Current trends and barriers to video management and analytics as a tool for surgeon skilling.

BACKGROUND: The benefits of intraoperative recording are well published in the literature; however, few studies have identified current practices, barriers, and subsequent solutions. The objective of this study was to better understand surgeon's current practices and perceptions of video management and gather blinded feedback on a new surgical video recording product with the potential to address these barriers effectively. METHODS: A structured questionnaire was used to survey 230 surgeons (general, gynecologic, and urologic) and hospital administrators across the US and Europe regarding their current video recording practices. The same questionnaire was used to evaluate a blinded concept describing a new intraoperative recording solution. RESULTS: 54% of respondents reported recording eligible cases, with the majority recording less than 35% of their total eligible caseload. Reasons for not recording included finding no value in recording simple procedures, forgetting to record, lack of access to equipment, legal concerns, labor intensity, and difficulty accessing videos. Among non-recording surgeons, 65% reported considering recording cases to assess surgical techniques, document practice, submit to conferences, share with colleagues, and aid in training. 35% of surgeons rejected recording due to medico-legal concerns, lack of perceived benefit, concerns about secure storage, and price. Regarding the concept of a recording solution, 74% of all respondents were very likely or quite likely to recommend the product for adoption at their facility. Appealing features to current recorders included the product's ease of use, use of AI to maintain patient and staff privacy, lack of manual downloads, availability of full-length procedural videos, and ease of access and storage. Non-recorders found the immediate access to videos and maintenance of patient/staff privacy appealing. CONCLUSION: Tools that address barriers to recording, accessing, and managing surgical case videos are critical for improving surgical skills. Touch Surgery Enterprise is a valuable tool that can help overcome these barriers.

Artificial intelligence automated surgical phases recognition in intraoperative videos of laparoscopic pancreatoduodenectomy.

BACKGROUND: Laparoscopic pancreatoduodenectomy (LPD) is one of the most challenging operations and has a long learning curve. Artificial intelligence (AI) automated surgical phase recognition in intraoperative videos has many potential applications in surgical education, helping shorten the learning curve, but no study has made this breakthrough in LPD. Herein, we aimed to build AI models to recognize the surgical phase in LPD and explore the performance characteristics of AI models. METHODS: Among 69 LPD videos from a single surgical team, we used 42 in the building group to establish the models and used the remaining 27 videos in the analysis group to assess the models' performance characteristics. We annotated 13 surgical phases of LPD, including 4 key phases and 9 necessary phases. Two minimal invasive pancreatic surgeons annotated all the videos. We built two AI models for the key phase and necessary phase recognition, based on convolutional neural networks. The overall performance of the AI models was determined mainly by mean average precision (mAP). RESULTS: Overall mAPs of the AI models in the test set of the building group were 89.7% and 84.7% for key phases and necessary phases, respectively. In the 27-video analysis group, overall mAPs were 86.8% and 71.2%, with maximum mAPs of 98.1% and 93.9%. We found commonalities between the error of model recognition and the differences of surgeon annotation, and the AI model exhibited bad performance in cases with anatomic variation or lesion involvement with adjacent organs. CONCLUSIONS: AI automated surgical phase recognition can be achieved in LPD, with outstanding performance in selective cases. This breakthrough may be the first step toward AI- and video-based surgical education in more complex surgeries.

Virtual pointer for gaze guidance in laparoscopic surgery.

