Level Of Surgical Skill Research Articles

Medical Students Cannot Assess Robotic Surgeons Performing Radical Prostatectomy.

Medical students have previously been shown to be just as effective for video rating as experts. We want to compare medical students to experienced surgeons as video assessors of simulated robot-assisted radical prostatectomy (RARP) performance. Video recordings of three RARP modules on the RobotiX (formerly Simbionix) simulator from a previous study were used. Five novice surgeons, five experienced robotic surgeons, and five experienced robotic surgeons in RARP performed a total of 45 video-recorded procedures. The videos were assessed with the modified Global Evaluative Assessment of Robotic Skills tool as both full-length and an edited edition that only included the first 5 minutes of the procedure. Fifty medical students and two experienced RARP surgeons (ES) performed a total of 680 video ratings of full-length videos and 5-minute videos (2-9 ratings per video). Medical students and ES showed poor agreement for both full-length videos and 5-minute videos (0.29 and -0.13, respectively). Medical students could not discriminate between the skill level of the surgeons in either full-length videos or 5-minute videos ( P = 0.053-0.36 and P = 0.21-0.82), whereas ES could discriminate between novice surgeons and experienced surgeons (full-length, P < 0.001, and 5 minutes, P = 0.007) and intermediate and experienced surgeons (full-length, P = 0.001, and 5 minutes, P = 0.01) in both full-length videos and 5-minute videos. We found that medical students cannot be used to assess RARP because they showed poor agreement with the ES rating for both full-length videos and 5-minute videos. Medical students could not discriminate between surgical skill levels.

Skill assessment based on clutch use in cross-platform robot-assisted surgery.

Many studies have investigated the transfer of skills between laparoscopic and robot-assisted surgery (RAS). These studies have considered time, error, and clinical outcomes in the assessment of skill transfer. However, little is known about the specific operations of the surgeon. Clutch control use is an important skill in RAS. Therefore, the present study aimed to propose a novel objective algorithm based on computer vision that can automatically evaluate a surgeon's clutch use. Additionally, the study aimed to evaluate the correlation between clutch metrics and surgical skill on different surgical robot platforms. The robotic surgery training center of Wuhan University trained 30 laparoscopic surgeons as the study group between 2023 and 2024. Laparoscopic surgeons were trained by combining robotic simulator exercises and RAS animal experiments. During the training, video and hand movement data were collected. Hand movements identified by a skin-color model were combined with labeling information to classify clutch use. The metrics were validated on different robotic platforms (dv-Trainer, EDGE MP1000, Toumai™ MT1000, and DaVinci Xi system) and among surgeons with different surgical skill levels. On the robotic simulator, clutch accuracy in the expert group was significantly higher than in the study group for all tasks. No significant differences were observed in the number of clutches between the expert and study groups. In the RAS experiment, the number of clutches decreased significantly for both study and expert groups. The accuracy was maintained at a high level in the expert group but decreased rapidly in the study group. We proposed a new objective assessment of surgical skills, clutch use metrics, in cross-platform RAS. Additionally, we verified that the metrics significantly correlated with the surgical skill levels of the surgeons.

Laparoscopic distal gastrectomy skill evaluation from video: a new artificial intelligence-based instrument identification system

The advent of Artificial Intelligence (AI)-based object detection technology has made identification of position coordinates of surgical instruments from videos possible. This study aimed to find kinematic differences by surgical skill level. An AI algorithm was developed to identify X and Y coordinates of surgical instrument tips accurately from video. Kinematic analysis including fluctuation analysis was performed on 18 laparoscopic distal gastrectomy videos from three expert and three novice surgeons (3 videos/surgeon, 11.6 h, 1,254,010 frames). Analysis showed the expert surgeon cohort moved more efficiently and regularly, with significantly less operation time and total travel distance. Instrument tip movement did not differ in velocity, acceleration, or jerk between skill levels. The evaluation index of fluctuation β was significantly higher in experts. ROC curve cutoff value at 1.4 determined sensitivity and specificity of 77.8% for experts and novices. Despite the small sample, this study suggests AI-based object detection with fluctuation analysis is promising because skill evaluation can be calculated in real time with potential for peri-operational evaluation.

