Self-assessment, and not continuous training, improves basic open suturing skills.

BACKGROUND: To develop and maintain suturing skills, clinical exposure is important. When clinical exposure cannot be guaranteed, an adequate training schedule for suturing skills is required. This study evaluates the effect of continuous training, 'reflection before practice' and self-assessment on basic open suturing skills. METHODS: Medical students performed four basic suturing tasks on a simulation set up before ('pre-test') and after their surgical rotation ('after-test'). Participants were divided in three groups; the 'clinical exposure group' (n = 44) had clinical exposure during their rotation only, the 'continuous training group' (n = 16) completed a suturing interval training during their rotation and the 'self-assessment group' (n = 16) also completed a suturing interval training, but with the use of reflection before practice and self-assessment. Parameters measured by a tracking system during the suturing tasks and a calculated 'composite score' were compared between groups and test-moments. RESULTS: A significantly better composite score was found at the after-test compared to the pre-test for all groups for all basic suturing tasks (0.001 ≤ p ≤ 0.049). The self-assessment group scored better at the pre-test than the other two groups for all tasks, except for 'knot tying by hand' (0.004 ≤ p ≤ 0.063). However, this group did not score better at the after-test for all tasks, compared to the other two groups. This resulted in a smaller delta of time ('transcutaneous suture', p = 0.013), distance ('Donati suture' and 'intracutaneous suture', 0.005 ≤ p ≤ 0.009) or composite score (all tasks, except for knot tying by hand, 0.007 ≤ p ≤ 0.061) in the self-assessment group. CONCLUSION: Reflection before practice and self-assessment during continuous training of basic open suturing tasks, may improve surgical skills at the start of the learning curve.

Concurrent validity of objective assessment by finger tracking for open surgical suturing training.

BACKGROUND: The aim of this study is to evaluate whether the parameters 'time' and 'distance', measured by SurgTrac, correlate with the assessment of the same skills by blinded experts. METHODS: Basic open suturing tasks were executed by medical students. SurgTrac software measured objective parameters by tracking fingers. The executed tasks were recorded by a tablet and additionally assessed by a blinded expert with a Competency Assessment Tool-form (CAT-form). A Pearson's correlation was used to investigate the correlation between the parameters and the outcomes of the expert assessment. RESULTS: A strong correlation between the measured parameters of SurgTrac and the expert-assessment was found for knot tying by hand (r â€‹= â€‹-0.703) and vertical mattress suture (r â€‹= â€‹-0.644) and a moderate correlation for transcutaneous suture (r â€‹= â€‹-0.555) and intracutaneous suture (r â€‹= â€‹-0.451). CONCLUSION: The use of finger tracking by SurgTrac showed a good concurrent validity for the basic open suturing tasks knot tying by hand, transcutaneous suture and vertical mattress suture.

Alcohol's impact on fine motor skills: Insights from minimally invasive surgical simulation.

Background: Alcohol misuse among medical professionals poses a significant concern, and there is a lack of clarity in (inter)national guidelines regarding alcohol use during work. Moreover, there exists an insufficient body of research on the specific impact of alcohol on fine motor skills within the medical sector, specifically surgery. This study aims to investigate the impact of alcohol on fine motor skills in a minimally invasive surgical setting. Methods: A cross-sectional study was conducted at Lowland Science on August 19th, 20th, and 21st, 2022, during the Lowlands music festival in Biddinghuizen, the Netherlands. Participants were divided into five groups based on measured alcohol consumption. Exclusion criteria included drug use, prior surgical experience, being underage, and previous participation. The main outcomes were the number of correctly transferred rings during the PEG transfer task and the number of errors. Blood alcohol concentration (BAC) was measured using a breathalyser. Results: A total of 1056 participants were included in the study. The results indicated an inverse relationship between BAC levels and surgical performance, with higher alcohol levels associated with a decrease in performance (p = 0.023). However, there was no significant difference in the number of errors among the five groups (p = 0.597). The group with the highest alcohol consumption (BAC >0.08 %) exhibited significantly worse performance compared to the group with a BAC of 0.0 % (p = 0.002). Conclusion: This study uncovers a negative impact of increased alcohol intake on fine motor skills in a minimally invasive surgery simulation exercise. While there was no effect on the occurrence of errors. Professional medical organizations should reconsider and explicate their position on alcohol use in (surgical) healthcare.

