Brandon Edelbach
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Brandon Edelbach has a total of 38 co-authors. Here are the top co-authors by number of shared publications:
- Miguel Angel Lopez-Gonzalez
Department of Neurosurgery, Loma Linda University Medical Center, Loma Linda, California, United States
3 shared publications Rasha Elbadry
Department of Neurosurgery, Loma Linda University Medical Center, Loma Linda, California, United States of America
3 shared publications- Ahmad Kareem Almekkawi
Department of Neurological Surgery, Mayo Clinic, Phoenix, Arizona, USA; Neurosurgery Simulation and Innovation Lab, Mayo Clinic, Phoenix, Arizona, USA; Precision Neuro-therapeutics Innovation Lab, Mayo Clinic, Phoenix, Arizona, USA
2 shared publications - Carlos Bagley
Saint Luke's Marion Bloch Neuroscience Institute Department of Neurosurgery, 4401 Wornall Rd., Kansas City, MO 64111, USA; University of Missouri-Kansas City School of Medicine, 2411 Holmes St, Kansas City, MO 64108, USA
2 shared publications James P. Caruso
The University of Texas Southwestern Department of Neurosurgery, 5151 Harry Hines Blvd, Dallas, TX 75235, USA
2 shared publicationsTarek Y. El Ahmadieh
Saint Luke's Marion Bloch Neuroscience Institute Department of Neurosurgery, 4401 Wornall Rd., Kansas City, MO 64111, USA
2 shared publications
Latest Publications
- Open Access
- 15 June 2026
- Published by JMIR Publications Inc. in JMIR Preprints
Performance of AtlasGPT and OpenEvidence on the United States Medical Licensing Examination: A comparative study of two retrieval-augmented medical AI systems on open-ended USMLE questions (Preprint)
BACKGROUND Large language models (LLMs) now routinely meet or exceed the passing threshold on the United States Medical Licensing Examination (USMLE), with systems such as OpenEvidence reported to achieve perfect scores. Despite this, most evaluations use multiple-choice formats and may not reflect performance on open-ended questions which more closely reflect real life scenarios. Specialized systems like OpenEvidence and AtlasGPT seek to improve reliability through retrieval-augmented generation and structured reasoning. OBJECTIVE To benchmark AtlasGPT against the widely used, high-performing OpenEvidence system by comparing their performance on identical sets of USMLE-style questions across Steps 1, 2, and 3, in an open-ended format. METHODS Two LLMs were tested on 366 publicly available USMLE questions. Each model was presented with only the question stem and generated a single final answer. Three physicians scored responses against official keys using a binary rubric. Inter-rater reliability (Fleiss’ κ) and head-to-head accuracy (McNemar’s test) were calculated; accuracy is reported with 95% confidence intervals and head-to-head differences with matched odds ratios. RESULTS Across 366 items, AtlasGPT achieved 71.86% accuracy (263/366; 95% CI, 0.670–0.762) versus OpenEvidence 65.30% (239/366; 95% CI, 0.603–0.700); head-to-head testing favored AtlasGPT (McNemar χ²=7.3472, p=0.0067; matched OR=2.00, 95% CI, 1.23–3.27). Effects were consistent by step—significant in Step 1 (OR=2.71, 95% CI, 1.14–6.46; p=0.031) and not significant in Step 2 (p=0.210) or Step 3 (p=0.405); no heterogeneity was detected (Q=0.843, p=0.656). Inter-rater agreement was substantial (κ: AtlasGPT 0.847, 95% CI, 0.796–0.893; OpenEvidence 0.809, 95% CI, 0.757–0.856); the difference was not significant (Δκ=0.0388, p=0.085). CONCLUSIONS AtlasGPT outperformed OpenEvidence on open-ended USMLE-style questions, suggesting that meaningful performance differences exist between advanced RAG systems even when both use curated medical corpora.