A Contemporary Review of Trachea, Nose, and Ear Cartilage Bioengineering and Additive Manufacturing.

The complex structure, chemical composition, and biomechanical properties of craniofacial cartilaginous structures make them challenging to reconstruct. Autologous grafts have limited tissue availability and can cause significant donor-site morbidity, homologous grafts often require immunosuppression, and alloplastic grafts may have high rates of infection or displacement. Furthermore, all these grafting techniques require a high level of surgical skill to ensure that the reconstruction matches the original structure. Current research indicates that additive manufacturing shows promise in overcoming these limitations. Autologous stem cells have been developed into cartilage when exposed to the appropriate growth factors and culture conditions, such as mechanical stress and oxygen deprivation. Additive manufacturing allows for increased precision when engineering scaffolds for stem cell cultures. Fine control over the porosity and structure of a material ensures adequate cell adhesion and fit between the graft and the defect. Several recent tissue engineering studies have focused on the trachea, nose, and ear, as these structures are often damaged by congenital conditions, trauma, and malignancy. This article reviews the limitations of current reconstructive techniques and the new developments in additive manufacturing for tracheal, nasal, and auricular cartilages.

The Role of Artificial Intelligence in Medical Education: A Systematic Review.

To examine the artificial intelligence (AI) tools currently being studied in modern medical education, and critically evaluate the level of validation and the quality of evidence presented in each individual study. This review (PROSPERO ID: CRD42023410752) was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement. A database search was conducted using PubMed, Embase, and Cochrane Library. Articles written in the English language between 2000 and March 2023 were reviewed retrospectively using the MeSH Terms "AI" and "medical education" A total of 4642 potentially relevant studies were found. After a thorough screening process, 36 studies were included in the final analysis. These studies consisted of 26 quantitative studies and 10 studies investigated the development and validation of AI tools. When examining the results of studies in which Support vector machines (SVMs) were employed, it has demonstrated high accuracy in assessing students' experiences, diagnosing acute abdominal pain, classifying skilled and novice participants, and evaluating surgical training levels. Particularly in the comparison of surgical skill levels, it has achieved an accuracy rate of over 92%. AI tools demonstrated effectiveness in improving practical skills, diagnosing diseases, and evaluating student performance. However, further research with rigorous validation is required to identify the most effective AI tools for medical education.

Is the robotic revolution stunting surgical skills?

This perspective piece aims to examine the impact of the growing utilization of robotic platforms in general and minimally invasive surgery on surgical trainee experience, skill level, and comfort in performing general surgical and minimally invasive procedures following completion of training. We review current literature and explore the application of robotic surgery to surgical training, where minimum case thresholds and breadth distribution are well defined, and where development of surgical technique is historically gained through delicate tissue handling with haptic feedback rather than relying on visual feedback alone. We call for careful consideration as to how best to incorporate robotics in surgical training in order to embrace technological advances without endangering the surgical proficiency of the surgeons of tomorrow. Key messageThe large-scale incorporation of robotics into general and minimally invasive surgical training is something that most, if not all, trainees must grapple with in today's world, and the proportion of robotics is increasing. This shift may significantly negatively affect trainees in terms of surgical skill upon completion of training and must be approached with an appropriate degree of concern and thoughtfulness so as to protect the surgeons of tomorrow.

Lack of stereopsis does not reduce surgical performance but prolongs the learning curve of vitreoretinal surgery.

To investigate whether individuals with long-term reduced stereopsis were able to obtain the same level of surgical skills in simulated vitreoretinal surgery on the Eyesi Surgical Simulator as individuals with normal stereopsis. Twenty-four medical students were recruited and divided into two groups according to their degree of stereopsis: Group 1 (n = 12) included subjects with normal stereopsis (60 arcsec or lower) and Group 2 (n = 12) included subjects with reduced stereopsis (120 arcsec or higher). Stereopsis was tested using the TNO test (Lameris Ootech BV, Nieuwegein, the Netherlands). The participants were trained in virtual reality-simulated vitreoretinal surgery and continuously measured using a test with solid validity evidence and a pre-defined pass-fail score. All data were analysed using the Wilcoxon rank sum test. We observed no differences in overall performance scores for any of the four modules. The participants with reduced stereopsis used 5.8 more attempts in bimanual training (p = 0.04), 8.8 more attempts in removal of posterior hyaloid (p = 0.04), 9.1 more attempts in navigation training (p = 0.20) and 0.3 fewer attempts in removal of internal limiting membrane (p = 0.69). The final performance scores on the Eyesi Surgical Simulator were independent of the degree of stereopsis. However, the number of attempts to achieve the pre-defined pass-fail score increased significantly with reduced stereopsis in two of four modules. These results indicate that a high degree of stereopsis is not necessary to become proficient in microsurgery but may prolong the learning curve.

