Mostafa Fatehi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Mostafa Fatehi has a total of 54 co-authors. Here are the top co-authors by number of shared publications:
- Ghassan Hamarneh
Simon Fraser University
4 shared publications - Weina Jin
Simon Fraser University
4 shared publications - Pedram Laghaei Farimani
Combined Neurosurgical and Orthopedic Spine Program, University of British Columbia, Vancouver, Canada
3 shared publications Brian D. Toyota
Division of Neurosurgery, University of California, San Francisco, San Francisco, CA, USA
2 shared publicationsBrian Toyota
Department of Surgery Kingston General Hospital Queens University Kingston Canada
2 shared publications- Camille Hunt
14 Division of Neurology, University of British Columbia, Vancouver, B.C, Canada
2 shared publications
Latest Publications
- 7 January 2026
- Published by Elsevier in Trends in Cancer
- Vol. 12 (3) , 248-261
Speed meets precision: rapid intraoperative diagnostics in neuro-oncologic surgery
- 7 January 2026
- Published by Elsevier in Trends in Cancer
- Vol. 12 (3) , 248-261
Speed meets precision: rapid intraoperative diagnostics in neuro-oncologic surgery
- 16 June 2025
- Published by Cambridge University Press (CUP) in Canadian Journal of Neurological Sciences / Journal Canadien des Sciences Neurologiques
- Vol. 53 (3) , 469-471
Primary Central Nervous System Melanoma with Bilateral Vestibular Lesions Mimicking Neurofibromatosis Type 2
Primary Central Nervous System Melanoma With Bilateral Vestibular Lesions Mimicking Neurofibromatosis Type 2
- 16 June 2025
- Published by Cambridge University Press (CUP) in Canadian Journal of Neurological Sciences / Journal Canadien des Sciences Neurologiques
- Vol. 53 (3) , 469-471
Primary Central Nervous System Melanoma with Bilateral Vestibular Lesions Mimicking Neurofibromatosis Type 2
Primary Central Nervous System Melanoma With Bilateral Vestibular Lesions Mimicking Neurofibromatosis Type 2
- Open Access
- 5 November 2024
- Published by Cambridge University Press (CUP) in Canadian Journal of Neurological Sciences / Journal Canadien des Sciences Neurologiques
- p. 1-5
The Effect of After-Hours Resection on the Outcomes in Patients with High-Grade Gliomas
Objective: The “weekend effect” is the finding that patients presenting for medical care outside of regular working hours tend to have worse outcomes. There is a paucity of literature in the neuro-oncology space exploring this effect. We investigated the extent of resection and complication rates in patients undergoing after-hours high-grade glioma resection. Methods: A retrospective review was conducted on patients with high-grade glioma requiring emergent surgery between January 1, 2021, and March 31, 2023. After hours was defined as surgical resection on the weekend and/or evening (>50% of surgical time between 1630 and 0659). These patients were matched to patients undergoing resection during regular working hours. Groups were compared on the basis of the extent of resection, postoperative complications and 6-month mortality rate. Results: A total of 38 patients were included in this study (19 after hours, 19 regular hours). There was no significant difference in age, sex, tumor grade and tumor size between the two groups (all p > 0.05). There was no significant difference in the extent of resection between the groups (p = 0.7442). There was no significant difference in the rate of intraoperative complications, postoperative complications, reoperation and death at 6 months between the groups (all p > 0.05). Estimated blood loss was significantly higher in the regular hours group (p = 0.0278). There was no significant difference in the total operative time (p = 0.0643) and length of stay (p = 0.0601). Conclusions: After-hours high-grade glioma surgery has similar outcomes to regular-hours surgery for lesions not requiring specialized functional mapping.
