skill evaluation, and other surgical tasks. The publicly available surgical datasets for laparoscopic phase recognition tasks are then detailed. The review highlights the research methods that have exhibited superior performance in these public datasets and identifies common characteristics of these high-performing methods. Based on the insights obtained, the commonly used phase recognition research and surgical skill evaluation methods and models in this field are summarized. In addition, this study briefly outlines the standards and methods for evaluating laparoscopic surgical skills. Finally, an analysis of the difficulties researchers face and potential future development directions is presented. Moreover, this paper aims to provide valuable references for researchers, promoting further advancements in this domain.

Vascular pedicle dissection time in laparoscopic colectomies as a novel marker of surgical skill: a prospective cohort study

BACKGROUND

Outcomes after colorectal resections depend on patient, pathology and operative factors. Existing validated surgical skills scores (such as the competency assessment tool (CAT)) are directly correlated with outcomes but are time-consuming to administer, limiting their clinical utility. The vascular pedicle dissection time (VPDT) is a novel, simple surgical skill assessment measure with the potential for computer vision automation. This study aimed to assess the VPDT and benchmark it against the CAT score.

METHODS

A prospective multicentre study was performed in New Zealand, recording videos of laparoscopic colorectal resections. Patient, operation and histology characteristics were also collected. The VPDT was calculated from retraction of the vascular pedicle to completion of medial dissection, including vascular division. Each laparoscopic video was scored by two independent colorectal surgeons, and the median CAT score was grouped into tertiles.

RESULTS

In total, 154 patients were included between December 2020 and November 2023 (74 (48.1%) right-sided and 80 (51.9%) left-sided resections). Median VPDT was significantly different between the CAT score groups for the right-sided resections (lower, 15 min; middle, 13 min; higher, 10 min; p = 0.036) and the left-sided resections (lower, 46 min; middle, 40 min; higher, 26 min; p =  < 0.001). There was no significant difference in R1 resection, anastomotic leak rate, the occurrence of Clavien-Dindo > 3 complications or re-admission between the CAT groups.

CONCLUSIONS

This study showed that the VPDT was inversely correlated with the CAT score, indicating that it quantifies operative technical skill. A current study is evaluating the suitability of VPDT for real-time measurement using computer vision algorithms. This could allow for automated assessment of surgeons' learning curve and skills.

Systematic review on the use of artificial intelligence to identify anatomical structures during laparoscopic cholecystectomy: a tool towards the future

PURPOSE

Bile duct injury (BDI) during laparoscopic cholecystectomy (LC) is a dreaded complication. Artificial intelligence (AI) has recently been introduced in surgery. This systematic review aims to investigate whether AI can guide surgeons in identifying anatomical structures to facilitate safer dissection during LC.

METHODS

Following PROSPERO registration CRD-42023478754, a Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)-compliant systematic search of MEDLINE (via PubMed), EMBASE, and Web of Science databases was conducted.

RESULTS

Out of 2304 articles identified, twenty-five were included in the analysis. The mean average precision for biliary structures detection reported in the included studies reaches 98%. The mean intersection over union ranges from 0.5 to 0.7, and the mean Dice/F1 spatial correlation index was greater than 0.7/1. AI system provided a change in the annotations in 27% of the cases, and 70% of these shifts were considered safer changes. The contribution to preventing BDI was reported at 3.65/4.

CONCLUSIONS

Although studies on the use of AI during LC are few and very heterogeneous, AI has the potential to identify anatomical structures, thereby guiding surgeons towards safer LC procedures.

Artificial intelligence integration in surgery through hand and instrument tracking: a systematic literature review

Objective

This systematic literature review of the integration of artificial intelligence (AI) applications in surgical practice through hand and instrument tracking provides an overview of recent advancements and analyzes current literature on the intersection of surgery with AI. Distinct AI algorithms and specific applications in surgical practice are also examined.

Methods

An advanced search using medical subject heading terms was conducted in Medline (via PubMed), SCOPUS, and Embase databases for articles published in English. A strict selection process was performed, adhering to PRISMA guidelines.

Results

A total of 225 articles were retrieved. After screening, 77 met inclusion criteria and were included in the review. Use of AI algorithms in surgical practice was uncommon during 2013-2017 but has gained significant popularity since 2018. Deep learning algorithms (n = 62) are increasingly preferred over traditional machine learning algorithms (n = 15). These technologies are used in surgical fields such as general surgery (n = 19), neurosurgery (n = 10), and ophthalmology (n = 9). The most common functional sensors and systems used were prerecorded videos (n = 29), cameras (n = 21), and image datasets (n = 7). The most common applications included laparoscopic (n = 13), robotic-assisted (n = 13), basic (n = 12), and endoscopic (n = 8) surgical skills training, as well as surgical simulation training (n = 8).

Conclusion

AI technologies can be tailored to address distinct needs in surgical education and patient care. The use of AI in hand and instrument tracking improves surgical outcomes by optimizing surgical skills training. It is essential to acknowledge the current technical and social limitations of AI and work toward filling those gaps in future studies.

Deep learning in surgical process modeling: A systematic review of workflow recognition

OBJECTIVE

The application of artificial intelligence (AI) in health care has led to a surge of interest in surgical process modeling (SPM). The objective of this study is to investigate the role of deep learning in recognizing surgical workflows and extracting reliable patterns from datasets used in minimally invasive surgery, thereby advancing the development of context-aware intelligent systems in endoscopic surgeries.

METHODS

We conducted a comprehensive search of articles related to SPM from 2018 to April 2024 in the PubMed, Web of Science, Google Scholar, and IEEE Xplore databases. We chose surgical videos with annotations to describe the article on surgical process modeling and focused on examining the specific methods and research results of each study.

RESULTS

The search initially yielded 2937 articles. After filtering on the basis of the relevance of titles, abstracts, and content, 59 articles were selected for full-text review. These studies highlight the widespread adoption of neural networks, and transformers for surgical workflow analysis (SWA). They focus on minimally invasive surgeries performed with laparoscopes and microscopes. However, the process of surgical annotation lacks detailed description, and there are significant differences in the annotation process for different surgical procedures.

CONCLUSION

Time and spatial sequences are key factors determining the identification of surgical phase. RNN, TCN, and transformer networks are commonly used to extract long-distance temporal relationships. Multimodal data input is beneficial, as it combines information from surgical instruments. However, publicly available datasets often lack clinical knowledge, and establishing large annotated datasets for surgery remains a challenge. To reduce annotation costs, methods such as semi supervised learning, self-supervised learning, contrastive learning, transfer learning, and active learning are commonly used.

Objective surgical skill assessment based on automatic recognition of dissection and exposure times in robot-assisted radical prostatectomy

PURPOSE

Assessing surgical skills is vital for training surgeons, but creating objective, automated evaluation systems is challenging, especially in robotic surgery. Surgical procedures generally involve dissection and exposure (D/E), and their duration and proportion can be used for skill assessment. This study aimed to develop an AI model to acquire D/E parameters in robot-assisted radical prostatectomy (RARP) and verify if these parameters could distinguish between novice and expert surgeons.

METHODS

This retrospective study used 209 RARP videos from 18 Japanese institutions. Dissection time was defined as the duration of forceps energy activation, and exposure time as the combined duration of manipulating the third arm and camera. To measure these times, an AI-based interface recognition model was developed to automatically extract instrument status from the da Vinci Surgical System® UI. We compared novices and experts by measuring dissection and exposure times from the model's output.