BACKGROUND: A challenge of laparoscopic surgery is learning how to interpret the indirect view of the operative field. Acquiring professional vision-understanding what to see and which information to attend to, is thereby an essential part of laparoscopic training and one in which trainers exert great effort to convey. We designed a virtual pointer (VP) that enables experts to point or draw free-hand sketches over an intraoperative laparoscopic video for a novice to see. This study aimed to investigate the efficacy of the virtual pointer in guiding novices' gaze patterns. METHODS: We conducted a counter-balanced, within-subject trial to compare the novices' gaze behaviors in laparoscopic training with the virtual pointer compared to a standard training condition, i.e., verbal instruction with un-mediated gestures. In the study, seven trainees performed four simulated laparoscopic tasks guided by an experienced surgeon as the trainer. A Tobii Pro X3-120 eye-tracker was used to capture the trainees' eye movements. The measures include fixation rate, i.e., the frequency of trainees' fixations, saccade amplitude, and fixation concentration, i.e., the closeness of trainees' fixations. RESULTS: No significant difference in fixation rate or saccade amplitude was found between the virtual pointer condition and the standard condition. In the virtual pointer condition, trainees' fixations were more concentrated (p = 0.039) and longer fixations were more clustered, compared to the Standard condition (p = 0.008). CONCLUSIONS: The virtual pointer effectively improved surgical trainees' in-the-moment gaze focus during the laparoscopic training by reducing their gaze dispersion and concentrating their attention on the anatomical target. These results suggest that technologies which support gaze training should be expert-driven and intraoperative to efficiently modify novices' gaze behaviors.

Resection of cervical ependymoma.

Intramedullary ependymomas are surgically curable tumors. However, their surgical resection poses several challenges. In this intraoperative video we illustrate the main steps for the surgical resection of a cervical intramedullary ependymoma. These critical steps include: adequate exposure of the entire length of the tumor; use of the intraoperative ultrasound; identification of the posterior median sulcus and separation of the posterior columns; Identification of the plane between the spinal cord and the tumor; mobilization and debulking of the tumor and disconnection of the vascular supply (usually from small anterior spinal artery branches). Following these basic steps a complete resection can be safely achieved in many cases. The video can be found here: http://youtu.be/QMYXC_F4O4U.

Resection of spinal hemangioblastoma.

Spinal cord hemangioblastomas occur as sporadic lesions or in the setting of Von Hippel-Lindau disease. In this intraoperative video we present a case of sporadic cervical cord hemangioblastoma and illustrate the main surgical steps to achieve safe and complete resection which include: identification and division of the feeding arteries; careful circumferential dissection of the tumor from the surrounding gliotic cord; identification, isolation and division of the main venous drainage and single piece removal of the tumor. The video can be found here: http://youtu.be/I7DxqRrfTxc.

Hemodynamic Analysis Shows High Wall Shear Stress Is Associated with Intraoperatively Observed Thin Wall Regions of Intracranial Aneurysms.

BACKGROUND: Studying the relationship between hemodynamics and local intracranial aneurysm (IA) pathobiology can help us understand the natural history of IA. We characterized the relationship between the IA wall appearance, using intraoperative imaging, and the hemodynamics from CFD simulations. METHODS: Three-dimensional geometries of 15 IAs were constructed and used for CFD. Two-dimensional intraoperative images were subjected to wall classification using a machine learning approach, after which the wall type was mapped onto the 3D surface. IA wall regions included thick (white), normal (purple-crimson), and thin/translucent (red) regions. IA-wide and local statistical analyses were performed to assess the relationship between hemodynamics and wall type. RESULTS: Thin regions of the IA sac had significantly higher WSS, Normalized WSS, WSS Divergence and Transverse WSS, compared to both normal and thick regions. Thicker regions tended to co-locate with significantly higher RRT than thin regions. These trends were observed on a local scale as well. Regression analysis showed a significant positive correlation between WSS and thin regions and a significant negative correlation between WSSD and thick regions. CONCLUSION: Hemodynamic simulation results were associated with the intraoperatively observed IA wall type. We consistently found that elevated WSS and WSSNorm were associated with thin regions of the IA wall rather than thick and normal regions.

A Guide to Annotation of Neurosurgical Intraoperative Video for Machine Learning Analysis and Computer Vision.