Causes of Iatrogenic Median Nerve Injury after Endoscopic Carpal Tunnel Release.

Background: Endoscopic carpal tunnel release (ECTR) is a less invasive procedure but has a higher risk of complications. We analysed ECTR cases dividing them into three periods according to a single surgeon's experience when the ECTR was performed: the initial, midterm and late period. Cases of iatrogenically induced median nerve injuries that occurred after ECTR were then noted and evaluated. Methods: We reviewed 195 ECTRs done with the 2-portal technique and divided the patients into three groups according to periods of when ECTR was done. The indications for ECTR surgery were limited to severe CTS cases. These groups of patients were similar in terms of age, duration of disease, electrophysiological study results and severity of the disease. The patients were evaluated for median neuropathy pre- and postoperatively using Semmes-Weinstein monofilament test (SWT), Disabilities of the arm, shoulder and hand (DASH) Score, Coin-flip test (CFT), postoperative paraesthesias and complications, such as pillar pain, and so on. Electrophysiological evaluation was performed only preoperatively. Results: Postoperative median nerve recovery was overall good. Normal recovery was noted in 181 cases (93%). SWT, DASH and CFT were all significantly improved upon follow-up in all three groups. In terms of iatrogenic neuropathy, median nerve palsy worsened (including those transiently worsened) after ECTR in 11 cases (5.6%), even in the later period. The sensory disturbance was equally worsening from the radial to the ulnar side. Conclusions: The fact that there were neurologically worsened cases even in the later period, when the operator is higher skilled in the technique, suggests that the surgical technique itself may be the one posing higher risk than the level of surgical skill. The most likely causes of aggravated nerve palsy were a direct injury by cannula insertion at the proximal portal, or additional median nerve compression during cannula insertion into the carpal tunnel. Level of Evidence: Level IV (Therapeutic).

Transition toward Technological Transformation: Challenges of Implementing Virtual Reality and Augmented Reality in the Health Sector

Abstract Background: Metaverse, the disruptive digital technology, has demonstrated significant effectiveness in the fields of preventive and cognitive therapy, diagnostics, surgical interventions and rehabilitation. Virtual Reality (VR), a part of Metaverse, integrates imaging data and input from users and deliver a 3D graphical output which can be visualised through a wearable headset. Augmented reality (AR) on the other hand, can control the presence of the user in the real world. Methodology: A review was undertaken of peer-reviewed literature on the emerging evidence on the applications of AR and VR in healthcare. Research studies carried out to identify effectiveness of AR and VR technologies were included. Result: AR &amp; VR have been effective in rehabilitation of patients of Autism Spectrum Disorders and Mild Cognitive Impairment by improving motor skills, social skills and various cognitive indices like task learning and attention. In the surgical field, AR head mounted device (HMD) can provide three-dimensional, patient specific anatomic information during surgery. It minimises surgical complications and improves patient satisfaction. AR is of particular interest in complicated spinal surgeries and orthopaedic manoeuvres which require high level of surgical skill. AR has also been used successfully in different types of robotic surgeries as well. In several countries AR technology have been used in basic medical and advanced surgical training. Major challenges in implementing AR and VR in the field of health care persist in the domains of cyber security, ethical issues and cost effectiveness. Conclusion: VR and AR technology can maximise patient outcomes and rapidly develop satisfactory patient management in fields of cognitive research and surgical interventions. More clinical trials with immersive digital technologies are required. Ethical and cyber security challenges are present but there are ways to overcome them. It is our duty as physicians to participate in the development of these innovations to ensure virtual health reality benefits for our patients in real-world setting.

A Deep Learning Approach to Classify Surgical Skill in Microsurgery Using Force Data from a Novel Sensorised Surgical Glove.

