Patient-specific prediction of SEEG electrode bending for stereotactic neurosurgical planning.

PURPOSE : Electrode bending observed after stereotactic interventions is typically not accounted for in either computer-assisted planning algorithms, where straight trajectories are assumed, or in quality assessment, where only metrics related to entry and target points are reported. Our aim is to provide a fully automated and validated pipeline for the prediction of stereo-electroencephalography (SEEG) electrode bending. METHODS : We transform electrodes of 86 cases into a common space and compare features-based and image-based neural networks on their ability to regress local displacement ([Formula: see text]) or electrode bending ([Formula: see text]). Electrodes were stratified into six groups based on brain structures at the entry and target point. Models, both with and without Monte Carlo (MC) dropout, were trained and validated using tenfold cross-validation. RESULTS : mage-based models outperformed features-based models for all groups, and models that predicted [Formula: see text] performed better than for [Formula: see text]. Image-based model prediction with MC dropout resulted in lower mean squared error (MSE) with improvements up to 12.9% ([Formula: see text]) and 39.9% ([Formula: see text]), compared to no dropout. Using an image of brain tissue types (cortex, white and deep grey matter) resulted in similar, and sometimes better performance, compared to using a T1-weighted MRI when predicting [Formula: see text]. When inferring trajectories of image-based models (brain tissue types), 86.9% of trajectories had an MSE[Formula: see text] mm. CONCLUSION : An image-based approach regressing local displacement with an image of brain tissue types resulted in more accurate electrode bending predictions compared to other approaches, inputs, and outputs. Future work will investigate the integration of electrode bending into planning and quality assessment algorithms.

Simulating various levels of clinical challenge in the assessment of clinical procedure competence.

STUDY OBJECTIVE: Immersive simulation is increasingly used for competency assessment of emergency physicians. This group's concept of hybrid simulation (HS) (combining simulated patients and part-task trainers (a simulator that simulates a limited component of a clinical procedure) to create a multimodal clinical context) requires clinicians to integrate technical and nontechnical skills in a holistic clinical performance for assessment. It also offers the potential to provide authentic simulation of a given clinical procedure across multiple levels of challenge. The aims of this study are to systematically design and validate 2 patient-focused HS scenarios (each combining a simulated patient with a part-task simulator) for assessment of the management of a commonly encountered problem in an emergency department (ED) at different levels of clinical challenge, and to explore the effect of level of challenge of the HS scenario on physicians' performance. METHODS: A simple (HS1) and a complex (HS2) HS scenario (based on the management of a patient with a traumatic skin laceration within the ED) was developed according to expert opinion through cognitive task analysis. Interns and emergency medicine residents (stratified into expert and novice groups according to experience) were recruited to participate in both scenarios. Participants were randomized to perform either the HS1 or HS2 scenario first. Participants completed a questionnaire for face validity (realism of simulation) and content validity (comprehensiveness of simulation). Performances were assessed by 2 independent raters using validated rating tools modified to the needs of this study: the Modified Objective Structured Assessment of Technical Skills-Task Specific Checklist, the Objective Structured Assessment of Technical Skills-Global Rating Score, and the Direct Observation of Procedural Skills. RESULTS: Ten novice and 10 expert clinicians completed both scenarios. Mean face and content validity ratings were high for both HS1 (mean 4.4 [SD 0.52] and 4.2 [SD 0.48], respectively) and HS2 scenarios (mean 4.5 [SD 0.35] and 4.3 [SD 0.43], respectively). In HS1, no difference was found between experts' and novices' Modified Objective Structured Assessment of Technical Skills-Task Specific Checklist, Objective Structured Assessment of Technical Skills-Global Rating Score, and Direct Observation of Procedural Skills ratings. Experts performed significantly better than novices in HS2 in terms of the 3 tools' ratings. Novices' Modified Objective Structured Assessment of Technical Skills-Task Specific Checklist and Direct Observation of Procedural Skills ratings were significantly worse in HS2 compared with HS1, but no difference was found with the Objective Structured Assessment of Technical Skills-Global Rating Score. No statistical difference was found in experts' Modified Objective Structured Assessment of Technical Skills-Task Specific Checklist, Objective Structured Assessment of Technical Skills-Global Rating Score, and Direct Observation of Procedural Skills ratings between HS2 and HS1 scenarios. CONCLUSION: Recreating clinical challenge is an important consideration in the design of simulation-based assessment of procedural skills of clinicians. In this study, we have demonstrated a systematic approach to developing HS scenarios, which may be able to recreate various levels of clinical challenge for purpose of assessment of procedural skills.