OBJECTIVE: Computer vision (CV) is a subset of artificial intelligence that performs computations on image or video data, permitting the quantitative analysis of visual information. Common CV tasks that may be relevant to surgeons include image classification, object detection and tracking, and extraction of higher order features. Despite the potential applications of CV to intraoperative video, however, few surgeons describe the use of CV. A primary roadblock in implementing CV is the lack of a clear workflow to create an intraoperative video dataset to which CV can be applied. We report general principles for creating usable surgical video datasets and the result of their applications. METHODS: Video annotations from cadaveric endoscopic endonasal skull base simulations (n = 20 trials of 1-5 minutes, size = 8 GB) were reviewed by 2 researcher-annotators. An internal, retrospective analysis of workflow for development of the intraoperative video annotations was performed to identify guiding practices. RESULTS: Approximately 34,000 frames of surgical video were annotated. Key considerations in developing annotation workflows include 1) overcoming software and personnel constraints; 2) ensuring adequate storage and access infrastructure; 3) optimization and standardization of annotation protocol; and 4) operationalizing annotated data. Potential tools for use include CVAT (Computer Vision Annotation Tool) and Vott: open-sourced annotation software allowing for local video storage, easy setup, and the use of interpolation. CONCLUSIONS: CV techniques can be applied to surgical video, but challenges for novice users may limit adoption. We outline principles in annotation workflow that can mitigate initial challenges groups may have when converting raw video into useable, annotated datasets.

Intraoperative fluorescein video angiography in intracranial aneurysm surgery: single-center, observational cohort study.

The aim of this study was to compare the incidence of postoperative, surgery-related complications in patients where fluorescein video angiography (FL-VA) was performed with those operated without intraoperative verification. This is an observational cohort study including 97 patients who were selected for microsurgical clipping due to intracranial aneurysm. First 52 patients enrolled in the study were operated prior to introduction of fluorescein fluorescence in our surgical workflow. These patients were considered as controls. The study group consisted of 45 consecutive patients operated with the use of fluorescein video angiography and by the same surgical team. Outcomes in both groups were compared using non-parametric test (Mann-Whitney U). Intraoperative fluorescein video angiography revealed aneurysm remnant or inadvertent vessel occlusion in 17.8% of patients. Following clip reposition, a repeated FL-VA was performed to confirm restoration of blood flow and/or complete aneurysm obliteration. Intraoperative findings were later confirmed using computed tomography angiography (CTA). None of the patients in our study group developed surgery-related complications; whereas in the control group, aneurysm remnant was discovered in 7.7%, brain ischemia in 9.6% and both of the latter in 5.8% of patients. Difference in treatment-related outcome was statistically significant (p < 0.05). Intraoperative fluorescein video angiography successfully identified aneurysm residual and adjacent artery occlusion leading to excellent outcome following clip reposition.

Use of surgical video-based automated performance metrics to predict blood loss and success of simulated vascular injury control in neurosurgery: a pilot study.

OBJECTIVE: Experts can assess surgeon skill using surgical video, but a limited number of expert surgeons are available. Automated performance metrics (APMs) are a promising alternative but have not been created from operative videos in neurosurgery to date. The authors aimed to evaluate whether video-based APMs can predict task success and blood loss during endonasal endoscopic surgery in a validated cadaveric simulator of vascular injury of the internal carotid artery. METHODS: Videos of cadaveric simulation trials by 73 neurosurgeons and otorhinolaryngologists were analyzed and manually annotated with bounding boxes to identify the surgical instruments in the frame. APMs in five domains were defined-instrument usage, time-to-phase, instrument disappearance, instrument movement, and instrument interactions-on the basis of expert analysis and task-specific surgical progressions. Bounding-box data of instrument position were then used to generate APMs for each trial. Multivariate linear regression was used to test for the associations between APMs and blood loss and task success (hemorrhage control in less than 5 minutes). The APMs of 93 successful trials were compared with the APMs of 49 unsuccessful trials. RESULTS: In total, 29,151 frames of surgical video were annotated. Successful simulation trials had superior APMs in each domain, including proportionately more time spent with the key instruments in view (p < 0.001) and less time without hemorrhage control (p = 0.002). APMs in all domains improved in subsequent trials after the participants received personalized expert instruction. Attending surgeons had superior instrument usage, time-to-phase, and instrument disappearance metrics compared with resident surgeons (p < 0.01). APMs predicted surgeon performance better than surgeon training level or prior experience. A regression model that included APMs predicted blood loss with an R2 value of 0.87 (p < 0.001). CONCLUSIONS: Video-based APMs were superior predictors of simulation trial success and blood loss than surgeon characteristics such as case volume and attending status. Surgeon educators can use APMs to assess competency, quantify performance, and provide actionable, structured feedback in order to improve patient outcomes. Validation of APMs provides a benchmark for further development of fully automated video assessment pipelines that utilize machine learning and computer vision.