- 5 November 2024
- Published by Cambridge University Press (CUP) in Canadian Journal of Neurological Sciences / Journal Canadien des Sciences Neurologiques
- p. 1-5
Open AccessThe Effect of After-Hours Resection on the Outcomes in Patients with High-Grade Gliomas
Objective: The “weekend effect” is the finding that patients presenting for medical care outside of regular working hours tend to have worse outcomes. There is a paucity of literature in the neuro-oncology space exploring this effect. We investigated the extent of resection and complication rates in patients undergoing after-hours high-grade glioma resection. Methods: A retrospective review was conducted on patients with high-grade glioma requiring emergent surgery between January 1, 2021, and March 31, 2023. After hours was defined as surgical resection on the weekend and/or evening (>50% of surgical time between 1630 and 0659). These patients were matched to patients undergoing resection during regular working hours. Groups were compared on the basis of the extent of resection, postoperative complications and 6-month mortality rate. Results: A total of 38 patients were included in this study (19 after hours, 19 regular hours). There was no significant difference in age, sex, tumor grade and tumor size between the two groups (all p > 0.05). There was no significant difference in the extent of resection between the groups (p = 0.7442). There was no significant difference in the rate of intraoperative complications, postoperative complications, reoperation and death at 6 months between the groups (all p > 0.05). Estimated blood loss was significantly higher in the regular hours group (p = 0.0278). There was no significant difference in the total operative time (p = 0.0643) and length of stay (p = 0.0601). Conclusions: After-hours high-grade glioma surgery has similar outcomes to regular-hours surgery for lesions not requiring specialized functional mapping.
- 2 January 2024
- Published by Elsevier in Artificial Intelligence in Medicine
- Vol. 148, 102751
Evaluating the clinical utility of artificial intelligence assistance and its explanation on the glioma grading task
- 2 January 2024
- Published by Elsevier in Artificial Intelligence in Medicine
- Vol. 148, 102751
Evaluating the clinical utility of artificial intelligence assistance and its explanation on the glioma grading task
- Open Access
Evaluating the Clinical Utility of Artificial Intelligence Assistance and its Explanation on Glioma Grading Task
Background: As a fast-advancing technology, artificial intelligence (AI) has considerable potential to assist physicians in various clinical tasks from disease identification to lesion segmentation. Despite much research, AI has not yet been applied to neurooncological imaging in a clinically meaningful way. To bridge the clinical implementation gap of AI in neuro-oncological settings, we conducted a clinical user-based evaluation, analogous to the phase II clinical trial, to evaluate the utility of AI for diagnostic predictions and the value of AI explanations on the glioma grading task. Method: Using the publicly-available BraTS dataset, we trained an AI model of 88.0% accuracy on the glioma grading task. We selected the SmoothGrad explainable AI Weina Jin and Mostafa Fatehi are co-first authors. algorithm based on the computational evaluation regarding explanation truthfulness among a candidate of 16 commonly-used algorithms. SmoothGrad could explain the AI model’s prediction using a heatmap overlaid on the MRI to highlight important regions for AI prediction. The evaluation is an online survey wherein the AI prediction and explanation are embedded. Each of the 35 neurosurgeon participants read 25 brain MRI scans of patients with gliomas, and gave their judgment on the glioma grading without and with the assistance of AI’s prediction and explanation. Result: Compared to the average accuracy of 82.5 ± 8.7% when physicians perform the task alone, physicians’ task performance increased to 87.7 ± 7.3% with statistical significance ( p -value = 0.002) when assisted by AI prediction, and remained at almost the same level of 88.5 ± 7.0% ( p -value = 0.35) with the additional AI explanation assistance. Conclusion: The evaluation shows the clinical utility of AI to assist physicians on the glioma grading task. It also reveals the limitations of applying existing AI explanation techniques in clinical settings. Key points: Phase II evaluation with 35 neurosurgeons on the clinical utility of AI and its explanation AI prediction assistance improved physicians’ performance on the glioma grading task Additional AI explanation assistance did not yield a performance boost Importance of the study: This study is the first phase II AI clinical evaluation in neuro-oncology. Evaluating AI is a prerequisite for its clinical deployment. The four phases of AI clinical evaluation are analogous to the four phases of clinical trials. Prior works that apply AI in neurooncology utilize phase I algorithmic evaluation, which do not reflect how AI can be used in clinical settings to support physician decision making. To bridge the research gap, we conducted the first clinical evaluation to assess the joint neurosurgeon-AI task performance. The evaluation also includes AI explanation as an indispensable feature for AI clinical deployment. Results from quantitative and qualitative data analysis are presented for a detailed examination of the clinical utility of AI and its explanation.