- 15 June 2026
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessPerformance of AtlasGPT and OpenEvidence on the United States Medical Licensing Examination: A comparative study of two retrieval-augmented medical AI systems on open-ended USMLE questions (Preprint)
BACKGROUND Large language models (LLMs) now routinely meet or exceed the passing threshold on the United States Medical Licensing Examination (USMLE), with systems such as OpenEvidence reported to achieve perfect scores. Despite this, most evaluations use multiple-choice formats and may not reflect performance on open-ended questions which more closely reflect real life scenarios. Specialized systems like OpenEvidence and AtlasGPT seek to improve reliability through retrieval-augmented generation and structured reasoning. OBJECTIVE To benchmark AtlasGPT against the widely used, high-performing OpenEvidence system by comparing their performance on identical sets of USMLE-style questions across Steps 1, 2, and 3, in an open-ended format. METHODS Two LLMs were tested on 366 publicly available USMLE questions. Each model was presented with only the question stem and generated a single final answer. Three physicians scored responses against official keys using a binary rubric. Inter-rater reliability (Fleiss’ κ) and head-to-head accuracy (McNemar’s test) were calculated; accuracy is reported with 95% confidence intervals and head-to-head differences with matched odds ratios. RESULTS Across 366 items, AtlasGPT achieved 71.86% accuracy (263/366; 95% CI, 0.670–0.762) versus OpenEvidence 65.30% (239/366; 95% CI, 0.603–0.700); head-to-head testing favored AtlasGPT (McNemar χ²=7.3472, p=0.0067; matched OR=2.00, 95% CI, 1.23–3.27). Effects were consistent by step—significant in Step 1 (OR=2.71, 95% CI, 1.14–6.46; p=0.031) and not significant in Step 2 (p=0.210) or Step 3 (p=0.405); no heterogeneity was detected (Q=0.843, p=0.656). Inter-rater agreement was substantial (κ: AtlasGPT 0.847, 95% CI, 0.796–0.893; OpenEvidence 0.809, 95% CI, 0.757–0.856); the difference was not significant (Δκ=0.0388, p=0.085). CONCLUSIONS AtlasGPT outperformed OpenEvidence on open-ended USMLE-style questions, suggesting that meaningful performance differences exist between advanced RAG systems even when both use curated medical corpora.
- 13 February 2026
- Published by Elsevier in Neurochirurgie
- Vol. 72 (3) , 101786
Single position prone versus dual-position in lateral transpsoas approach for lumbar interbody fusion and fixation: propensity matched non-inferiority analysis for spondylosis
- 13 February 2026
- Published by Elsevier in Neurochirurgie
- Vol. 72 (3) , 101786
Single position prone versus dual-position in lateral transpsoas approach for lumbar interbody fusion and fixation: propensity matched non-inferiority analysis for spondylosis
- 31 July 2025
- Published by Elsevier in Neurochirurgie
- Vol. 71 (5) , 101707
Response to Critique of “On the Origin of Olfactory Schwannomas: A Systematic Review and Analysis of Attachment Site Variability and Clinical Implications”
- 31 July 2025
- Published by Elsevier in Neurochirurgie
- Vol. 71 (5) , 101707
Response to Critique of “On the Origin of Olfactory Schwannomas: A Systematic Review and Analysis of Attachment Site Variability and Clinical Implications”
- Open Access
- 23 July 2025
- Published by Elsevier in World Neurosurgery
- Vol. 201, 124321
Dural Venous Sinus Stenting in Pediatric Refractory Idiopathic Intracranial Hypertension: A Case Series and Literature Review
- 23 July 2025
- Published by Elsevier in World Neurosurgery
- Vol. 201, 124321
Open AccessDural Venous Sinus Stenting in Pediatric Refractory Idiopathic Intracranial Hypertension: A Case Series and Literature Review
- 11 June 2025
- Published by Elsevier in Journal of Clinical Neuroscience
- Vol. 138, 111371
Outcomes of surgically managed brainstem cavernous Malformations: A systematic review and Meta-Analysis
- 11 June 2025
- Published by Elsevier in Journal of Clinical Neuroscience
- Vol. 138, 111371
Outcomes of surgically managed brainstem cavernous Malformations: A systematic review and Meta-Analysis
- Open Access
- 29 May 2025
- Published by Elsevier in World Neurosurgery
- Vol. 200, 124129
En Bloc Spondylectomy versus Separation Surgery for Spinal Metastases: A Systematic Review and Meta-Analysis
- 29 May 2025
- Published by Elsevier in World Neurosurgery
- Vol. 200, 124129
Open AccessEn Bloc Spondylectomy versus Separation Surgery for Spinal Metastases: A Systematic Review and Meta-Analysis
- 16 April 2025
- Published by Elsevier in Neurochirurgie
- Vol. 71 (4) , 101673
On the origin of olfactory schwannomas: A systematic review and analysis of attachment site variability and clinical implications
- 16 April 2025
- Published by Elsevier in Neurochirurgie
- Vol. 71 (4) , 101673
On the origin of olfactory schwannomas: A systematic review and analysis of attachment site variability and clinical implications
- Open Access
- 1 January 2025
- Published by Wiley in Behavioural Neurology
- Vol. 2025 (1) , 5717503
Predictors of Postoperative Epilepsy/Seizures in Patients Readmitted After Meningioma Resection