Microsurgery serves as the foundation for numerous operative procedures. Given its highly technical nature, the assessment of surgical skill becomes an essential component of clinical practice and microsurgery education. The interaction forces between surgical tools and tissues play a pivotal role in surgical success, making them a valuable indicator of surgical skill. In this study, we employ six distinct deep learning architectures (LSTM, GRU, Bi-LSTM, CLDNN, TCN, Transformer) specifically designed for the classification of surgical skill levels. We use force data obtained from a novel sensorized surgical glove utilized during a microsurgical task. To enhance the performance of our models, we propose six data augmentation techniques. The proposed frameworks are accompanied by a comprehensive analysis, both quantitative and qualitative, including experiments conducted with two cross-validation schemes and interpretable visualizations of the network's decision-making process. Our experimental results show that CLDNN and TCN are the top-performing models, achieving impressive accuracy rates of 96.16% and 97.45%, respectively. This not only underscores the effectiveness of our proposed architectures, but also serves as compelling evidence that the force data obtained through the sensorized surgical glove contains valuable information regarding surgical skill.

Surgical skill level classification model development using EEG and eye-gaze data and machine learning algorithms

The aim of this study was to develop machine learning classification models using electroencephalogram (EEG) and eye-gaze features to predict the level of surgical expertise in robot-assisted surgery (RAS). EEG and eye-gaze data were recorded from 11 participants who performed cystectomy, hysterectomy, and nephrectomy using the da Vinci robot. Skill level was evaluated by an expert RAS surgeon using the modified Global Evaluative Assessment of Robotic Skills (GEARS) tool, and data from three subtasks were extracted to classify skill levels using three classification models—multinomial logistic regression (MLR), random forest (RF), and gradient boosting (GB). The GB algorithm was used with a combination of EEG and eye-gaze data to classify skill levels, and differences between the models were tested using two-sample t tests. The GB model using EEG features showed the best performance for blunt dissection (83% accuracy), retraction (85% accuracy), and burn dissection (81% accuracy). The combination of EEG and eye-gaze features using the GB algorithm improved the accuracy of skill level classification to 88% for blunt dissection, 93% for retraction, and 86% for burn dissection. The implementation of objective skill classification models in clinical settings may enhance the RAS surgical training process by providing objective feedback about performance to surgeons and their teachers.

Developing surgical skill level classification model using visual metrics and a gradient boosting algorithm.

Assessment of surgical skills is crucial for improving training standards and ensuring the quality of primary care. This study aimed to develop a gradient boosting classification model (GBM) to classify surgical expertise into inexperienced, competent, and experienced levels in robot-assisted surgery (RAS) using visual metrics. Eye gaze data were recorded from 11 participants performing four subtasks; blunt dissection, retraction, cold dissection, and hot dissection using live pigs and the da Vinci robot. Eye gaze data were used to extract the visual metrics. One expert RAS surgeon evaluated each participant's performance and expertise level using the modified Global Evaluative Assessment of Robotic Skills (GEARS) assessment tool. The extracted visual metrics were used to classify surgical skill levels and to evaluate individual GEARS metrics. Analysis of Variance (ANOVA) was used to test the differences for each feature across skill levels. Classification accuracies for blunt dissection, retraction, cold dissection, and burn dissection were 95%, 96%, 96%, and 96%, respectively. The time to complete only the retraction was significantly different among the 3 skill levels (p-value = 0.04). Performance was significantly different for 3 categories of surgical skill level for all subtasks (p-values<0.01). The extracted visual metrics were strongly associated with GEARS metrics (R2>0.7 for GEARS metrics evaluation models). Machine learning (ML) algorithms trained by visual metrics of RAS surgeons can classify surgical skill levels and evaluate GEARS measures. The time to complete a surgical subtask may not be considered a stand-alone factor for skill level assessment.

Mental workload during endoscopic sinus surgery is associated with surgeons' skill levels.