Congenital Vascular Malformations in Children: From Historical Perspective to a Multidisciplinary Approach in the Modern Era-A Comprehensive Review.

Congenital vascular malformations (CVMs) are the result of an aberrant development during embryogenesis. Although these lesions are present at birth, they are not always visible yet. Once symptomatic, patients suffer from pain, bleeding, ulcers, infections or lymphatic leakage, depending on the subtype of vessels involved. Treatment includes conservative management, surgery, sclerotherapy, embolization and pharmacological therapy. The clinical presentation varies widely and treatment can be challenging due to the rarity of the disease and potential difficulties of treatment. This review gives an overview of the historical developments in diagnosis and classification and exposes the key elements of innovations in the past decades on the identification of genetic mutations and personalized treatment. These advances in the field and a multidisciplinary approach are highly valuable in the optimization of clinical care aimed at both curing or stabilizing the CVM and pursuing physical and psychosocial wellbeing.

Experts in Minimally Invasive Surgery are Outperformed by Trained Novices on Suturing Skills.

INTRODUCTION: Learning minimally invasive suturing can be challenging, creating a barrier to further implementation, especially with the development of easier methods. Nevertheless, mastering intracorporeal knot tying is crucial when alternative techniques prove inadequate. Therefore, the minimally invasive surgery (MIS) suturing skills of MIS experts are compared with a group of novices during their learning curve on a simulator. METHODS: The novice participants repeatedly performed the intracorporeal suturing task on the EoSim MIS simulator (up to a maximum of 20 repetitions). The experts (>50 MIS procedures and advanced MIS experience) completed the same task once. The first and last exercises of the novices and the expert tasks were all blindly recorded and assessed by two independent assessors using the Laparoscopic Suturing Competency Assessment Tool (LS-CAT). Additionally, objective assessment parameters, "time" and "distance", using instrument tracking, were collected. The scores of the experts were then compared with the novices. RESULTS: At the end of the training, novices significantly outperformed the experts on both the expert assessment (LS-CAT: 16.8 versus 26.8, P = 0.001) and objective parameters (median time: 190 s versus 161 s, P < 0.001; median distance: 6.1 m versus 3.6 m, P < 0.001). Although the experts showed slightly better performance than the novices during their first task, the difference was not significant on the expert assessment (LS-CAT experts 16.8, novices 20.5, P = 0.057). CONCLUSIONS: Our findings underscore the significance of continued MIS suturing training for both residents and surgeons. In this study, trained novices demonstrated a significant outperformance of experts on both quantitative and qualitative outcome parameters within a simulated setting.

Where is VALDO? VAscular Lesions Detection and segmentatiOn challenge at MICCAI 2021.

Imaging markers of cerebral small vessel disease provide valuable information on brain health, but their manual assessment is time-consuming and hampered by substantial intra- and interrater variability. Automated rating may benefit biomedical research, as well as clinical assessment, but diagnostic reliability of existing algorithms is unknown. Here, we present the results of the VAscular Lesions DetectiOn and Segmentation (Where is VALDO?) challenge that was run as a satellite event at the international conference on Medical Image Computing and Computer Aided Intervention (MICCAI) 2021. This challenge aimed to promote the development of methods for automated detection and segmentation of small and sparse imaging markers of cerebral small vessel disease, namely enlarged perivascular spaces (EPVS) (Task 1), cerebral microbleeds (Task 2) and lacunes of presumed vascular origin (Task 3) while leveraging weak and noisy labels. Overall, 12 teams participated in the challenge proposing solutions for one or more tasks (4 for Task 1-EPVS, 9 for Task 2-Microbleeds and 6 for Task 3-Lacunes). Multi-cohort data was used in both training and evaluation. Results showed a large variability in performance both across teams and across tasks, with promising results notably for Task 1-EPVS and Task 2-Microbleeds and not practically useful results yet for Task 3-Lacunes. It also highlighted the performance inconsistency across cases that may deter use at an individual level, while still proving useful at a population level.