Development of the User Experience (UX) and Video Quality Evaluation (VQE) Instruments for Assessment of Intraoperative Video Capture Technology.

OBJECTIVE: There exists a lack of technology to reliably and routinely capture high-quality video of open surgical procedures. To critically evaluate and compare new and existing technology solutions, we must have widely accepted evaluation criteria for intraoperative camera devices. The objective, therefore, was to develop evaluation criteria for intraoperative camera devices, as well as the video product they produce. DESIGN: A modified Delphi process that included 2 iterative surveys was used to build expert consensus and develop 2 evaluation instruments: one to evaluate the user experience (UX) of using an intraoperative camera device, and the second for video quality evaluation (VQE) of the video product. SETTING: Global, through iterative online surveys. PARTICIPANTS: Surgeons who perform open surgery and have experience with intraoperative video capture. RESULTS: Eighty-six experts participated in the first iteration of the survey and 46 in the second. Ten factors met the a priori cutoff for >80% agreement for the UX survey: (1) ease of setup/integration with current practice, (2) comfort, (3) distracting during case, (4) overall satisfaction with wearing the device, (5) would you use this device again, (6) would you recommend this device to colleagues, (7) the weight of wearing the device, (8) sufficient battery life, (9) ability to control device while operating, and (10) degree to which the device interferes or is incompatible with other surgical accessories. Six factors met the cutoff for the VQE survey: (1) camera stability, (2) brightness/exposure, (3) resolution/sharpness, (4) unobstructed view of the surgical field, (5) appropriate field of view, and (6) overall satisfaction with video quality. CONCLUSIONS: These instruments can be used to critically evaluate camera technologies for intraoperative video capture of open surgery.

Can Intraoperative Video Recordings Contribute to Improving Laparoscopic Percutaneous Extraperitoneal Closure in Children with Inguinal Hernia and Prevent Recurrence? A Pilot Study.

Aim: We reviewed intraoperative video recordings (IVRs) of laparoscopic percutaneous extraperitoneal closure (LPEC) for inguinal hernia in children blindly to assess performance. Methods: IVRs of 183 LPEC performed between April 2013 and March 2016, graded by the operating surgeon as difficult (D; n = 8), straightforward (S; n = 96), or easy (E; n = 79), were scored by a panel of reviewers with advanced (group A; >400 LPEC cases; n = 5), intermediate (group I; 50-150 cases; n = 5), and basic (group B; <10 cases; n = 5) experience, according to suturing, dissection plane, vas/vessel dissection, bleeding, and peritoneal injury. They also allocated a recurrence risk rank (RRR; highest = 6; lowest = 1) for each IVR. Mean score variance for each IVR was also compared between reviewers. Results: There was one recurrence (R; 4-year-old male; level E). RRR were: 1, 2, and 2 for reviewers A, I, and B, respectively. Reviewer A scores for "suturing" and "bleeding," and reviewer I scores for "dissection plane" and "peritoneal injury" correlated significantly with RRR. No reviewer B scores correlated with RRR. Score variance between A and I and A and B for cases D1 and D2 were statistically significant. Conclusion: Advanced reviewers showed greatest variance, questioning the validity of whether experience alone improves surgical technique.