Surgeons' mental workload during endoscopic sinus surgery (ESS) has not been fully evaluated. The assessment was challenging due to the great diversity of each patient's anatomy and the consequence variety of surgical difficulties. In this study, we examined the mental workload of surgeons with various surgical skill levels during ESS under the standardized condition provided by novel-designed 3D sinus models. Forty-seven participants performed a high-fidelity ESS simulation with 3D-printed sinus models. Surgeons' mental workload was assessed with the national aeronautics and space administration-task load index (NASA-TLX). Associations between the total and subscales score of NASA-TLX and surgical skill index, including the board certification status, the number of experienced ESS cases, and the objective structured assessment of technical skills (OSATS), were analyzed. In addition, 10 registrars repeated the simulation surgery, and their NASA-TLX score was compared before and after the repetitive training. The total NASA-TLX score was significantly associated with OSATS score (p = 0.0001). Primary component analysis classified the surgeons' mental burden into three different categories: (1) the skill-level-dependent factors (temporal demand, effort, and performance), (2) the skill-level-independent factors (mental and physical demand), and (3) frustration. After the repetitive training, the skill-level-dependent factors were alleviated (temporal demand; z = -2.3664, p = 0.0091, effort; z = -2.1704, p = 0.0346, and performance; z = -2.5992, p = 0.0017), the independent factors were increased (mental demand; z = -2.5992, p = 0.0023 and physical demand; z = -2.2509, p = 0.0213), and frustration did not change (p = 0.3625). Some of the mental workload during ESS is associated with surgical skill level and alleviated with repetitive training. However, other aspects remain a burden or could worsen even when surgeons have gained surgical experience. Routine assessment of registrars' mental burdens would be necessary during surgical training to sustain their mental health.

Motion Smoothness-Based Assessment of Surgical Expertise: The Importance of Selecting Proper Metrics.

The smooth movement of hand/surgical instruments is considered an indicator of skilled, coordinated surgical performance. Jerky surgical instrument movements or hand tremors can cause unwanted damages to the surgical site. Different methods have been used in previous studies for assessing motion smoothness, causing conflicting results regarding the comparison among surgical skill levels. We recruited four attending surgeons, five surgical residents, and nine novices. The participants conducted three simulated laparoscopic tasks, including peg transfer, bimanual peg transfer, and rubber band translocation. Tooltip motion smoothness was computed using the mean tooltip motion jerk, logarithmic dimensionless tooltip motion jerk, and 95% tooltip motion frequency (originally proposed in this study) to evaluate their capability of surgical skill level differentiation. The results revealed that logarithmic dimensionless motion jerk and 95% motion frequency were capable of distinguishing skill levels, indicated by smoother tooltip movements observed in high compared to low skill levels. Contrarily, mean motion jerk was not able to distinguish the skill levels. Additionally, 95% motion frequency was less affected by the measurement noise since it did not require the calculation of motion jerk, and 95% motion frequency and logarithmic dimensionless motion jerk yielded a better motion smoothness assessment outcome in distinguishing skill levels than mean motion jerk.

Novel Low Memory Footprint DNN Models for Edge Classification of Surgeons’ Postures

Skill assessment is fundamental to enhance current laparoscopic surgical training and reduce the incidence of musculoskeletal injuries from performing these procedures. Recently, deep neural networks (DNNs) have been used to improve human posture and surgeons’ skills training. While they work well in the lab, they normally require significant computational power which makes it impossible to use them on edge devices. This letter presents two low memory footprint DNN models used for classifying laparoscopic surgical skill levels at the edge. Trained models were deployed on three Arm Cortex-M processors using the X-Cube-AI and Tensorflow Lite Micro (TFLM) libraries. Results show that the CUBE-AI-based models give the best relative performance, memory footprint, and accuracy tradeoffs when executed on the Cortex-M7.

3D Autonomous Surgeon’s Hand Movement Assessment Using a Cascaded Fuzzy Supervisor in Multi-Thread Video Processing