Open AccessEvaluating the Clinical Utility of Artificial Intelligence Assistance and its Explanation on Glioma Grading Task
Background: As a fast-advancing technology, artificial intelligence (AI) has considerable potential to assist physicians in various clinical tasks from disease identification to lesion segmentation. Despite much research, AI has not yet been applied to neurooncological imaging in a clinically meaningful way. To bridge the clinical implementation gap of AI in neuro-oncological settings, we conducted a clinical user-based evaluation, analogous to the phase II clinical trial, to evaluate the utility of AI for diagnostic predictions and the value of AI explanations on the glioma grading task. Method: Using the publicly-available BraTS dataset, we trained an AI model of 88.0% accuracy on the glioma grading task. We selected the SmoothGrad explainable AI Weina Jin and Mostafa Fatehi are co-first authors. algorithm based on the computational evaluation regarding explanation truthfulness among a candidate of 16 commonly-used algorithms. SmoothGrad could explain the AI model’s prediction using a heatmap overlaid on the MRI to highlight important regions for AI prediction. The evaluation is an online survey wherein the AI prediction and explanation are embedded. Each of the 35 neurosurgeon participants read 25 brain MRI scans of patients with gliomas, and gave their judgment on the glioma grading without and with the assistance of AI’s prediction and explanation. Result: Compared to the average accuracy of 82.5 ± 8.7% when physicians perform the task alone, physicians’ task performance increased to 87.7 ± 7.3% with statistical significance ( p -value = 0.002) when assisted by AI prediction, and remained at almost the same level of 88.5 ± 7.0% ( p -value = 0.35) with the additional AI explanation assistance. Conclusion: The evaluation shows the clinical utility of AI to assist physicians on the glioma grading task. It also reveals the limitations of applying existing AI explanation techniques in clinical settings. Key points: Phase II evaluation with 35 neurosurgeons on the clinical utility of AI and its explanation AI prediction assistance improved physicians’ performance on the glioma grading task Additional AI explanation assistance did not yield a performance boost Importance of the study: This study is the first phase II AI clinical evaluation in neuro-oncology. Evaluating AI is a prerequisite for its clinical deployment. The four phases of AI clinical evaluation are analogous to the four phases of clinical trials. Prior works that apply AI in neurooncology utilize phase I algorithmic evaluation, which do not reflect how AI can be used in clinical settings to support physician decision making. To bridge the research gap, we conducted the first clinical evaluation to assess the joint neurosurgeon-AI task performance. The evaluation also includes AI explanation as an indispensable feature for AI clinical deployment. Results from quantitative and qualitative data analysis are presented for a detailed examination of the clinical utility of AI and its explanation.
- Top Cited Papers
- 17 November 2022
- Published by Elsevier in Medical Image Analysis
- Vol. 84, 102684
Guidelines and evaluation of clinical explainable AI in medical image analysis
- 17 November 2022
- Published by Elsevier in Medical Image Analysis
- Vol. 84, 102684
Top Cited PapersGuidelines and evaluation of clinical explainable AI in medical image analysis
- Open Access
- 23 August 2022
- Published by Springer Nature in Cureus
- Vol. 14 (8) , e28301
Intraparenchymal Mucosa-Associated Lymphoid Tissue Lymphoma: A Case Report
Marginal zone B-cell lymphoma (MZBCL) of mucosa-associated lymphoid tissue (MALT) type, which is primary to the central nervous system (CNS), is a rare lesion, with those originating within the parenchyma even more so. We present the case of a 64-year-old male with weakness in the left hand and focal motor seizures of his arm, who was found to have a right frontal intraparenchymal lesion. Following resection, histopathological and immunohistochemical evaluations were completed, leading to a diagnosis of a primary CNS MZBCL of MALT type in the context of a negative workup of systemic disease. Neuroimaging, histopathological, and immunohistochemical findings, as well as a comprehensive literature review of similar cases, are discussed.