RESULTS

The overall accuracies of the UI recognition model for recognizing the forceps type, energy activation status, and camera usage status were 0.991, 0.998, and 0.991, respectively. Dissection time was 45.2 vs. 35.1 s (novice vs. expert, p = 0.374), exposure time was 195.7 vs. 89.7 s (novice vs. expert, p < 0.001), and the D/E ratio was 0.174 vs. 0.315 (novice vs. expert, p = 0.003).

CONCLUSIONS

We successfully developed a model to automatically acquire dissection and exposure parameters for RARP. Exposure time may serve as an objective parameter to distinguish between novices and experts in RARP, and automated technical evaluation in RARP is feasible.

TRIAL REGISTRATION NUMBER AND DATE

This study was approved by the Institutional Review Board of the National Cancer Center Hospital East (No.2020 - 329) on January 28, 2021.

Artificial Intelligence for context-aware surgical guidance in complex robot-assisted oncological procedures: An exploratory feasibility study

INTRODUCTION

Complex oncological procedures pose various surgical challenges including dissection in distinct tissue planes and preservation of vulnerable anatomical structures throughout different surgical phases. In rectal surgery, violation of dissection planes increases the risk of local recurrence and autonomous nerve damage resulting in incontinence and sexual dysfunction. This work explores the feasibility of phase recognition and target structure segmentation in robot-assisted rectal resection (RARR) using machine learning.

MATERIALS AND METHODS

A total of 57 RARR were recorded and subsets of these were annotated with respect to surgical phases and exact locations of target structures (anatomical structures, tissue types, static structures, and dissection areas). For surgical phase recognition, three machine learning models were trained: LSTM, MSTCN, and Trans-SVNet. Based on pixel-wise annotations of target structures in 9037 images, individual segmentation models based on DeepLabv3 were trained. Model performance was evaluated using F1 score, Intersection-over-Union (IoU), accuracy, precision, recall, and specificity.

RESULTS

The best results for phase recognition were achieved with the MSTCN model (F1 score: 0.82 ± 0.01, accuracy: 0.84 ± 0.03). Mean IoUs for target structure segmentation ranged from 0.14 ± 0.22 to 0.80 ± 0.14 for organs and tissue types and from 0.11 ± 0.11 to 0.44 ± 0.30 for dissection areas. Image quality, distorting factors (i.e. blood, smoke), and technical challenges (i.e. lack of depth perception) considerably impacted segmentation performance.

CONCLUSION

Machine learning-based phase recognition and segmentation of selected target structures are feasible in RARR. In the future, such functionalities could be integrated into a context-aware surgical guidance system for rectal surgery.

Automated surgical skill assessment in colorectal surgery using a deep learning-based surgical phase recognition model

BACKGROUND

There is an increasing demand for automated surgical skill assessment to solve issues such as subjectivity and bias that accompany manual assessments. This study aimed to verify the feasibility of assessing surgical skills using a surgical phase recognition model.

METHODS

A deep learning-based model that recognizes five surgical phases of laparoscopic sigmoidectomy was constructed, and its ability to distinguish between three skill-level groups-the expert group, with a high Endoscopic Surgical Skill Qualification System (ESSQS) score (26 videos); the intermediate group, with a low ESSQS score (32 videos); and the novice group, with an experience of < 5 colorectal surgeries (27 videos)-was assessed. Furthermore, 1 272 videos were divided into three groups according to the ESSQS score: ESSQS-high, ESSQS-middle, and ESSQS-low groups, and whether they could be distinguished by the score calculated by multiple regression analysis of the parameters from the model was also evaluated.

RESULTS

The time for mobilization of the colon, time for dissection of the mesorectum plus transection of the rectum plus anastomosis, and phase transition counts were significantly shorter or less in the expert group than in the intermediate (p = 0.0094, 0.0028, and < 0.001, respectively) and novice groups (all p < 0.001). Mesorectal excision time was significantly shorter in the expert group than in the novice group (p = 0.0037). The group with higher ESSQS scores also had higher AI scores.

CONCLUSION

This model has the potential to be applied to automated skill assessments.

Machine learning perioperative applications in visceral surgery: a narrative review

Artificial intelligence in surgery has seen an expansive rise in research and clinical implementation in recent years, with many of the models being driven by machine learning. In the preoperative setting, machine learning models have been utilized to guide indications for surgery, appropriate timing of operations, calculation of risks and prognostication, along with improving estimations of time and resources required for surgeries. Intraoperative applications that have been demonstrated are visual annotations of the surgical field, automated classification of surgical phases and prediction of intraoperative patient decompensation. Postoperative applications have been studied the most, with most efforts put towards prediction of postoperative complications, recurrence patterns of malignancy, enhanced surgical education and assessment of surgical skill. Challenges to implementation of these models in clinical practice include the need for more quantity and quality of standardized data to improve model performance, sufficient resources and infrastructure to train and use machine learning, along with addressing ethical and patient acceptance considerations.

Deep learning-based surgical step recognition for laparoscopic right-sided colectomy

PURPOSE

Understanding the complex anatomy and surgical steps involved in laparoscopic right-sided colectomy (LAP-RC) is essential for standardizing the surgical procedure. Deep-learning (DL)-based computer vision can achieve this. This study aimed to develop a step recognition model for LAP-RC using a dataset of surgical videos with annotated step information and evaluate its recognition performance.

METHODS

This single-center retrospective study utilized a video dataset of laparoscopic ileocecal resection (LAP-ICR) and laparoscopic right-sided hemicolectomy (LAP-RHC) for right-sided colon cancer performed between January 2018 and March 2022. The videos were split into still images, which were divided into training, validation, and test sets using 66%, 17%, and 17% of the data, respectively. Videos were manually classified into eight main steps: 1) medial mobilization, 2) central vascular ligation, 3) dissection of the superior mesenteric vein, 4) retroperitoneal mobilization, 5) lateral mobilization, 6) cranial mobilization, 7) mesocolon resection, and 8) intracorporeal anastomosis. In a simpler version, consecutive surgical steps were combined, resulting in five steps. Precision, recall, F1 scores, and overall accuracy were assessed to evaluate the model's performance in the surgical step classification task.

RESULTS

Seventy-eight patients were included; LAP-ICR and LAP-RHC were performed in 35 (44%) and 44 (56%) patients, respectively. The overall accuracy was 72.1% and 82.9% for the eight-step and combined five-step classification tasks, respectively.

CONCLUSIONS

The automatic surgical step-recognition model for LAP-RCs, developed using a DL algorithm, exhibited a fairly high classification performance. A model that understands the complex steps of LAP-RC will aid the standardization of the surgical procedure.

Real-time segmentation of biliary structure in pure laparoscopic donor hepatectomy

Pure laparoscopic donor hepatectomy (PLDH) has become a standard practice for living donor liver transplantation in expert centers. Accurate understanding of biliary structures is crucial during PLDH to minimize the risk of complications. This study aims to develop a deep learning-based segmentation model for real-time identification of biliary structures, assisting surgeons in determining the optimal transection site during PLDH. A single-institution retrospective feasibility analysis was conducted on 30 intraoperative videos of PLDH. All videos were selected for their use of the indocyanine green near-infrared fluorescence technique to identify biliary structure. From the analysis, 10 representative frames were extracted from each video specifically during the bile duct division phase, resulting in 300 frames. These frames underwent pixel-wise annotation to identify biliary structures and the transection site. A segmentation task was then performed using a DeepLabV3+ algorithm, equipped with a ResNet50 encoder, focusing on the bile duct (BD) and anterior wall (AW) for transection. The model's performance was evaluated using the dice similarity coefficient (DSC). The model predicted biliary structures with a mean DSC of 0.728 ± 0.01 for BD and 0.429 ± 0.06 for AW. Inference was performed at a speed of 15.3 frames per second, demonstrating the feasibility of real-time recognition of anatomical structures during surgery. The deep learning-based semantic segmentation model exhibited promising performance in identifying biliary structures during PLDH. Future studies should focus on validating the clinical utility and generalizability of the model and comparing its efficacy with current gold standard practices to better evaluate its potential clinical applications.