Health-related quality of life in children with congenital vascular malformations.

A cross-sectional study was performed to evaluate health-related quality of life (HRQOL) in children with congenital vascular malformations (CVM) and to investigate factors associated with an impaired HRQOL. Children (2-17 years) with CVMs who visited the HECOVAN expertise center between 2016-2018 were included. The PedsQL 4.0 Generic Core Scales were used and a score ≥ 1.0 SD below the normative mean was defined as an impaired HRQOL. Factors associated with impairment were investigated using univariate and multivariate logistic regression analysis. The median overall HRQOL was 84.8/100 (n = 207; 41% boys, 59% girls; self-reported IQR 73.9-92.4 and parent-reported IQR 71.4-92.4). Patients aged 13-17 years reported significantly worse physical functioning than those aged 8-12 years (median 84.4, IQR 71.1-93.8 versus median 90.6, IQR 81.3-96.9; p = 0.02). Parents reported a significantly lower overall HRQOL than their children (median 80.4, IQR 70.7-90.8 versus median 85.9, IQR 76.1-92.4; p = 0.001). HRQOL was impaired in 25% of patients. Impairment occurred significantly more often in lower extremity CVMs (38%, p = 0.01) and multifocal CVMs (47%, p = 0.01) compared to CVMs in the head/neck region (13%). Other associated factors included invasive management (31% versus 14%; p = 0.01), age at first treatment ≤ 5 years (48% versus 25%; p = 0.02) and ongoing treatment (38% versus 18%; p = 0.004). After correction for other factors, significance remained for lower extremity CVMs and ongoing invasive treatment. CONCLUSIONS: Overall median HRQOL was reasonable and not significantly different from the norm sample. Parental ratings were significantly lower than their children's ratings. A quarter of the patients had an impaired HRQOL, which seemed to worsen with age. Independently associated factors included a lower extremity CVM and invasive management. WHAT IS KNOWN: • Congenital vascular malformations could affect health-related quality of life (HRQOL). • Studies on pediatric patients are limited and either very small or in combination with adult patient series. WHAT IS NEW: • This study raises awareness of an impaired HRQOL in 25% of pediatric patients with congenital vascular malformations. • Associated factors included a lower extremity CVM and invasive management.

Assessing Quantitative Parenchymal Features at Baseline Dynamic Contrast-enhanced MRI and Cancer Occurrence in Women with Extremely Dense Breasts.

Background Automated identification of quantitative breast parenchymal enhancement features on dynamic contrast-enhanced (DCE) MRI scans could provide added value in assessment of breast cancer risk in women with extremely dense breasts. Purpose To automatically identify quantitative properties of the breast parenchyma on baseline DCE MRI scans and assess their association with breast cancer occurrence in women with extremely dense breasts. Materials and Methods This study represents a secondary analysis of the Dense Tissue and Early Breast Neoplasm Screening trial. MRI was performed in eight hospitals between December 2011 and January 2016. After segmentation of fibroglandular tissue, quantitative features (including volumetric density, volumetric morphology, and enhancement characteristics) of the parenchyma were extracted from baseline MRI scans. Principal component analysis was used to identify parenchymal measures with the greatest variance. Multivariable Cox proportional hazards regression was applied to assess the association between breast cancer occurrence and quantitative parenchymal features, followed by stratification of significant features into tertiles. Results A total of 4553 women (mean age, 55.7 years ± 6 [SD]) with extremely dense breasts were included; of these women, 122 (3%) were diagnosed with breast cancer. Five principal components representing 96% of the variance were identified, and the component explaining the greatest independent variance (42%) consisted of MRI features relating to volume of enhancing parenchyma. Multivariable analysis showed that volume of enhancing parenchyma was associated with breast cancer occurrence (hazard ratio [HR], 1.09; 95% CI: 1.01, 1.18; P = .02). Additionally, women in the high tertile of volume of enhancing parenchyma showed a breast cancer occurrence twice that of women in the low tertile (HR, 2.09; 95% CI: 1.25, 3.61; P = .005). Conclusion In women with extremely dense breasts, a high volume of enhancing parenchyma on baseline DCE MRI scans was associated with increased occurrence of breast cancer as compared with a low volume of enhancing parenchyma. © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Grimm in this issue.