- 23 August 2022
- Published by Springer Nature in Cureus
- Vol. 14 (8) , e28301
Open AccessIntraparenchymal Mucosa-Associated Lymphoid Tissue Lymphoma: A Case Report
Marginal zone B-cell lymphoma (MZBCL) of mucosa-associated lymphoid tissue (MALT) type, which is primary to the central nervous system (CNS), is a rare lesion, with those originating within the parenchyma even more so. We present the case of a 64-year-old male with weakness in the left hand and focal motor seizures of his arm, who was found to have a right frontal intraparenchymal lesion. Following resection, histopathological and immunohistochemical evaluations were completed, leading to a diagnosis of a primary CNS MZBCL of MALT type in the context of a negative workup of systemic disease. Neuroimaging, histopathological, and immunohistochemical findings, as well as a comprehensive literature review of similar cases, are discussed.
- 5 May 2022
- Published by Cambridge University Press (CUP) in Canadian Journal of Neurological Sciences / Journal Canadien des Sciences Neurologiques
- Vol. 50 (2) , 278-281
Current Trends in Neurosurgical Management of Adult Diffuse Low-Grade Gliomas in Canada
There is considerable variability in the management of diffuse low-grade gliomas (LGGs). To characterize treatment paradigms, a survey of Canadian neurosurgeons was performed with forty neurosurgeons responding. Their responses show that the management of patients with LGGs has evolved in the past decade and findings from the RTOG9802 trial have been integrated into the practice of Canadian neurosurgeons. Most respondents stated that the patient selection and treatment strategy advocated by the RTOG9802 trial needs further evaluation. Overall, there is a trend toward more aggressive surgical resections, and future investigations will have to more accurately stratify patient risk profiles.
- 5 May 2022
- Published by Cambridge University Press (CUP) in Canadian Journal of Neurological Sciences / Journal Canadien des Sciences Neurologiques
- Vol. 50 (2) , 278-281
Current Trends in Neurosurgical Management of Adult Diffuse Low-Grade Gliomas in Canada
There is considerable variability in the management of diffuse low-grade gliomas (LGGs). To characterize treatment paradigms, a survey of Canadian neurosurgeons was performed with forty neurosurgeons responding. Their responses show that the management of patients with LGGs has evolved in the past decade and findings from the RTOG9802 trial have been integrated into the practice of Canadian neurosurgeons. Most respondents stated that the patient selection and treatment strategy advocated by the RTOG9802 trial needs further evaluation. Overall, there is a trend toward more aggressive surgical resections, and future investigations will have to more accurately stratify patient risk profiles.
- Open Access
- 11 April 2022
- Published by Cambridge University Press (CUP) in Canadian Journal of Neurological Sciences / Journal Canadien des Sciences Neurologiques
- Vol. 50 (2) , 301-302
The Inception of the Canadian Medical Student Interest Group in Neurosurgery (CaMSIGN): A Student-Led Platform Dedicated to Neurosurgical Education, Research, Mentorship, and Advocacy
The Inception of the Canadian Medical Student Interest Group in Neurosurgery (CaMSIGN): A Student-Led Platform Dedicated to Neurosurgical Education, Research, Mentorship, and Advocacy
- 11 April 2022
- Published by Cambridge University Press (CUP) in Canadian Journal of Neurological Sciences / Journal Canadien des Sciences Neurologiques
- Vol. 50 (2) , 301-302
Open AccessThe Inception of the Canadian Medical Student Interest Group in Neurosurgery (CaMSIGN): A Student-Led Platform Dedicated to Neurosurgical Education, Research, Mentorship, and Advocacy
The Inception of the Canadian Medical Student Interest Group in Neurosurgery (CaMSIGN): A Student-Led Platform Dedicated to Neurosurgical Education, Research, Mentorship, and Advocacy