Role of artificial intelligence in gastrointestinal surgery

Artificial intelligence is rapidly evolving and its application is increasing day-by-day in the medical field. The application of artificial intelligence is also valuable in gastrointestinal diseases, by calculating various scoring systems, evaluating radiological images, preoperative and intraoperative assistance, processing pathological slides, prognosticating, and in treatment responses. This field has a promising future and can have an impact on many management algorithms. In this minireview, we aimed to determine the basics of artificial intelligence, the role that artificial intelligence may play in gastrointestinal surgeries and malignancies, and the limitations thereof.

Surgical step recognition in laparoscopic distal gastrectomy using artificial intelligence: a proof-of-concept study

PURPOSE

Laparoscopic distal gastrectomy (LDG) is a difficult procedure for early career surgeons. Artificial intelligence (AI)-based surgical step recognition is crucial for establishing context-aware computer-aided surgery systems. In this study, we aimed to develop an automatic recognition model for LDG using AI and evaluate its performance.

METHODS

Patients who underwent LDG at our institution in 2019 were included in this study. Surgical video data were classified into the following nine steps: (1) Port insertion; (2) Lymphadenectomy on the left side of the greater curvature; (3) Lymphadenectomy on the right side of the greater curvature; (4) Division of the duodenum; (5) Lymphadenectomy of the suprapancreatic area; (6) Lymphadenectomy on the lesser curvature; (7) Division of the stomach; (8) Reconstruction; and (9) From reconstruction to completion of surgery. Two gastric surgeons manually assigned all annotation labels. Convolutional neural network (CNN)-based image classification was further employed to identify surgical steps.

RESULTS

The dataset comprised 40 LDG videos. Over 1,000,000 frames with annotated labels of the LDG steps were used to train the deep-learning model, with 30 and 10 surgical videos for training and validation, respectively. The classification accuracies of the developed models were precision, 0.88; recall, 0.87; F1 score, 0.88; and overall accuracy, 0.89. The inference speed of the proposed model was 32 ps.

CONCLUSION

The developed CNN model automatically recognized the LDG surgical process with relatively high accuracy. Adding more data to this model could provide a fundamental technology that could be used in the development of future surgical instruments.

Real-time detection of active bleeding in laparoscopic colectomy using artificial intelligence

BACKGROUND

Most intraoperative adverse events (iAEs) result from surgeons' errors, and bleeding is the majority of iAEs. Recognizing active bleeding timely is important to ensure safe surgery, and artificial intelligence (AI) has great potential for detecting active bleeding and providing real-time surgical support. This study aimed to develop a real-time AI model to detect active intraoperative bleeding.

METHODS

We extracted 27 surgical videos from a nationwide multi-institutional surgical video database in Japan and divided them at the patient level into three sets: training (n = 21), validation (n = 3), and testing (n = 3). We subsequently extracted the bleeding scenes and labeled distinctively active bleeding and blood pooling frame by frame. We used pre-trained YOLOv7_6w and developed a model to learn both active bleeding and blood pooling. The Average Precision at an Intersection over Union threshold of 0.5 (AP.50) for active bleeding and frames per second (FPS) were quantified. In addition, we conducted two 5-point Likert scales (5 = Excellent, 4 = Good, 3 = Fair, 2 = Poor, and 1 = Fail) questionnaires about sensitivity (the sensitivity score) and number of overdetection areas (the overdetection score) to investigate the surgeons' assessment.

RESULTS

We annotated 34,117 images of 254 bleeding events. The AP.50 for active bleeding in the developed model was 0.574 and the FPS was 48.5. Twenty surgeons answered two questionnaires, indicating a sensitivity score of 4.92 and an overdetection score of 4.62 for the model.

CONCLUSIONS

We developed an AI model to detect active bleeding, achieving real-time processing speed. Our AI model can be used to provide real-time surgical support.

Weakly Supervised Pose Estimation of Surgical Instrument from a Single Endoscopic Image

Instrument pose estimation is a key demand in computer-aided surgery, and its main challenges lie in two aspects: Firstly, the difficulty of obtaining stable corresponding image feature points due to the instruments' high refraction and complicated background, and secondly, the lack of labeled pose data. This study aims to tackle the pose estimation problem of surgical instruments in the current endoscope system using a single endoscopic image. More specifically, a weakly supervised method based on the instrument's image segmentation contour is proposed, with the effective assistance of synthesized endoscopic images. Our method consists of the following three modules: a segmentation module to automatically detect the instrument in the input image, followed by a point inference module to predict the image locations of the implicit feature points of the instrument, and a point back-propagatable Perspective-n-Point module to estimate the pose from the tentative 2D-3D corresponding points. To alleviate the over-reliance on point correspondence accuracy, the local errors of feature point matching and the global inconsistency of the corresponding contours are simultaneously minimized. Our proposed method is validated with both real and synthetic images in comparison with the current state-of-the-art methods.

The development of a deep learning model for automated segmentation of the robotic pancreaticojejunostomy

BACKGROUND

Minimally invasive surgery provides an unprecedented opportunity to review video for assessing surgical performance. Surgical video analysis is time-consuming and expensive. Deep learning provides an alternative for analysis. Robotic pancreaticoduodenectomy (RPD) is a complex and morbid operation. Surgeon technical performance of pancreaticojejunostomy (PJ) has been associated with postoperative pancreatic fistula. In this work, we aimed to utilize deep learning to automatically segment PJ RPD videos.

METHODS

This was a retrospective review of prospectively collected videos from 2011 to 2022 that were in libraries at tertiary referral centers, including 111 PJ videos. Each frame of a robotic PJ video was categorized based on 6 tasks. A 3D convolutional neural network was trained for frame-level visual feature extraction and classification. All the videos were manually annotated for the start and end of each task.

RESULTS

Of the 100 videos assessed, 60 videos were used for the training the model, 10 for hyperparameter optimization, and 30 for the testing of performance. All the frames were extracted (6 frames/second) and annotated. The accuracy and mean per-class F1 scores were 88.01% and 85.34% for tasks.

CONCLUSION

The deep learning model performed well for automated segmentation of PJ videos. Future work will focus on skills assessment and outcome prediction.

An Automated Video Analysis System for Retrospective Assessment and Real-Time Monitoring of Endoscopic Procedures (with Video)

BACKGROUND AND AIMS

Accurate recognition of endoscopic instruments facilitates quantitative evaluation and quality control of endoscopic procedures. However, no relevant research has been reported. In this study, we aimed to develop a computer-assisted system, EndoAdd, for automated endoscopic surgical video analysis based on our dataset of endoscopic instrument images.