Magnitude and relevance of change in health-related quality of life in patients with vascular malformations treated with sirolimus.

Introduction: Vascular malformations are rare congenital anomalies of the vascular system, which can involve the capillaries, veins, arteries, lymphatics, or a combination of vessel types. Patients with vascular malformations experience an impaired health-related quality of life (HRQoL) because of their symptoms (e.g., pain, swelling, and bleeding) and psychosocial distress. Sirolimus is an effective drug used in the medical treatment of these patients; however, relatively little is known about the effect of sirolimus on specific changes in the HRQoL domains and its magnitude. Methods: The magnitude of change (effect size) following intervention is more informative to clinical practitioners than statistically significant but clinically unimportant changes; therefore, this study aimed to examine the magnitude and meaningfulness of change in the HRQoL of children and adults with vascular malformations following sirolimus treatment using low target levels. Results: In total, 50 patients with vascular malformations (19 children, 31 adults) were included in this study. These patients experienced a lower HRQoL than the general population, with the adults reporting a significantly lower score in almost all domains. A 6-month sirolimus treatment improved the HRQoL in 29 patients, including 77.8% of the children (Pediatric Quality of Life Inventory score [PedsQL]) and 57.7% of the adults (Short Form 36 [SF-36]). The effect sizes of sirolimus for each SF-36/PedsQL domain ranged from 0.19 to 1.02. The clinically relevant moderate magnitude of changes was seen in the domains of the children's reports: "Physical functioning" and "Social functioning" and in the domains of the parent reports: "Social functioning," "School functioning," and "Psychosocial." A high-magnitude change was seen in the domains "Emotional functioning" and "Psychosocial" in the children's reports and "Physical functioning" in the parent reports. In addition, the moderate magnitude of changes was also seen in the adults SF-36: in all domains except for "Role limitations-physical problems," "Role limitations-emotional problems," and "General health perception." Conclusion: We believe this is the first study showing the magnitude of change in HRQoL after sirolimus treatment in patients with vascular malformations. Before treatment, these patients experienced an impaired HRQoL compared with the general Dutch population. A 6-month sirolimus treatment with low target levels led to moderate-to-high clinically relevant changes in multiple domains, which significantly improved the HRQoL. Clinical trial registration: https://clinicaltrials.gov/ct2/show/NCT03987152?cond=Vascular+Malformations&cntry=NL&city=Nijmegen&draw=2&rank=1, identifier: NCT03987152.

Radiologic versus Segmentation Measurements to Quantify Wilms Tumor Volume on MRI in Pediatric Patients.

Wilms tumor is a common pediatric solid tumor. To evaluate tumor response to chemotherapy and decide whether nephron-sparing surgery is possible, tumor volume measurements based on magnetic resonance imaging (MRI) are important. Currently, radiological volume measurements are based on measuring tumor dimensions in three directions. Manual segmentation-based volume measurements might be more accurate, but this process is time-consuming and user-dependent. The aim of this study was to investigate whether manual segmentation-based volume measurements are more accurate and to explore whether these segmentations can be automated using deep learning. We included the MRI images of 45 Wilms tumor patients (age 0-18 years). First, we compared radiological tumor volumes with manual segmentation-based tumor volume measurements. Next, we created an automated segmentation method by training a nnU-Net in a five-fold cross-validation. Segmentation quality was validated by comparing the automated segmentation with the manually created ground truth segmentations, using Dice scores and the 95th percentile of the Hausdorff distances (HD95). On average, manual tumor segmentations result in larger tumor volumes. For automated segmentation, the median dice was 0.90. The median HD95 was 7.2 mm. We showed that radiological volume measurements underestimated tumor volume by about 10% when compared to manual segmentation-based volume measurements. Deep learning can potentially be used to replace manual segmentation to benefit from accurate volume measurements without time and observer constraints.