The purpose of the Fundamentals of Laparoscopic Surgery (FLS) training is to develop laparoscopic surgery skills by using simulation experiences. Several advanced training methods based on simulation have been created to enable training in a non-patient environment. Laparoscopic box trainers-cheap, portable devices-have been deployed for a while to offer training opportunities, competence evaluations, and performance reviews. However, the trainees must be under the supervision of medical experts who can evaluate their abilities, which is an expensive and time-consuming operation. Thus, a high level of surgical skill, determined by assessment, is necessary to prevent any intraoperative issues and malfunctions during a real laparoscopic procedure and during human intervention. To guarantee that the use of laparoscopic surgical training methods results in surgical skill improvement, it is necessary to measure and assess surgeons' skills during tests. We used our intelligent box-trainer system (IBTS) as a platform for skill training. The main aim of this study was to monitor the surgeon's hands' movement within a predefined field of interest. To evaluate the surgeons' hands' movement in 3D space, an autonomous evaluation system using two cameras and multi-thread video processing is proposed. This method works by detecting laparoscopic instruments and using a cascaded fuzzy logic assessment system. It is composed of two fuzzy logic systems executing in parallel. The first level assesses the left and right-hand movements simultaneously. Its outputs are cascaded by the final fuzzy logic assessment at the second level. This algorithm is completely autonomous and removes the need for any human monitoring or intervention. The experimental work included nine physicians (surgeons and residents) from the surgery and obstetrics/gynecology (OB/GYN) residency programs at WMU Homer Stryker MD School of Medicine (WMed) with different levels of laparoscopic skills and experience. They were recruited to participate in the peg-transfer task. The participants' performances were assessed, and the videos were recorded throughout the exercises. The results were delivered autonomously about 10 s after the experiments were concluded. In the future, we plan to increase the computing power of the IBTS to achieve real-time performance assessment.

Comparison Between Lumbar Endoscopic Unilateral Laminotomy for Bilateral Decompression and Other Decompression Approaches for Lumbar Spinal Stenosis: A Systematic Review

Lumbar spinal stenosis affects numerous people globally. Full-endoscopic uniportal interlaminar decompression (FEUID) for lumbar spinal stenosis results in satisfactory outcomes. In this systematic review, we compared technical methods, surgical outcomes, and complications among different types of surgical techniques and discussed the effect of different surgical skill levels. A systematic review of studies published from 1990 to January 2022 was performed. Studies related to FEUID were identified using the keywords "interlaminar decompression," "endoscopy," "uniportal," and "percutaneous." The outcomes measured were operative time, blood loss, hospital stay, complications, visual analog scale scores, Oswestry Disability Index scores, and the Macnab criteria. Ten of 306 studies were eligible for inclusion. For FEUID, data for 580 patients and more than 367 levels were collected. All the studies reported significant improvement in mean visual analog scale and Oswestry Disability Index scores, and the mean overall complication rate was 9.5%. Compared with other surgical techniques, FEUID resulted in lower visual analog scale and Oswestry Disability Index scores, complication rates, and blood loss and shorter hospital stay. These surgical parameters were considerably affected by a surgeon's skill level. FEUID results in better patient satisfaction with more favorable surgical outcomes and fewer complications. Although more prospective randomized controlled studies are required to confirm these findings, our results indicate that FEUID is a reasonable alternative to traditional lumbar spinal surgery.

7943 Modification of the “Aagl Endometriosis Classification” to a 3-Stage System: Diagnostic Accuracy & Inter-Observer Reproducibility Study

<h3>Study Objective</h3> We aim to assess the performance of a modified AAGL (mAAGL) endometriosis classification system and its inter-observer reproducibility. <h3>Design</h3> Retrospective diagnostic accuracy & inter-observer reproducibility study. <h3>Setting</h3> 2 gynecology units in metropolitan Sydney, Australia. <h3>Patients or Participants</h3> Patients with suspected endometriosis who underwent laparoscopy between January 2016 and October 2021 (n=272). <h3>Interventions</h3> Patients with complete surgical data included. The original 4-stage American Association of Gynecologic Laparoscopists (AAGL) staging system was modified to three stages by merging AAGL stages 2 and 3 and surgical complexity levels B and C. Three observers independently apportioned an AAGL stage 1-3, blinded to surgical skill level (A, B and C). End points were diagnostic performance of each observer to predict level of surgical complexity for each stage, i.e., mAAGL stage 1 to predict level A, mAAGL stage 2 to predict level B, mAAGL stage 3 to predict level C. Inter-observer reproducibility was assessed using OR of a higher grade, between each pair of observers. <h3>Measurements and Main Results</h3> 272 surgical cases analyzed. The diagnostic accuracy for all observers to predict each level in terms of sensitivity, specificity, PPV and NPV were: stage 1 to predict level A: 98.0-98.7%, 60.2-64.2%, 75.0-76.9% & 96.3-97.5%; stage 2 to predict level B: 31.1-36.9%, 97.0-98.2%, 86.5-92.6% & 69.8 – 71.9%; stage 3 to predict level C: 90.0-95.0%, 90.1-91.7%, 41.8 - 47.5% & 99.1-99.6%. Inter-observer reproducibility between observer 1 & 2 was 0.81, between observer 1 & 3 was 0.69 and between observer 2 & 3 was 0.43. Overall percentage agreement between observers 1 & 2 was 93.4%, between 1 & 3 was 93.4% & between 2 & 3 was 96.3%. <h3>Conclusion</h3> The mAAGL system appears to perform well for both low and high stage endometriosis disease. The 4-stage AAGL staging system could be improved by merging AAGL stages 2 and 3 and in turn using a 3-stage system.

