Hakim Baazaoui
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Hakim Baazaoui has a total of 145 co-authors. Here are the top co-authors by number of shared publications:
- Susanne Wegener
Department of Neurology, University Hospital and University of Zurich, Zurich, Switzerland
3 shared publications - Benedikt Wiestler
Technical University of Munich
2 shared publications - Bjoern Menze
Department of Quantitative Biomedicine, University of Zurich, Zurich, Switzerland
2 shared publications - Daniel Rueckert
Department of Computing, Imperial College London, London, United Kingdom
2 shared publications - Evamaria Olga Riedel
Department of Neuroradiology, School of Medicine and Health, Technical University of Munich, Munich, Germany
2 shared publications - Ezequiel de la Rosa
Department of Quantitative Biomedicine, University of Zurich, Zurich, Switzerland
2 shared publications
Latest Publications
- 1 November 2026
- Published by Wolters Kluwer Health in Stroke
- Vol. 57 (11) , 3441-3450
Responsible AI for Acute Stroke Management
Artificial intelligence is increasingly integrated into acute stroke management, demonstrating strong performance in tasks such as large vessel occlusion detection, Alberta Stroke Program Early Computed Tomography Score scoring, and functional outcome prediction. However, concerns persist regarding both the black-box nature of these models and algorithmic bias that may exacerbate disparities in stroke incidence, treatment, and outcomes across racial and socioeconomic subgroups. This review synthesizes the current literature on explainable artificial intelligence and fairness in artificial intelligence applications for acute stroke management, identifies persistent challenges, and outlines recommendations for the development of equitable and trustworthy systems in stroke care. Recent studies have increasingly adopted post hoc explainability methods, though these are limited by approximation and misleading interpretations, especially since explanations are rarely formally tested. Explainability and fairness remain largely disconnected, with fairness evaluation remaining uncommon due to limited demographic metadata, regulatory constraints, and the absence of stroke-specific fairness criteria. Generalizability also remains a concern due to suboptimal data set partitioning strategies and inadequate reporting practices. Responsible artificial intelligence for acute stroke management requires unified evaluation frameworks that jointly assess explainability, fairness, and generalizability.
- 1 November 2026
- Published by Wolters Kluwer Health in Stroke
- Vol. 57 (11) , 3441-3450
Responsible AI for Acute Stroke Management
Artificial intelligence is increasingly integrated into acute stroke management, demonstrating strong performance in tasks such as large vessel occlusion detection, Alberta Stroke Program Early Computed Tomography Score scoring, and functional outcome prediction. However, concerns persist regarding both the black-box nature of these models and algorithmic bias that may exacerbate disparities in stroke incidence, treatment, and outcomes across racial and socioeconomic subgroups. This review synthesizes the current literature on explainable artificial intelligence and fairness in artificial intelligence applications for acute stroke management, identifies persistent challenges, and outlines recommendations for the development of equitable and trustworthy systems in stroke care. Recent studies have increasingly adopted post hoc explainability methods, though these are limited by approximation and misleading interpretations, especially since explanations are rarely formally tested. Explainability and fairness remain largely disconnected, with fairness evaluation remaining uncommon due to limited demographic metadata, regulatory constraints, and the absence of stroke-specific fairness criteria. Generalizability also remains a concern due to suboptimal data set partitioning strategies and inadequate reporting practices. Responsible artificial intelligence for acute stroke management requires unified evaluation frameworks that jointly assess explainability, fairness, and generalizability.
- 23 July 2026
- Published by Massachusetts Medical Society in NEJM AI
- Vol. 3 (8)
The TopCoW Challenge — Topology-Aware Circle of Willis Segmentation for CT and MR Angiography
- 23 July 2026
- Published by Massachusetts Medical Society in NEJM AI
- Vol. 3 (8)
The TopCoW Challenge — Topology-Aware Circle of Willis Segmentation for CT and MR Angiography
- Open Access
- 23 July 2026
- Published by BMJ in BMJ Case Reports
- Vol. 19 (7) , e272451
Recurrent embolic cerebellar infarction due to bow hunter’s syndrome of the non-dominant vertebral artery
We describe a rare presentation of bow hunter’s syndrome (BHS) involving the non-dominant vertebral artery (VA), leading to recurrent embolic cerebellar infarction. A male in his mid-70s presented with sudden-onset immobilising vertigo. MRI revealed acute cerebellar ischaemia with occlusion of the left superior cerebellar artery. Dynamic duplex ultrasonography and angiography revealed an occlusion of the left VA in the V2 segment during head rotation to the left due to an osteophyte protruding from the C3/C4 vertebrae. The patient was diagnosed with embolic strokes due to compression of the left, non-dominant VA. He was treated with osteophyte resection and anterior cervical discectomy and fusion. BHS is the medical syndrome associated with the compression of the VA induced by movement or rotation of the head, resulting in symptoms of vertebrobasilar insufficiency or stroke. As a rare cause of stroke, it can be diagnosed through dynamic duplex ultrasound and angiography with head rotation. This case is an important addition to the literature because it highlights three key learning points: (1) BHS is a frequently underdiagnosed and underreported cause of transient ischaemic attack, stroke, disability and death; (2) early diagnosis and intervention can prevent devastating outcomes; (3) BHS can cause embolic stroke even from the non-dominant VA.