METHODS

Large training and validation datasets containing 45,143 images of 10 different endoscopic instruments and a test dataset of 18,375 images collected from several medical centers were used in this research. Annotated image frames were used to train the state-of-the-art object detection model, YOLO-v5, to identify the instruments. Based on the frame-level prediction results, we further developed a hidden Markov model to perform video analysis and generate heatmaps to summarize the videos.

RESULTS

EndoAdd achieved high accuracy (>97%) on the test dataset for all 10 endoscopic instrument types. The mean average accuracy, precision, recall, and F1-score were 99.1%, 92.0%, 88.8%, and 89.3%, respectively. The area under the curve values exceeded 0.94 for all instrument types. Heatmaps of endoscopic procedures were generated for both retrospective and real-time analyses.

CONCLUSIONS

We successfully developed an automated endoscopic video analysis system, EndoAdd, which supports retrospective assessment and real-time monitoring. It can be used for data analysis and quality control of endoscopic procedures in clinical practice.

Assessment of Automated Identification of Phases in Videos of Total Hip Arthroplasty Using Deep Learning Techniques

Background

As the population ages, the rates of hip diseases and fragility fractures are increasing, making total hip arthroplasty (THA) one of the best methods for treating elderly patients. With the increasing number of THA surgeries and diverse surgical methods, there is a need for standard evaluation protocols. This study aimed to use deep learning algorithms to classify THA videos and evaluate the accuracy of the labelling of these videos.

Methods

In our study, we manually annotated 7 phases in THA, including skin incision, broaching, exposure of acetabulum, acetabular reaming, acetabular cup positioning, femoral stem insertion, and skin closure. Within each phase, a second trained annotator marked the beginning and end of instrument usages, such as the skin blade, forceps, Bovie, suction device, suture material, retractor, rasp, femoral stem, acetabular reamer, head trial, and real head.

Results

In our study, we utilized YOLOv3 to collect 540 operating images of THA procedures and create a scene annotation model. The results of our study showed relatively high accuracy in the clear classification of surgical techniques such as skin incision and closure, broaching, acetabular reaming, and femoral stem insertion, with a mean average precision (mAP) of 0.75 or higher. Most of the equipment showed good accuracy of mAP 0.7 or higher, except for the suction device, suture material, and retractor.

Conclusions

Scene annotation for the instrument and phases in THA using deep learning techniques may provide potentially useful tools for subsequent documentation, assessment of skills, and feedback.

Development of deep learning framework for anatomical landmark detection and guided dissection line during laparoscopic cholecystectomy

Background

Bile duct injuries during laparoscopic cholecystectomy can arise from misinterpretation of biliary anatomy, leading to dissection in improper areas. The integration of a deep learning framework into laparoscopic procedures offers the potential for real-time anatomical landmark recognition, ensuring accurate dissection. The objective of this study is to develop a deep learning framework that can precisely identify anatomical landmarks, including Rouviere's sulcus and the liver base of segment IV, and provide a guided dissection line during laparoscopic cholecystectomy.

Methods

We retrospectively collected 40 laparoscopic cholecystectomy videos and extracted 80 images form each video to establish the dataset. Three surgeons annotated the bounding boxes of anatomical landmarks on a total of 3200 images. The YOLOv7 model was trained to detect Rouviere's sulcus and the liver base of segment IV as anatomical landmarks. Additionally, the guided dissection line was generated between these two landmarks by the proposed algorithm. To evaluate the performance of the detection model, mean average precision (mAP), precision, and recall were calculated. Furthermore, the accuracy of the guided dissection line was evaluated by three surgeons. The performance of the detection model was compared to the scaled-YOLOv4 and YOLOv5 models. Finally, the proposed framework was deployed in the operating room for real-time detection and visualization.

Results

The overall performance of the YOLOv7 model on validation set and testing set were 98.1 % and 91.3 %, respectively. Surgeons accepted the visualization of guide dissection line with a rate of 95.71 %. In the operating room, the well-trained model accurately identified the anatomical landmarks and generated the guided dissection line in real-time.

Conclusions

The proposed framework effectively identifies anatomical landmarks and generates a guided dissection line in real-time during laparoscopic cholecystectomy. This research underscores the potential of using deep learning models as computer-assisted tools in surgery, providing an assistant tool to accommodate with surgeons.

Automatic surgical phase recognition-based skill assessment in laparoscopic distal gastrectomy using multicenter videos

BACKGROUND

Gastric surgery involves numerous surgical phases; however, its steps can be clearly defined. Deep learning-based surgical phase recognition can promote stylization of gastric surgery with applications in automatic surgical skill assessment. This study aimed to develop a deep learning-based surgical phase-recognition model using multicenter videos of laparoscopic distal gastrectomy, and examine the feasibility of automatic surgical skill assessment using the developed model.

METHODS

Surgical videos from 20 hospitals were used. Laparoscopic distal gastrectomy was defined and annotated into nine phases and a deep learning-based image classification model was developed for phase recognition. We examined whether the developed model's output, including the number of frames in each phase and the adequacy of the surgical field development during the phase of supra-pancreatic lymphadenectomy, correlated with the manually assigned skill assessment score.

RESULTS

The overall accuracy of phase recognition was 88.8%. Regarding surgical skill assessment based on the number of frames during the phases of lymphadenectomy of the left greater curvature and reconstruction, the number of frames in the high-score group were significantly less than those in the low-score group (829 vs. 1,152, P < 0.01; 1,208 vs. 1,586, P = 0.01, respectively). The output score of the adequacy of the surgical field development, which is the developed model's output, was significantly higher in the high-score group than that in the low-score group (0.975 vs. 0.970, P = 0.04).

CONCLUSION

The developed model had high accuracy in phase-recognition tasks and has the potential for application in automatic surgical skill assessment systems.

Automated segmentation of phases, steps, and tasks in laparoscopic cholecystectomy using deep learning

BACKGROUND

Video-based review is paramount for operative performance assessment but can be laborious when performed manually. Hierarchical Task Analysis (HTA) is a well-known method that divides any procedure into phases, steps, and tasks. HTA requires large datasets of videos with consistent definitions at each level. Our aim was to develop an AI model for automated segmentation of phases, steps, and tasks for laparoscopic cholecystectomy videos using a standardized HTA.

METHODS

A total of 160 laparoscopic cholecystectomy videos were collected from a publicly available dataset known as cholec80 and from our own institution. All videos were annotated for the beginning and ending of a predefined set of phases, steps, and tasks. Deep learning models were then separately developed and trained for the three levels using a 3D Convolutional Neural Network architecture.

RESULTS

Four phases, eight steps, and nineteen tasks were defined through expert consensus. The training set for our deep learning models contained 100 videos with an additional 20 videos for hyperparameter optimization and tuning. The remaining 40 videos were used for testing the performance. The overall accuracy for phases, steps, and tasks were 0.90, 0.81, and 0.65 with the average F1 score of 0.86, 0.76 and 0.48 respectively. Control of bleeding and bile spillage tasks were most variable in definition, operative management, and clinical relevance.

CONCLUSION

The use of hierarchical task analysis for surgical video analysis has numerous applications in AI-based automated systems. Our results show that our tiered method of task analysis can successfully be used to train a DL model.

Deep Learning in Surgical Workflow Analysis: A Review of Phase and Step Recognition

OBJECTIVE

In the last two decades, there has been a growing interest in exploring surgical procedures with statistical models to analyze operations at different semantic levels. This information is necessary for developing context-aware intelligent systems, which can assist the physicians during operations, evaluate procedures afterward or help the management team to effectively utilize the operating room. The objective is to extract reliable patterns from surgical data for the robust estimation of surgical activities performed during operations. The purpose of this article is to review the state-of-the-art deep learning methods that have been published after 2018 for analyzing surgical workflows, with a focus on phase and step recognition.