Objective assessment for open surgical suturing training by finger tracking can discriminate novices from experts.

It is difficult, time consuming and expensive to assess manual skills in open surgery. The aim of this study is to investigate the construct validity of a low-cost, easily accessible tracking technique for basic open suturing tasks. Medical master students, surgical residents, and surgeons at the Radboud University Medical Center were recruited between September 2020 until September 2021. The participants were divided, according to experience, in a novice group (≤10 sutures performed) and an expert group (>50 sutures performed). For objective tracking, a tablet with SurgTrac software was used, which tracked a blue and a red tag placed on respectively their left and right index finger. The participants executed four basic tasks on a suturing model: 1) knot tying by hand, 2) transcutaneous suturing with an instrument knot, 3) 'Donati' (vertical mattress suture) with an instrument knot and 4) continuous intracutaneous suturing without a knot. In total 76 participants were included: 57 novices and 19 experts. All four tasks showed significant differences between the novice group and expert group for the parameters time (p<0.001), distance (p<0.001 for Task 1, 2 and 3 and p=0.034 for Task 4) and smoothness (p<0.001). Additionally, Task 3 showed a significant difference for the parameter handedness (p=0.006) and Task 4 for speed (p=0.033). Tracking index finger movements using SurgTrac software on a tablet while executing basic open suturing skills on a simulator shows excellent construct validity for time, distance and motion smoothness in all four suturing tasks.

Deep Learning-Based Segmentation of Locally Advanced Breast Cancer on MRI in Relation to Residual Cancer Burden: A Multi-Institutional Cohort Study.

BACKGROUND: While several methods have been proposed for automated assessment of breast-cancer response to neoadjuvant chemotherapy on breast MRI, limited information is available about their performance across multiple institutions. PURPOSE: To assess the value and robustness of deep learning-derived volumes of locally advanced breast cancer (LABC) on MRI to infer the presence of residual disease after neoadjuvant chemotherapy. STUDY TYPE: Retrospective. SUBJECTS: Training cohort: 102 consecutive female patients with LABC scheduled for neoadjuvant chemotherapy (NAC) from a single institution (age: 25-73 years). Independent testing cohort: 55 consecutive female patients with LABC from four institutions (age: 25-72 years). FIELD STRENGTH/SEQUENCE: Training cohort: single vendor 1.5 T or 3.0 T. Testing cohort: multivendor 3.0 T. Gradient echo dynamic contrast-enhanced sequences. ASSESSMENT: A convolutional neural network (nnU-Net) was trained to segment LABC. Based on resulting tumor volumes, an extremely randomized tree model was trained to assess residual cancer burden (RCB)-0/I vs. RCB-II/III. An independent model was developed using functional tumor volume (FTV). Models were tested on an independent testing cohort and response assessment performance and robustness across multiple institutions were assessed. STATISTICAL TESTS: The receiver operating characteristic (ROC) was used to calculate the area under the ROC curve (AUC). DeLong's method was used to compare AUCs. Correlations were calculated using Pearson's method. P values <0.05 were considered significant. RESULTS: Automated segmentation resulted in a median (interquartile range [IQR]) Dice score of 0.87 (0.62-0.93), with similar volumetric measurements (R = 0.95, P < 0.05). Automated volumetric measurements were significantly correlated with FTV (R = 0.80). Tumor volume-derived from deep learning of DCE-MRI was associated with RCB, yielding an AUC of 0.76 to discriminate between RCB-0/I and RCB-II/III, performing similar to the FTV-based model (AUC = 0.77, P = 0.66). Performance was comparable across institutions (IQR AUC: 0.71-0.84). DATA CONCLUSION: Deep learning-based segmentation estimates changes in tumor load on DCE-MRI that are associated with RCB after NAC and is robust against variations between institutions. EVIDENCE LEVEL: 2. TECHNICAL EFFICACY: Stage 4.