Endoscope-Integrated Fluorescence Video Angiography for the Surgery of Ventrally Located Perimedullary Arteriovenous Fistula at Craniocervical Junction.

BACKGROUND: Intraoperative confirmation of the vascular anatomy and blood flow contributes to the safety of the surgery for perimedullary arteriovenous fistulas (PAVF). However, because the PAVF at the craniocervical junction (CCJ) is mainly located on the ventral spinal cord surface, it is difficult to observe the entire pathology by a conventional surgical approach. To achieve increased viewing angle and visualization of real time blood flow, we introduced endoscope-integrated fluorescein video angiography in the treatment for PAVF at the CCJ for the first time. CASE DESCRIPTION: A 63-year-old man presented with subarachnoid hemorrhage due to rupture of PAVF at the CCJ, fed by both the right C1 radiculomedullary artery and the anterior spinal artery (ASA). Suboccipital craniotomy and C1 hemilaminotomy was performed and microscopic observation revealed partial anatomy of the PAVF covered by subarachnoid clots on the ventrolateral surface at the right C1 nerve root level. However, pathology ventral to the C1 nerve root was obscure and an endoscope-integrated fluorescein video angiography was introduced, which clearly demonstrated the PAVF components and the ASA. CONCLUSIONS: According to these findings, the PAVF was coagulated and the ASA was preserved. Endoscope-integrated fluorescein video angiography allowed to visualize its real-time blood flow, leading to a safe and reliable treatment.

Anterior Approach Combined with Endoscopic Fluorescence Video Angiography for a Cervical Perimedullary Arteriovenous Fistula.

BACKGROUND: Perimedullary arteriovenous fistulas (pAVFs) of the anterior cervical spinal cord are rare and difficult to eradicate by surgery because of the limitations of the approach routes. Because of the anatomic relationships, an anterior approach with corpectomy can provide direct observation. However, a narrow corridor to the lesion is the drawback of this approach. Therefore, to overcome this limitation, we introduced angled endoscopes integrated with fluorescence video angiography to observe the real-time blood flow. CASE DESCRIPTION: A 47-year-old woman was incidentally found to have a pAVF fed by multiple radicular arteries, and she underwent direct surgery via the anterior approach. Although observation of the entire lesion was difficult with the microscope alone, the introduction of the angled endoscope made it possible to observe the lateral portion of the spinal cord hidden behind the dura mater. Furthermore, endoscopic fluorescein video angiography visualized residual fine feeding arteries that were then electrocoagulated, which contributed to complete obliteration of the shunt. CONCLUSIONS: The anterior approach with endoscopic assistance is a reasonable strategy for the treatment of ventrally located cervical pAVFs. Furthermore, integration of a fluorescence video angiography system with the endoscope enables confirmation of the complicated real-time hemodynamics of the pAVFs, contributing to reliable treatment.

Quantitative videographic analysis of intraoperative total shoulder arthroplasty is predictive of radiographic implant loosening.

BACKGROUND: It is believed that both patient and surgeon factors contribute to premature implant loosening. This video study was designed to answer the following questions: Can orthopedic surgeons reliably differentiate between procedures done well and those that will lead to early glenoid failure? Do the difficulty of the operation and the surgeon's performance predict a patient's outcome? Does the presence of a Walch B2 glenoid result in surgery that is evidently more difficult and performed in such a way to suggest early glenoid component failure? METHODS: Eleven upper extremity surgeons blindly graded a set of intraoperative videos of 15 total shoulder arthroplasty patients (grouped by outcome at 2 years). Evaluation questionnaires consisted of questions about the perceived difficulty and the surgeon's performance. Total and partial patient scores were calculated for each video. Higher calculated score would indicate worse postsurgical outcome. RESULTS: The loosening group had a significantly higher total score (P = .0057). Also, patients with B2 glenoids scored significantly higher than patients with other wear type. The analysis of overall procedure performance indicated difference between outcome groups (P = .0063). CONCLUSION: Our results indicate that surgeons could review surgical videos and differentiate the cases that were difficult or those that were more likely to lead to loosening of the glenoid component. The presence of a B2 glenoid was predictive of difficult surgery. The results of this study should serve as a starting point for surgeons interested in critically evaluating performance and also for those interested in finding ways to maximize patient outcomes after total shoulder arthroplasty.