METHODS

Three databases, IEEE Xplore, Scopus, and PubMed were searched, and additional studies are added through a manual search. After the database search, 343 studies were screened and a total of 44 studies are selected for this review.

CONCLUSION

The use of temporal information is essential for identifying the next surgical action. Contemporary methods used mainly RNNs, hierarchical CNNs, and Transformers to preserve long-distance temporal relations. The lack of large publicly available datasets for various procedures is a great challenge for the development of new and robust models. As supervised learning strategies are used to show proof-of-concept, self-supervised, semi-supervised, or active learning methods are used to mitigate dependency on annotated data.

SIGNIFICANCE

The present study provides a comprehensive review of recent methods in surgical workflow analysis, summarizes commonly used architectures, datasets, and discusses challenges.

Development of an artificial intelligence system for real-time intraoperative assessment of the Critical View of Safety in laparoscopic cholecystectomy

BACKGROUND

The Critical View of Safety (CVS) was proposed in 1995 to prevent bile duct injury during laparoscopic cholecystectomy (LC). The achievement of CVS was evaluated subjectively. This study aimed to develop an artificial intelligence (AI) system to evaluate CVS scores in LC.

MATERIALS AND METHODS

AI software was developed to evaluate the achievement of CVS using an algorithm for image classification based on a deep convolutional neural network. Short clips of hepatocystic triangle dissection were converted from 72 LC videos, and 23,793 images were labeled for training data. The learning models were examined using metrics commonly used in machine learning.

RESULTS

The mean values of precision, recall, F-measure, specificity, and overall accuracy for all the criteria of the best model were 0.971, 0.737, 0.832, 0.966, and 0.834, respectively. It took approximately 6 fps to obtain scores for a single image.

CONCLUSIONS

Using the AI system, we successfully evaluated the achievement of the CVS criteria using still images and videos of hepatocystic triangle dissection in LC. This encourages surgeons to be aware of CVS and is expected to improve surgical safety.

Analysis of Specimen Mammography with Artificial Intelligence to Predict Margin Status

BACKGROUND

Intraoperative specimen mammography is a valuable tool in breast cancer surgery, providing immediate assessment of margins for a resected tumor. However, the accuracy of specimen mammography in detecting microscopic margin positivity is low. We sought to develop an artificial intelligence model to predict the pathologic margin status of resected breast tumors using specimen mammography.

METHODS

A dataset of specimen mammography images matched with pathologic margin status was collected from our institution from 2017 to 2020. The dataset was randomly split into training, validation, and test sets. Specimen mammography models pretrained on radiologic images were developed and compared with models pretrained on nonmedical images. Model performance was assessed using sensitivity, specificity, and area under the receiver operating characteristic curve (AUROC).

RESULTS

The dataset included 821 images, and 53% had positive margins. For three out of four model architectures tested, models pretrained on radiologic images outperformed nonmedical models. The highest performing model, InceptionV3, showed sensitivity of 84%, specificity of 42%, and AUROC of 0.71. Model performance was better among patients with invasive cancers, less dense breasts, and non-white race.

CONCLUSIONS

This study developed and internally validated artificial intelligence models that predict pathologic margins status for partial mastectomy from specimen mammograms. The models' accuracy compares favorably with published literature on surgeon and radiologist interpretation of specimen mammography. With further development, these models could more precisely guide the extent of resection, potentially improving cosmesis and reducing reoperations.

Bringing Artificial Intelligence to the operating room: edge computing for real-time surgical phase recognition

BACKGROUND

Automation of surgical phase recognition is a key effort toward the development of Computer Vision (CV) algorithms, for workflow optimization and video-based assessment. CV is a form of Artificial Intelligence (AI) that allows interpretation of images through a deep learning (DL)-based algorithm. The improvements in Graphic Processing Unit (GPU) computing devices allow researchers to apply these algorithms for recognition of content in videos in real-time. Edge computing, where data is collected, analyzed, and acted upon in close proximity to the collection source, is essential meet the demands of workflow optimization by providing real-time algorithm application. We implemented a real-time phase recognition workflow and demonstrated its performance on 10 Robotic Inguinal Hernia Repairs (RIHR) to obtain phase predictions during the procedure.

METHODS

Our phase recognition algorithm was developed with 211 videos of RIHR originally annotated into 14 surgical phases. Using these videos, a DL model with a ResNet-50 backbone was trained and validated to automatically recognize surgical phases. The model was deployed to a GPU, the Nvidia® Jetson Xavier™ NX edge computing device.

RESULTS

This model was tested on 10 inguinal hernia repairs from four surgeons in real-time. The model was improved using post-recording processing methods such as phase merging into seven final phases (peritoneal scoring, mesh placement, preperitoneal dissection, reduction of hernia, out of body, peritoneal closure, and transitionary idle) and averaging of frames. Predictions were made once per second with a processing latency of approximately 250 ms. The accuracy of the real-time predictions ranged from 59.8 to 78.2% with an average accuracy of 68.7%.

CONCLUSION

A real-time phase prediction of RIHR using a CV deep learning model was successfully implemented. This real-time CV phase segmentation system can be useful for monitoring surgical progress and be integrated into software to provide hospital workflow optimization.

Preliminary stage in the development of an artificial intelligence algorithm: Variations between 100 surgeons in phase annotation in a video of internal fixation of distal radius fracture

INTRODUCTION

In order to be used naturally and widely, an artificial intelligence algorithm of phase detection in surgical videos presupposes an expert consensus defining phases.

OBJECTIVES

The aim of the present study was to seek consensus in defining the various phases of a surgical technique in wrist traumatology.

METHODS

Three thousand two hundred and twenty-nine surgeons were sent a video showing anterior plate fixation of the distal radius and a questionnaire on the number of phases they distinguished and the visual cues signaling the beginning of each phase. Three experimenters predefined the number of phases (5: installation, approach, fixation, verification, closure) and sub-phases (3a: introduction of plate; 3b: positioning distal screws; 3c: positioning proximal screws) and the cues signaling the beginning of each. The numbers of the responses per item were collected.

RESULTS

Only 216 (6.7%) surgeons opened the questionnaire, and 100 answered all questions (3.1%). Most respondents claimed 5/5 expertise. Number of phases identified ranged between 3 and 10. More than two-thirds of respondents identified the same phase cue as defined by the 3 experimenters in most cases, except for "verification" and "positioning proximal screws".

DISCUSSION

Surgical procedures comprise a succession of phases, the beginning or end of which can be defined by a precise visual cue on video, either beginning with the appearance of the cue or the disappearance of the cue defining the preceding phase.

CONCLUSION

These cues need to be defined very precisely before attempting manual annotation of surgical videos in order to develop an artificial intelligence algorithm.

LEVEL OF EVIDENCE

II.

Process model analysis of parenchyma sparing laparoscopic liver surgery to recognize surgical steps and predict impact of new technologies

BACKGROUND

Surgical process model (SPM) analysis is a great means to predict the surgical steps in a procedure as well as to predict the potential impact of new technologies. Especially in complicated and high-volume treatments, such as parenchyma sparing laparoscopic liver resection (LLR), profound process knowledge is essential for enabling improving surgical quality and efficiency.