Long-term Survival in Breast Cancer Patients Is Associated with Contralateral Parenchymal Enhancement at MRI: Outcomes of the SELECT Study.

Background Several single-center studies found that high contralateral parenchymal enhancement (CPE) at breast MRI was associated with improved long-term survival in patients with estrogen receptor (ER)-positive and human epidermal growth factor receptor 2 (HER2)-negative breast cancer. Due to varying sample sizes, population characteristics, and follow-up times, consensus of the association is currently lacking. Purpose To confirm whether CPE is associated with long-term survival in a large multicenter retrospective cohort, and to investigate if CPE is associated with endocrine therapy effectiveness. Materials and Methods This multicenter observational cohort included women with unilateral ER-positive HER2-negative breast cancer (tumor size ≤50 mm and ≤three positive lymph nodes) who underwent MRI from January 2005 to December 2010. Overall survival (OS), recurrence-free survival (RFS), and distant RFS (DRFS) were assessed. Kaplan-Meier analysis was performed to investigate differences in absolute risk after 10 years, stratified according to CPE tertile. Multivariable Cox proportional hazards regression analysis was performed to investigate whether CPE was associated with prognosis and endocrine therapy effectiveness. Results Overall, 1432 women (median age, 54 years [IQR, 47-63 years]) were included from 10 centers. Differences in absolute OS after 10 years were stratified according to CPE tertile as follows: 88.5% (95% CI: 88.1, 89.1) in tertile 1, 85.8% (95% CI: 85.2, 86.3) in tertile 2, and 85.9% (95% CI: 85.4, 86.4) in tertile 3. CPE was independently associated with OS, with a hazard ratio (HR) of 1.17 (95% CI: 1.0, 1.36; P = .047), but was not associated with RFS (HR, 1.11; P = .16) or DRFS (HR, 1.11; P = .19). The effect of endocrine therapy on survival could not be accurately assessed; therefore, the association between endocrine therapy efficacy and CPE could not reliably be estimated. Conclusion High contralateral parenchymal enhancement was associated with a marginally decreased overall survival in patients with estrogen receptor-positive and human epidermal growth factor receptor 2-negative breast cancer, but was not associated with recurrence-free survival (RFS) or distant RFS. Published under a CC BY 4.0 license. Supplemental material is available for this article. See also the editorial by Honda and Iima in this issue.

Clinical differences in sirolimus treatment with low target levels between children and adults with vascular malformations - A nationwide trial.

The clinical presentation of patients with slow-flow vascular malformations is very heterogeneous. High clinical burden and subsequent reduced health-related quality of life is something they have in common. There is an unmet medical need for these patients for whom regular treatments like surgery and embolization are either insufficient or technically impossible. Sirolimus has been reported to be effective and overall well-tolerated in most patients. However, the main limitation of sirolimus is the reported high toxicity, especially when target levels of 10-15 ng/mL are being used. We report the results of a phase IIB single-arm open-label clinical trial consisting of 68 (67 in the challenge phase and 68 in the rechallenge phase) evaluable patients (children n = 33 and adults n = 35) demonstrating that treatment with low sirolimus target levels (4-10 ng/mL) is effective in 79.1% of the patients. When sirolimus treatment was stopped, the majority of patients experienced a recurrence of symptoms, supporting prolonged or even lifelong treatment requirement. Adults experienced a higher baseline pain score compared with children, having an estimated marginal mean of 6.2 versus 4.1, p < 0.05; however, they showed a similar decrease to children. Furthermore, the pediatric population experienced less often a sirolimus-related grade I-IV adverse event (35.9% vs. 64.1%, p > 0.05) compared with adults. Additionally, response rates were higher in children compared with adults (93.8% vs. 65.7%, p < 0.05), and children responded faster (28 vs. 91 days, p < 0.05). These results suggest benefits of sirolimus in patients with slow-flow vascular malformations and support its initiation as young as possible.