Intraoperative FLOW 800 Analysis for Intramedullary Cystic Lesion: A Technical Case Report.

BACKGROUND AND IMPORTANCE: During intramedullary lesion surgery, the lesion site and the posterior median sulcus (PMS) should be accurately identified prior to myelotomy to avoid severe injury of the posterior funiculus. However, intramedullary lesions are fundamentally invisible until the myelotomy is performed. Furthermore, the PMS location is frequently unclear due to lesion-induced swelling or distortion of the spinal cord. Intraoperative indocyanine green videoangiography (ICG-VA) followed by FLOW 800 analysis, which shows vascularization of the spinal parenchyma, may provide a solution for these problems in specific cases. CLINICAL PRESENTATION: A 61-year-old woman suffering from claudication visited our department. Magnetic resonance imaging (MRI) revealed a cystic lesion at the level of Th11. A solid portion was not detected in the T1-weighted images following gadolinium administration. We made a diagnosis of ventriculus terminalis and performed a lesion resection. Prior to opening the PMS, ICG-VA was performed, which revealed an avascular area representing the intramedullary cyst. The PMS was the most avascular area observed in the time-intensity analysis executed using FLOW 800 software (Zeiss, Oberkochen, Germany). Thus, it was helpful in determining the site for myelotomy, which should be performed at the center of the extent of the lesion. The patient was discharged 23 days after the operation, ambulating independently. CONCLUSION: Intraoperative ICG-VA followed by FLOW 800 analysis was applied to a case of intramedullary cystic lesion. This technique may be helpful in performing safer intramedullary cystic lesion surgery because it enables visualization of the lesion location and confirmation of the PMS.

In Vivo Retinal Vein Bypass Surgery in a Porcine Model.

PURPOSE: To evaluate the feasibility of retinal vein bypass surgery for induced branch retinal-vein occlusion (BRVO) in the living porcine eye. METHODS: Fifteen minipigs were used in the study. Seven days before vascular surgery, hyaluronidase and plasmin were intravitreally injected for induction of posterior vitreous detachment. Aspirin and warfarin were oral administered daily starting 5 d prior to vascular surgery for anti-coagulation. The minipigs were anethetized with an intraperitoneal injection of 300 mg/kg chloral hydrate for intravitreal injection procedure and vascular surgery. Temporary keratoprosthesis vitrectomy was performed, and intraoperative video fluorescein angiography (VFA) was possible. The central and posterior vitreous was removed together with the posterior hyaloid membrane to facilitate vascular maneuvers. BRVO was induced by bipolar diathermy on the vein at the main vein's first branching. Polyimide tubes (50.8-µm internal diameter and 7.6-µm wall thickness) were used as artificial vessels. Vascular manipulation was performed in a bimanual manner. Both end of a prepared tubing was inserted into venous lumen by puncturing and catheterization, and the vein bypass bridging the occlusion was created. Then, the patency of the bypass graft was assessed by intraoperative VFA. RESULTS: The retinal vein bypass surgery was surgically accomplished in 33% (5/15) of the eyes, and the immediate graft patency was confirmed by intraoperative VFA only in one eye. We observed and recorded fluorescein flow from the branch vein to the main vein through the bypass graft which bridging the occlusive vein segment. CONCLUSIONS: We demonstrated the feasibility of retinal vein bypass for induced BRVO in the living porcine eye, and the immediate graft patency was successfully evaluated by intraoperative VFA. Despite the potential, there are still some significant hurdles in vivo retinal vein bypass surgery, and modification of both surgical instruments and maneuvers is needed for further study.