METHODS

Videos of thirteen parenchyma sparing LLR were analyzed to extract the duration and sequence of surgical steps according to the process model. The videos were categorized into three groups, based on the tumor locations. Next, a detailed discrete events simulation model (DESM) of LLR was built, based on the process model and the process data obtained from the endoscopic videos. Furthermore, the impact of using a navigation platform on the total duration of the LLR was studied with the simulation model by assessing three different scenarios: (i) no navigation platform, (ii) conservative positive effect, and (iii) optimistic positive effect.

RESULTS

The possible variations of sequences of surgical steps in performing parenchyma sparing depending on the tumor locations were established. The statistically most probable chain of surgical steps was predicted, which could be used to improve parenchyma sparing surgeries. In all three categories (i-iii) the treatment phase covered the major part (~ 40%) of the total procedure duration (bottleneck). The simulation results predict that a navigation platform could decrease the total surgery duration by up to 30%.

CONCLUSION

This study showed a DESM based on the analysis of steps during surgical procedures can be used to predict the impact of new technology. SPMs can be used to detect, e.g., the most probable workflow paths which enables predicting next surgical steps, improving surgical training systems, and analyzing surgical performance. Moreover, it provides insight into the points for improvement and bottlenecks in the surgical process.

Automatic Surgical Skill Assessment System Based on Concordance of Standardized Surgical Field Development Using Artificial Intelligence

Importance

Automatic surgical skill assessment with artificial intelligence (AI) is more objective than manual video review-based skill assessment and can reduce human burden. Standardization of surgical field development is an important aspect of this skill assessment.

Objective

To develop a deep learning model that can recognize the standardized surgical fields in laparoscopic sigmoid colon resection and to evaluate the feasibility of automatic surgical skill assessment based on the concordance of the standardized surgical field development using the proposed deep learning model.

Design, Setting, and Participants

This retrospective diagnostic study used intraoperative videos of laparoscopic colorectal surgery submitted to the Japan Society for Endoscopic Surgery between August 2016 and November 2017. Data were analyzed from April 2020 to September 2022.

Interventions

Videos of surgery performed by expert surgeons with Endoscopic Surgical Skill Qualification System (ESSQS) scores higher than 75 were used to construct a deep learning model able to recognize a standardized surgical field and output its similarity to standardized surgical field development as an AI confidence score (AICS). Other videos were extracted as the validation set.

Main Outcomes and Measures

Videos with scores less than or greater than 2 SDs from the mean were defined as the low- and high-score groups, respectively. The correlation between AICS and ESSQS score and the screening performance using AICS for low- and high-score groups were analyzed.

Results

The sample included 650 intraoperative videos, 60 of which were used for model construction and 60 for validation. The Spearman rank correlation coefficient between the AICS and ESSQS score was 0.81. The receiver operating characteristic (ROC) curves for the screening of the low- and high-score groups were plotted, and the areas under the ROC curve for the low- and high-score group screening were 0.93 and 0.94, respectively.

Conclusions and Relevance

The AICS from the developed model strongly correlated with the ESSQS score, demonstrating the model's feasibility for use as a method of automatic surgical skill assessment. The findings also suggest the feasibility of the proposed model for creating an automated screening system for surgical skills and its potential application to other types of endoscopic procedures.

Artificial intelligence in colorectal surgery: an AI-powered systematic review

Artificial intelligence (AI) has the potential to revolutionize surgery in the coming years. Still, it is essential to clarify what the meaningful current applications are and what can be reasonably expected. This AI-powered review assessed the role of AI in colorectal surgery. A Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)-compliant systematic search of PubMed, Embase, Scopus, Cochrane Library databases, and gray literature was conducted on all available articles on AI in colorectal surgery (from January 1 1997 to March 1 2021), aiming to define the perioperative applications of AI. Potentially eligible studies were identified using novel software powered by natural language processing (NLP) and machine learning (ML) technologies dedicated to systematic reviews. Out of 1238 articles identified, 115 were included in the final analysis. Available articles addressed the role of AI in several areas of interest. In the preoperative phase, AI can be used to define tailored treatment algorithms, support clinical decision-making, assess the risk of complications, and predict surgical outcomes and survival. Intraoperatively, AI-enhanced surgery and integration of AI in robotic platforms have been suggested. After surgery, AI can be implemented in the Enhanced Recovery after Surgery (ERAS) pathway. Additional areas of applications included the assessment of patient-reported outcomes, automated pathology assessment, and research. Available data on these aspects are limited, and AI in colorectal surgery is still in its infancy. However, the rapid evolution of technologies makes it likely that it will increasingly be incorporated into everyday practice.

Towards an Action Recognition Framework for Endovascular Surgery

Objective knowledge about instrument manoeuvres in endovascular surgery is essential for evaluating surgical skills and developing advanced technologies for cathlab routines. To the recent day, endovascular navigation has been exclusively assessed in laboratory scenarios. By contrast, information contained in available fluoroscopy data from clinical cases has been disregarded. In this work, we pioneer a learning-based framework for motion activity recognition in fluoroscopy sequences. The architecture is composed of two networks for instrument segmentation and action recognition. In this preliminary study, we demonstrate feasibility of recognising instrument manoeuvres automatically in our ex vivo datasets.Clinical relevance-The proposed framework contributes to image-based and automated assessment of endovascular tasks. This facilitates robotic control development, surgical education, and smart clinical documentation.

Artificial Intelligence in Surgery and Its Potential for Gastric Cancer

Artificial intelligence (AI) has made significant progress in recent years, and many medical fields are attempting to introduce AI technology into clinical practice. Currently, much research is being conducted to evaluate that AI can be incorporated into surgical procedures to make them safer and more efficient, subsequently to obtain better outcomes for patients. In this paper, we review basic AI research regarding surgery and discuss the potential for implementing AI technology in gastric cancer surgery. At present, research and development is focused on AI technologies that assist the surgeon's understandings and judgment during surgery, such as anatomical navigation. AI systems are also being developed to recognize in which the surgical phase is ongoing. Such a surgical phase recognition systems is considered for effective storage of surgical videos and education, in the future, for use in systems to objectively evaluate the skill of surgeons. At this time, it is not considered practical to let AI make intraoperative decisions or move forceps automatically from an ethical standpoint, too. At present, AI research on surgery has various limitations, and it is desirable to develop practical systems that will truly benefit clinical practice in the future.

Evolution of a surgical system using deep learning in minimally invasive surgery (Review)

Recently, artificial intelligence (AI) has been applied in various fields due to the development of new learning methods, such as deep learning, and the marked progress in computational processing speed. AI is also being applied in the medical field for medical image recognition and omics analysis of genomes and other data. Recently, AI applications for videos of minimally invasive surgeries have also advanced, and studies on such applications are increasing. In the present review, studies that focused on the following topics were selected: i) Organ and anatomy identification, ii) instrument identification, iii) procedure and surgical phase recognition, iv) surgery-time prediction, v) identification of an appropriate incision line, and vi) surgical education. The development of autonomous surgical robots is also progressing, with the Smart Tissue Autonomous Robot (STAR) and RAVEN systems being the most reported developments. STAR, in particular, is currently being used in laparoscopic imaging to recognize the surgical site from laparoscopic images and is in the process of establishing an automated suturing system, albeit in animal experiments. The present review examined the possibility of fully autonomous surgical robots in the future.