Comparison of the Recently Published “AAGL Endometriosis Classification” with Other Staging Systems to Predict Surgical Complexity

<h3>Study Objective</h3> We aim to assess the performance of the "2021 AAGL Endometriosis Classification" compared with revised American Society of Reproductive Medicine (rASRM), Ultrasound Based Endometriosis Staging System (UBESS) and modified Ultrasound Based Endometriosis Staging System (mUBESS) to predict AAGL levels of surgical complexity. <h3>Design</h3> Multicenter retrospective diagnostic accuracy study. <h3>Setting</h3> 2 gyneacology units in metropolitan Sydney, Australia. <h3>Patients or Participants</h3> Patients with suspected endometriosis who underwent laparoscopy between January 2016 and October 2021 (n=272). <h3>Interventions</h3> Database of patients with comprehensive surgical data including location & severity of endometriosis & surgical procedure performed was analysed. AAGL surgical complexity level (A to D) was pre-determined by consensus between three observers. For the three stage UBESS to be compared with the four levels of AAGL surgical complexity, levels B and C were merged. rASRM stage 1, 2, 3 and 4 predicted AAGL level A, B, C and D, respectively. UBESS stage 1, 2 and 3 predicted AAGL level A, B/C and D, respectively. mUBESS stage 1, 2, 3 and 4 predicted AAGL level A, B, C and D, respectively. The performance of all staging systems was assessed using kappa, accuracy, sensitivity, specificity, positive predictive value (PPV,) negative predictive value (NPV) and likelihood ratios. <h3>Measurements and Main Results</h3> Across all diagnostic accuracy measurements there was no clear difference between the four staging tools. For overall comparison, kappa/weighted kappa scores respectively were: For AAGL 0.42 & 0.55, for rASRM 0.41 & 0.55, for UBESS 0.44 & 0.57 and for mUBESS 0.47 & 0.58. <h3>Conclusion</h3> Across all tools, there was good concordance for mild and severe disease to predict surgical skill level. For intermediate disease there was weak concordance. There was no clearly superior tool.

Improving situation recognition using endoscopic videos and navigation information for endoscopic sinus surgery.

Endoscopic sinus surgery (ESS) is widely used to treat chronic sinusitis. However, it involves the use of surgical instruments in a narrow surgical field in close proximity to vital organs, such as the brain and eyes. Thus, an advanced level of surgical skill is expected of surgeons performing this surgery. In a previous study, endoscopic images and surgical navigation information were used to develop an automatic situation recognition method in ESS. In this study, we aimed to develop a more accurate automatic surgical situation recognition method for ESS by improving the method proposed in our previous study and adding post-processing to remove incorrect recognition. We examined the training model parameters and the number of long short-term memory (LSTM) units, modified the input data augmentation method, and added post-processing. We also evaluated the modified method using clinical data. The proposed improvements improved the overall scene recognition accuracy compared with the previous study. However, phase recognition did not exhibit significant improvement. In addition, the application of the one-dimensional median filter significantly reduced short-time false recognition compared with the time series results. Furthermore, post-processing was required to set constraints on the transition of the scene to further improve recognition accuracy. We suggested that the scene recognition could be improved by considering the model parameter, adding the one-dimensional filter and post-processing. However, the scene recognition accuracy remained unsatisfactory. Thus, a more accurate scene recognition and appropriate post-processing method is required.