Validation of Combined Deep Learning Triaging and Computer-Aided Diagnosis in 2901 Breast MRI Examinations From the Second Screening Round of the Dense Tissue and Early Breast Neoplasm Screening Trial.

OBJECTIVES: Computer-aided triaging (CAT) and computer-aided diagnosis (CAD) of screening breast magnetic resonance imaging have shown potential to reduce the workload of radiologists in the context of dismissing normal breast scans and dismissing benign disease in women with extremely dense breasts. The aim of this study was to validate the potential of integrating CAT and CAD to reduce workload and workup on benign lesions in the second screening round of the DENSE trial, without missing cancer. METHODS: We included 2901 breast magnetic resonance imaging scans, obtained from 8 hospitals in the Netherlands. Computer-aided triaging and CAD were previously developed on data from the first screening round. Computer-aided triaging dismissed examinations without lesions. Magnetic resonance imaging examinations triaged to radiological reading were counted and subsequently processed by CAD. The number of benign lesions correctly classified by CAD was recorded. The false-positive fraction of the CAD was compared with that of unassisted radiological reading in the second screening round. Receiver operating characteristics (ROC) analysis was performed and the generalizability of CAT and CAD was assessed by comparing results from first and second screening rounds. RESULTS: Computer-aided triaging dismissed 950 of 2901 (32.7%) examinations with 49 lesions in total; none were malignant. Subsequent CAD classified 132 of 285 (46.3%) lesions as benign without misclassifying any malignant lesion. Together, CAT and CAD yielded significantly fewer false-positive lesions, 53 of 109 (48.6%) and 89 of 109 (78.9%), respectively ( P = 0.001), than radiological reading alone. Computer-aided triaging had a smaller area under the ROC curve in the second screening round compared with the first, 0.83 versus 0.76 ( P = 0.001), but this did not affect the negative predictive value at the 100% sensitivity operating threshold. Computer-aided diagnosis was not associated with significant differences in area under the ROC curve (0.857 vs 0.753, P = 0.08). At the operating thresholds, the specificities of CAT (39.7% vs 41.0%, P = 0.70) and CAD (41.0% vs 38.2%, P = 0.62) were successfully reproduced in the second round. CONCLUSION: The combined application of CAT and CAD showed potential to reduce workload of radiologists and to reduce number of biopsies on benign lesions. Computer-aided triaging (CAT) correctly dismissed 950 of 2901 (32.7%) examinations with 49 lesions in total; none were malignant. Subsequent computer-aided diagnosis (CAD) classified 132 of 285 (46.3%) lesions as benign without misclassifying any malignant lesion. Together, CAT and CAD yielded significantly fewer false-positive lesions, 53 of 109 (48.6%) and 89 of 109 (78.9%), respectively ( P = 0.001), than radiological reading alone.

Assessment of Minimally Invasive Suturing Skills: Is Instrument Tracking an Accurate Prediction?

Introduction: Minimally invasive surgery (MIS) suturing demands advanced surgical skills. Therefore, it is important these skills are adequately trained and assessed. Assessment and feedback can consist of judgments and scores of expert observers or objective parameters using instrument tracking. The aim of this study was to determine to what extent objective parameters correspond to expert assessment. Methods: Participants performed an intracorporeal suturing task on the EoSim simulator repeatedly (maximum 20 repetitions) during training. The best discriminating parameters, which previously shown construct validation, were combined into a composite score, using regression analysis. All videos were blinded and assessed by 2 independent reviewers using the validated laparoscopic suturing competency assessment tool (LS-CAT). These scores were compared with the composite score. Results: A 100 videos of 16 trainees, during separate points on their learning curve, and 8 experts were used. The parameters "time" and "distance" were statistically significantly correlated with all LS-CAT domains. The composite score (calculated from "time" and "distance") showed improvement between the first and the last knot (57% versus 94%, P < .001). Also the LS-CAT score improved (28 versus 17, P < .001). However, the correlation of the composite score with the LS-CAT score was weak (R: 0.351), with an accuracy of 55/100 when pooling the outcomes based on inadequate, adequate, or good performance. Conclusion: Instrument tracking parameters (using Surgtrac) could give an indication of the skill level, however, it missed important elements, essential for reliable assessment. Therefore, expert assessment remains superior to determine the skill level in MIS suturing skills.