Surgical Phase Recognition in Inguinal Hernia Repair-AI-Based Confirmatory Baseline and Exploration of Competitive Models

Video-recorded robotic-assisted surgeries allow the use of automated computer vision and artificial intelligence/deep learning methods for quality assessment and workflow analysis in surgical phase recognition. We considered a dataset of 209 videos of robotic-assisted laparoscopic inguinal hernia repair (RALIHR) collected from 8 surgeons, defined rigorous ground-truth annotation rules, then pre-processed and annotated the videos. We deployed seven deep learning models to establish the baseline accuracy for surgical phase recognition and explored four advanced architectures. For rapid execution of the studies, we initially engaged three dozen MS-level engineering students in a competitive classroom setting, followed by focused research. We unified the data processing pipeline in a confirmatory study, and explored a number of scenarios which differ in how the DL networks were trained and evaluated. For the scenario with 21 validation videos of all surgeons, the Video Swin Transformer model achieved ~0.85 validation accuracy, and the Perceiver IO model achieved ~0.84. Our studies affirm the necessity of close collaborative research between medical experts and engineers for developing automated surgical phase recognition models deployable in clinical settings.

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.

An Automated Skill Assessment Framework Based on Visual Motion Signals and a Deep Neural Network in Robot-Assisted Minimally Invasive Surgery

Surgical skill assessment can quantify the quality of the surgical operation via the motion state of the surgical instrument tip (SIT), which is considered one of the effective primary means by which to improve the accuracy of surgical operation. Traditional methods have displayed promising results in skill assessment. However, this success is predicated on the SIT sensors, making these approaches impractical when employing the minimally invasive surgical robot with such a tiny end size. To address the assessment issue regarding the operation quality of robot-assisted minimally invasive surgery (RAMIS), this paper proposes a new automatic framework for assessing surgical skills based on visual motion tracking and deep learning. The new method innovatively combines vision and kinematics. The kernel correlation filter (KCF) is introduced in order to obtain the key motion signals of the SIT and classify them by using the residual neural network (ResNet), realizing automated skill assessment in RAMIS. To verify its effectiveness and accuracy, the proposed method is applied to the public minimally invasive surgical robot dataset, the JIGSAWS. The results show that the method based on visual motion tracking technology and a deep neural network model can effectively and accurately assess the skill of robot-assisted surgery in near real-time. In a fairly short computational processing time of 3 to 5 s, the average accuracy of the assessment method is 92.04% and 84.80% in distinguishing two and three skill levels. This study makes an important contribution to the safe and high-quality development of RAMIS.

SurgSmart: an artificial intelligent system for quality control in laparoscopic cholecystectomy: an observational study

BACKGROUND

The rate of bile duct injury in laparoscopic cholecystectomy (LC) continues to be high due to low critical view of safety (CVS) achievement and the absence of an effective quality control system. The development of an intelligent system enables the automatic quality control of LC surgery and, eventually, the mitigation of bile duct injury. This study aims to develop an intelligent surgical quality control system for LC and using the system to evaluate LC videos and investigate factors associated with CVS achievement.

MATERIALS AND METHODS

SurgSmart, an intelligent system capable of recognizing surgical phases, disease severity, critical division action, and CVS automatically, was developed using training datasets. SurgSmart was also applied in another multicenter dataset to validate its application and investigate factors associated with CVS achievement.

RESULTS

SurgSmart performed well in all models, with the critical division action model achieving the highest overall accuracy (98.49%), followed by the disease severity model (95.45%) and surgical phases model (88.61%). CVSI, CVSII, and CVSIII had an accuracy of 80.64, 97.62, and 78.87%, respectively. CVS was achieved in 4.33% in the system application dataset. In addition, the analysis indicated that surgeons at a higher hospital level had a higher CVS achievement rate. However, there was still considerable variation in CVS achievement among surgeons in the same hospital.

CONCLUSIONS

SurgSmart, the surgical quality control system, performed admirably in our study. In addition, the system's initial application demonstrated its broad potential for use in surgical quality control.

Application of artificial intelligence in diagnosis and treatment of colorectal cancer: A novel Prospect

In the past few decades, according to the rapid development of information technology, artificial intelligence (AI) has also made significant progress in the medical field. Colorectal cancer (CRC) is the third most diagnosed cancer worldwide, and its incidence and mortality rates are increasing yearly, especially in developing countries. This article reviews the latest progress in AI in diagnosing and treating CRC based on a systematic collection of previous literature. Most CRCs transform from polyp mutations. The computer-aided detection systems can significantly improve the polyp and adenoma detection rate by early colonoscopy screening, thereby lowering the possibility of mutating into CRC. Machine learning and bioinformatics analysis can help screen and identify more CRC biomarkers to provide the basis for non-invasive screening. The Convolutional neural networks can assist in reading histopathologic tissue images, reducing the experience difference among doctors. Various studies have shown that AI-based high-level auxiliary diagnostic systems can significantly improve the readability of medical images and help clinicians make more accurate diagnostic and therapeutic decisions. Moreover, Robotic surgery systems such as da Vinci have been more and more commonly used to treat CRC patients, according to their precise operating performance. The application of AI in neoadjuvant chemoradiotherapy has further improved the treatment and efficacy evaluation of CRC. In addition, AI represented by deep learning in gene sequencing research offers a new treatment option. All of these things have seen that AI has a promising prospect in the era of precision medicine.

Computer Vision Analysis of Specimen Mammography to Predict Margin Status

Intra-operative specimen mammography is a valuable tool in breast cancer surgery, providing immediate assessment of margins for a resected tumor. However, the accuracy of specimen mammography in detecting microscopic margin positivity is low. We sought to develop a deep learning-based model to predict the pathologic margin status of resected breast tumors using specimen mammography. A dataset of specimen mammography images matched with pathology reports describing margin status was collected. Models pre-trained on radiologic images were developed and compared with models pre-trained on non-medical images. Model performance was assessed using sensitivity, specificity, and area under the receiver operating characteristic curve (AUROC). The dataset included 821 images and 53% had positive margins. For three out of four model architectures tested, models pre-trained on radiologic images outperformed domain-agnostic models. The highest performing model, InceptionV3, showed a sensitivity of 84%, a specificity of 42%, and AUROC of 0.71. These results compare favorably with the published literature on surgeon and radiologist interpretation of specimen mammography. With further development, these models could assist clinicians with identifying positive margins intra-operatively and decrease the rate of positive margins and re-operation in breast-conserving surgery.

Capturing fine-grained details for video-based automation of suturing skills assessment

OBJECTIVES

Manually-collected suturing technical skill scores are strong predictors of continence recovery after robotic radical prostatectomy. Herein, we automate suturing technical skill scoring through computer vision (CV) methods as a scalable method to provide feedback.

METHODS

Twenty-two surgeons completed a suturing exercise three times on the Mimic™ Flex VR simulator. Instrument kinematic data (XYZ coordinates of each instrument and pose) were captured at 30 Hz. After standardized training, three human raters manually video segmented suturing task into four sub-stitch phases (Needle handling, Needle targeting, Needle driving, Needle withdrawal) and labeled the corresponding technical skill domains (Needle positioning, Needle entry, Needle driving, and Needle withdrawal). The CV framework extracted RGB features and optical flow frames using a pre-trained AlexNet. Additional CV strategies including auxiliary supervision (using kinematic data during training only) and attention mechanisms were implemented to improve performance.

