ABSTRACT
Background: Diffuse midline gliomas (DMG) are aggressive pediatric brain tumors that are diagnosed and monitored through MRI. We developed an automatic pipeline to segment subregions of DMG and select radiomic features that predict patient overall survival (OS). Methods: We acquired diagnostic and post-radiation therapy (RT) multisequence MRI (T1, T1ce, T2, and T2 FLAIR) and manual segmentations from 2 centers: 53 from 1 center formed the internal cohort and 16 from the other center formed the external cohort. We pretrained a deep learning model on a public adult brain tumor data set (BraTS 2021), and finetuned it to automatically segment tumor core (TC) and whole tumor (WT) volumes. PyRadiomics and sequential feature selection were used for feature extraction and selection based on the segmented volumes. Two machine learning models were trained on our internal cohort to predict patient 12-month survival from diagnosis. One model used only data obtained at diagnosis prior to any therapy (baseline study) and the other used data at both diagnosis and post-RT (post-RT study). Results: Overall survival prediction accuracy was 77% and 81% for the baseline study, and 85% and 78% for the post-RT study, for internal and external cohorts, respectively. Homogeneous WT intensity in baseline T2 FLAIR and larger post-RT TC/WT volume ratio indicate shorter OS. Conclusions: Machine learning analysis of MRI radiomics has potential to accurately and noninvasively predict which pediatric patients with DMG will survive less than 12 months from the time of diagnosis to provide patient stratification and guide therapy.
ABSTRACT
OBJECTIVE: Neurosurgery subinternships are a critical portion of the medical student application to neurosurgery residency programs, allowing programs to assess the student's clinical knowledge, interpersonal skills, work ethic, and character. Despite how critical these auditions are, many students have a poor understanding of expectations prior to beginning these subinternships. Thomas Jefferson University hosted a combined in-person and virtual boot camp session open to all medical students interested in neurosurgery. The authors sought to determine the effectiveness of this inaugural course. METHODS: A total of 304 registered participants were sent a survey assessing their attitudes toward neurosurgery subinternships, beliefs about their abilities, and their comfort with various neurosurgical skills. All participants were sent a postsession survey composed of the same questions. The mean scores for responses to pre- and postsession survey questions were recorded based on graduating year and by medical school type (US allopathic [US MD], US osteopathic [US DO], or foreign degree/international medical graduate [IMG]). Differences in means between pre- and postsession survey responses were analyzed using the Student t-test, and statistical significance was set at p < 0.05. RESULTS: A total of 112 presession surveys and 64 postsession surveys were completed, yielding a presession survey response rate of 36.8% and a postsession survey response rate of 21.1%. Seventy-five percent of the postsession survey respondents attended virtually, and 25% were in-person. US MD, US DO, and IMG attendees demonstrated a significantly increased understanding of the expectations of a neurosurgery subintern (p < 0.001). All students had significantly increased confidence in their ability to succeed as subinterns (US MD students and IMGs p < 0.001, US DO students p < 0.05). Regarding procedural confidence, US MD students had increased confidence in craniotomies and cranial plating (p < 0.001). When comparing responses by graduation year, students in the classes of 2024 and 2025 (rising 4th-year and rising 3rd-year medical students, respectively) demonstrated significantly increased understanding of expectations and confidence in their ability to succeed (< 0.001). Seventy-five percent of our postsession survey respondents attended virtually, and 25% were in-person. The in-person cohort had greater improvements in comfort with procedures such as craniotomies, cranial plating, and extraventricular drain placement (in-person vs Zoom mean differences: craniotomies and cranial plating, -2.29, extraventricular drain placement, -2.31) (p < 0.05). CONCLUSIONS: The boot camp successfully delineated the expectations of neurosurgery subinterns and enhanced the attendees' confidence in their abilities. The authors concluded that a hybrid virtual and in-person format is beneficial and feasible in increasing accessibility to information about neurosurgery subinternships.