Background: Epilepsy/seizures in meningioma patients may occur pre‐ or postoperatively, causing significant morbidity and impaired quality of life. Surgical excision is considered a standard management with variable rates of epilepsy/seizure resolution reported after surgery. Employing a national database, we examined the pre‐ and postoperative incidences of epilepsy/seizures and risk factors associated with postoperative epilepsy/seizures in patients readmitted within 30 days and/or 90 days following meningioma resection. Methods: The 2010–2014 Nationwide Readmissions Database was analyzed. Consecutive patients undergoing surgery for meningioma resection were identified using appropriate ICD‐9‐CM codes. Standard descriptive techniques and multivariate regression were used to identify predictors of postoperative epilepsy/seizure after discharge. Results: Among 46,107 patients undergoing meningioma resection at index hospitalization, 20.40% ( n = 9408) had preoperative epilepsy/seizure diagnosis. The mean patient age was 58.37 ± 13.85 years. Patients with preoperative epilepsy/seizures were more likely to be male ( p < 0.001), frail ( p < 0.001), and with higher comorbidity index scores ( p < 0.001). The overall readmission rate was 30.36% and was higher among patients with preoperative epilepsy/seizures (36.66% vs. 28.75%, p < 0.001). Respectively, 30‐ and 90‐day readmission rates were higher among patients (13.22% vs. 11.73%, p < 0.001) and (23.25% vs 20.30%, p = 0.04) with epilepsy/seizure diagnosis at index admission. Predictors of postoperative epilepsy/seizures at 30‐ and 90‐day readmissions included the preoperative epilepsy/seizure, malignant meningioma, peritumoral cerebral edema, and higher comorbidity index scores, while male sex was significant only at 30‐day readmissions. Intraoperative electrocorticography was associated with a decreased likelihood of postoperative epilepsy/seizures. Conclusion: Development of epilepsy/seizures after meningioma resection is likely multifactorial. Identifying factors associated with postoperative epilepsy/seizures after discharge is important in triaging and closer monitoring of at‐risk patients and for adapting management to help improve outcomes.
- 1 January 2025
- Published by Wiley in Behavioural Neurology
- Vol. 2025 (1) , 5717503
Open AccessPredictors of Postoperative Epilepsy/Seizures in Patients Readmitted After Meningioma Resection
Background: Epilepsy/seizures in meningioma patients may occur pre‐ or postoperatively, causing significant morbidity and impaired quality of life. Surgical excision is considered a standard management with variable rates of epilepsy/seizure resolution reported after surgery. Employing a national database, we examined the pre‐ and postoperative incidences of epilepsy/seizures and risk factors associated with postoperative epilepsy/seizures in patients readmitted within 30 days and/or 90 days following meningioma resection. Methods: The 2010–2014 Nationwide Readmissions Database was analyzed. Consecutive patients undergoing surgery for meningioma resection were identified using appropriate ICD‐9‐CM codes. Standard descriptive techniques and multivariate regression were used to identify predictors of postoperative epilepsy/seizure after discharge. Results: Among 46,107 patients undergoing meningioma resection at index hospitalization, 20.40% ( n = 9408) had preoperative epilepsy/seizure diagnosis. The mean patient age was 58.37 ± 13.85 years. Patients with preoperative epilepsy/seizures were more likely to be male ( p < 0.001), frail ( p < 0.001), and with higher comorbidity index scores ( p < 0.001). The overall readmission rate was 30.36% and was higher among patients with preoperative epilepsy/seizures (36.66% vs. 28.75%, p < 0.001). Respectively, 30‐ and 90‐day readmission rates were higher among patients (13.22% vs. 11.73%, p < 0.001) and (23.25% vs 20.30%, p = 0.04) with epilepsy/seizure diagnosis at index admission. Predictors of postoperative epilepsy/seizures at 30‐ and 90‐day readmissions included the preoperative epilepsy/seizure, malignant meningioma, peritumoral cerebral edema, and higher comorbidity index scores, while male sex was significant only at 30‐day readmissions. Intraoperative electrocorticography was associated with a decreased likelihood of postoperative epilepsy/seizures. Conclusion: Development of epilepsy/seizures after meningioma resection is likely multifactorial. Identifying factors associated with postoperative epilepsy/seizures after discharge is important in triaging and closer monitoring of at‐risk patients and for adapting management to help improve outcomes.
- 1 October 2024
- Published by Elsevier in Neurochirurgie
- Vol. 70 (6) , 101601
Systematic review and comparative analysis of endovascular and microsurgical management of giant ruptured fusiform mca aneurysms with illustrative cases
- 1 October 2024
- Published by Elsevier in Neurochirurgie
- Vol. 70 (6) , 101601
Systematic review and comparative analysis of endovascular and microsurgical management of giant ruptured fusiform mca aneurysms with illustrative cases