The feasibility and benefit of unsupervised at-home training of minimally invasive surgical skills.

BACKGROUND: Simulation-based training may be used to acquire MIS skills. While mostly done in a simulation center, it is proposed that this training can be undertaken at-home as well. The aim of this study is to evaluate whether unsupervised at-home training and assessment of MIS skills is feasible and results in increased MIS skills. METHODS: Medical doctors and senior medical students were tested on their innate abilities by performing a pre-test on a take-home simulator. Henceforth, they followed a two-week interval training practicing two advanced MIS skills (an interrupted suture with knot tying task and a precise peg transfer task) and subsequently performed a post-test. Both tests and all training moments were performed at home. Performance was measured using motion analysis software (SurgTrac) and by expert-assessment and self-assessment using a competency assessment tool for MIS suturing (LS-CAT). RESULTS: A total of 38 participants enrolled in the study. Participants improved significantly between the pre-test and the post-test for both tasks. They were faster (632 s vs. 213 s, p < 0.001) and more efficient (distance of instrument tips: 9.8 m vs. 3.4 m, p = 0.001) in the suturing task. Total LS-CAT scores, rated by an expert, improved significantly with a decrease from 36 at the pre-test to 20 at the post-test (p < 0.001) and showed a strong correlation with self-assessment scores (R 0.771, p < 0.001). The precise peg transfer task was completed faster (300 s vs. 163 s, p < 0.001) and more efficient as well (14.8 m vs. 5.7 m, p = 0.005). Additionally, they placed more rings correctly (7 vs. 12, p = 0.010). CONCLUSION: Unsupervised at-home training and assessment of MIS skills is feasible and resulted in an evident increase in skills. Especially in times of less exposure in the clinical setting and less education on training locations this can aid in improving MIS skills.

Evaluating the Use of a Take-Home Minimally Invasive Surgery Box Training for At-Home Training Sessions Before and During the COVID Pandemic.

Introduction: Acquiring and retaining minimally invasive surgery (MIS) skills may be aided by using MIS take-home box simulators. This study aims to evaluate the use of a take-home MIS box trainer for training sessions at-home. Methods: Trainees who previously bought the LaparoscopyBoxx (between 2016 and 2020) were asked to complete an online questionnaire. The first part was on their demographic data, and the second part was on their use and opinion on the box trainer (on a 5-point Likert scale). Results: A total of 39 participants (9 surgeons and 30 trainees) completed the questionnaire (response rate 40%). Overall, participants had a positive opinion on the use of the box trainer and would recommend it to others for training (mean 4.2). Basic tasks, such as needle trail and ring transfer, were practiced most often and were regarded most relevant (mean 4.1-4.3). They regarded practicing on the stimulator of added value to practicing in the clinic (mean 4.2), were more confident during clinical procedures (mean 3.9), and felt that their skills had improved (mean 4.1). Of quarter of the participants who bought the MIS box trainer during the COVID-19 pandemic reported an increased use during times of less clinical exposure. Conclusion: Surgeons and trainees have a positive opinion on using a take-home MIS box trainer for at-home training for general as well as for pediatric MIS skills. Especially during periods of less exposure in the clinical setting and less hands-on courses, such as during the COVID-19 pandemic, at-home training may be a valuable addition.

Explainable artificial intelligence (XAI) in deep learning-based medical image analysis.

With an increase in deep learning-based methods, the call for explainability of such methods grows, especially in high-stakes decision making areas such as medical image analysis. This survey presents an overview of explainable artificial intelligence (XAI) used in deep learning-based medical image analysis. A framework of XAI criteria is introduced to classify deep learning-based medical image analysis methods. Papers on XAI techniques in medical image analysis are then surveyed and categorized according to the framework and according to anatomical location. The paper concludes with an outlook of future opportunities for XAI in medical image analysis.