RESULTS

This study included data from 15 expert surgeons (median caseload 300 [IQR 165-750]) and 7 training surgeons (0 [IQR 0-8]). In all, 226 virtual sutures were captured. Automated assessments for Needle positioning performed best with the simplest approach (1 s video; AUC 0.749). Remaining skill domains exhibited improvements with the implementation of auxiliary supervision and attention mechanisms when deployed separately (AUC 0.604-0.794). All techniques combined produced the best performance, particularly for Needle driving and Needle withdrawal (AUC 0.959 and 0.879, respectively).

CONCLUSIONS

This study demonstrated the best performance of automated suturing technical skills assessment to date using advanced CV techniques. Future work will determine if a "human in the loop" is necessary to verify surgeon evaluations.

Evolution of the digital operating room: the place of video technology in surgery

PURPOSE

The aim of this review was to collate current evidence wherein digitalisation, through the incorporation of video technology and artificial intelligence (AI), is being applied to the practice of surgery. Applications are vast, and the literature investigating the utility of surgical video and its synergy with AI has steadily increased over the last 2 decades. This type of technology is widespread in other industries, such as autonomy in transportation and manufacturing.

METHODS

Articles were identified primarily using the PubMed and MEDLINE databases. The MeSH terms used were "surgical education", "surgical video", "video labelling", "surgery", "surgical workflow", "telementoring", "telemedicine", "machine learning", "deep learning" and "operating room". Given the breadth of the subject and the scarcity of high-level data in certain areas, a narrative synthesis was selected over a meta-analysis or systematic review to allow for a focussed discussion of the topic.

RESULTS

Three main themes were identified and analysed throughout this review, (1) the multifaceted utility of surgical video recording, (2) teleconferencing/telemedicine and (3) artificial intelligence in the operating room.

CONCLUSIONS

Evidence suggests the routine collection of intraoperative data will be beneficial in the advancement of surgery, by driving standardised, evidence-based surgical care and personalised training of future surgeons. However, many barriers stand in the way of widespread implementation, necessitating close collaboration between surgeons, data scientists, medicolegal personnel and hospital policy makers.

Scene-dependent, feedforward eye gaze metrics can differentiate technical skill levels of trainees in laparoscopic surgery

INTRODUCTION

In laparoscopic surgery, looking in the target areas is an indicator of proficiency. However, gaze behaviors revealing feedforward control (i.e., looking ahead) and their importance have been under-investigated in surgery. This study aims to establish the sensitivity and relative importance of different scene-dependent gaze and motion metrics for estimating trainee proficiency levels in surgical skills.

METHODS

Medical students performed the Fundamentals of Laparoscopic Surgery peg transfer task while recording their gaze on the monitor and tool activities inside the trainer box. Using computer vision and fixation algorithms, five scene-dependent gaze metrics and one tool speed metric were computed for 499 practice trials. Cluster analysis on the six metrics was used to group the trials into different clusters/proficiency levels, and ANOVAs were conducted to test differences between proficiency levels. A Random Forest model was trained to study metric importance at predicting proficiency levels.

RESULTS

Three clusters were identified, corresponding to three proficiency levels. The correspondence between the clusters and proficiency levels was confirmed by differences between completion times (F2,488 = 38.94, p < .001). Further, ANOVAs revealed significant differences between the three levels for all six metrics. The Random Forest model predicted proficiency level with 99% out-of-bag accuracy and revealed that scene-dependent gaze metrics reflecting feedforward behaviors were more important for prediction than the ones reflecting feedback behaviors.

CONCLUSION

Scene-dependent gaze metrics revealed skill levels of trainees more precisely than between experts and novices as suggested in the literature. Further, feedforward gaze metrics appeared to be more important than feedback ones at predicting proficiency.

Intraoperative Detection of Surgical Gauze Using Deep Convolutional Neural Network

During laparoscopic surgery, surgical gauze is usually inserted into the body cavity to help achieve hemostasis. Retention of surgical gauze in the body cavity may necessitate reoperation and increase surgical risk. Using deep learning technology, this study aimed to propose a neural network model for gauze detection from the surgical video to record the presence of the gauze. The model was trained by the training group using YOLO (You Only Look Once)v5x6, then applied to the testing group. Positive predicted value (PPV), sensitivity, and mean average precision (mAP) were calculated. Furthermore, a timeline of gauze presence in the video was drawn by the model as well as human annotation to evaluate the accuracy. After the model was well-trained, the PPV, sensitivity, and mAP in the testing group were 0.920, 0.828, and 0.881, respectively. The inference time was 11.3 ms per image. The average accuracy of the model adding a marking and filtering process was 0.899. In conclusion, surgical gauze can be successfully detected using deep learning in the surgical video. Our model provided a fast detection of surgical gauze, allowing further real-time gauze tracing in laparoscopic surgery that may help surgeons recall the location of the missing gauze.

Role of artificial intelligence in risk prediction, prognostication, and therapy response assessment in colorectal cancer: current state and future directions

Artificial Intelligence (AI) is a branch of computer science that utilizes optimization, probabilistic and statistical approaches to analyze and make predictions based on a vast amount of data. In recent years, AI has revolutionized the field of oncology and spearheaded novel approaches in the management of various cancers, including colorectal cancer (CRC). Notably, the applications of AI to diagnose, prognosticate, and predict response to therapy in CRC, is gaining traction and proving to be promising. There have also been several advancements in AI technologies to help predict metastases in CRC and in Computer-Aided Detection (CAD) Systems to improve miss rates for colorectal neoplasia. This article provides a comprehensive review of the role of AI in predicting risk, prognosis, and response to therapies among patients with CRC.

Deep Learning Applications in Surgery: Current Uses and Future Directions

Deep learning (DL) is a subset of machine learning that is rapidly gaining traction in surgical fields. Its tremendous capacity for powerful data-driven problem-solving has generated computational breakthroughs in many realms, with the fields of medicine and surgery becoming increasingly prominent avenues. Through its multi-layer architecture of interconnected neural networks, DL enables feature extraction and pattern recognition of highly complex and large-volume data. Across various surgical specialties, DL is being applied to optimize both preoperative planning and intraoperative performance in new and innovative ways. Surgeons are now able to integrate deep learning tools into their practice to improve patient safety and outcomes. Through this review, we explore the applications of deep learning in surgery and related subspecialties with an aim to shed light on the practical utilization of this technology in the present and near future.

Video labelling robot-assisted radical prostatectomy and the role of artificial intelligence (AI): training a novice

Video labelling is the assigning of meaningful information to raw videos. With the evolution of artificial intelligence and its intended incorporation into the operating room, video datasets can be invaluable tools for education and the training of intelligent surgical workflow systems through computer vision. However, the process of manual labelling of video datasets can prove costly and time-consuming for already busy practising surgeons. Twenty-five robot-assisted radical prostatectomy (RARP) procedures were recorded on Proximie, an augmented reality platform, anonymised and access given to a novice, who was trained to develop the knowledge and skills needed to accurately segment a full-length RARP procedure on a video labelling platform. A labelled video was subsequently randomly selected for assessment of accuracy by four practising urologists. Of the 25 videos allocated, 17 were deemed suitable for labelling, and 8 were excluded on the basis of procedure length and video quality. The labelled video selected for assessment was graded for accuracy of temporal labelling, with an average score of 93.1%, and a range of 85.6–100%. The self-training of a novice in the accurate segmentation of a surgical video to the standard of a practising urologist is feasible and practical for the RARP procedure. The assigning of temporal labels on a video labelling platform was also studied and proved feasible throughout the study period.