- 23 July 2026
- Published by BMJ in BMJ Case Reports
- Vol. 19 (7) , e272451
Open AccessRecurrent embolic cerebellar infarction due to bow hunter’s syndrome of the non-dominant vertebral artery
We describe a rare presentation of bow hunter’s syndrome (BHS) involving the non-dominant vertebral artery (VA), leading to recurrent embolic cerebellar infarction. A male in his mid-70s presented with sudden-onset immobilising vertigo. MRI revealed acute cerebellar ischaemia with occlusion of the left superior cerebellar artery. Dynamic duplex ultrasonography and angiography revealed an occlusion of the left VA in the V2 segment during head rotation to the left due to an osteophyte protruding from the C3/C4 vertebrae. The patient was diagnosed with embolic strokes due to compression of the left, non-dominant VA. He was treated with osteophyte resection and anterior cervical discectomy and fusion. BHS is the medical syndrome associated with the compression of the VA induced by movement or rotation of the head, resulting in symptoms of vertebrobasilar insufficiency or stroke. As a rare cause of stroke, it can be diagnosed through dynamic duplex ultrasound and angiography with head rotation. This case is an important addition to the literature because it highlights three key learning points: (1) BHS is a frequently underdiagnosed and underreported cause of transient ischaemic attack, stroke, disability and death; (2) early diagnosis and intervention can prevent devastating outcomes; (3) BHS can cause embolic stroke even from the non-dominant VA.
- 1 May 2026
- Published by Radiological Society of North America (RSNA) in Radiology: Artificial Intelligence
- Vol. 8 (3) , e250603
The Ischemic Stroke Lesion Segmentation Challenge (ISLES)’24 Dataset: A Multimodal Stroke Imaging Dataset with Hyperacute CT, Acute Postinterventional MRI, and 3-month Clinical Outcomes
A multicenter longitudinal stroke dataset with hyperacute CT, acute MRI 2–9 days after reperfusion, and 3-month clinical outcomes is presented to support machine learning models for lesion identification and outcome prediction.
- 1 May 2026
- Published by Radiological Society of North America (RSNA) in Radiology: Artificial Intelligence
- Vol. 8 (3) , e250603
The Ischemic Stroke Lesion Segmentation Challenge (ISLES)’24 Dataset: A Multimodal Stroke Imaging Dataset with Hyperacute CT, Acute Postinterventional MRI, and 3-month Clinical Outcomes
A multicenter longitudinal stroke dataset with hyperacute CT, acute MRI 2–9 days after reperfusion, and 3-month clinical outcomes is presented to support machine learning models for lesion identification and outcome prediction.
- 26 March 2026
- Published by American Society of Neuroradiology (ASNR) in American Journal of Neuroradiology
- Vol. 47 (11) , 2726-2732
Smokers with Large Vessel Occlusion Stroke Show Poor Leptomeningeal Collaterals
In ischemic stroke, the collateral circulation influences tissue survival, but the factors that affect collateral function remain understudied. Smoking is a major vascular risk factor, yet some studies have suggested a "smokers' paradox", with increased resilience of the brain's vascular network and better outcomes in patients with a positive smoking history. Here, we aimed to assess associations between smoking and leptomeningeal collateral status using DSA, the reference standard for collateral evaluation. We retrospectively analysed data from 921 patients with confirmed vessel occlusion from our multicentre stroke image and clinical data repository. Leptomeningeal collaterals were graded based on DSA. Smoking status was classified as never versus former or current smoker. Univariate and multivariate analyses were used to investigate the associations between smoking and collateral status, adjusting for relevant clinical features. 357 patients were included in the analysis (53% women), 19% were current or former smokers. Stroke patients with a positive smoking history were significantly younger and less likely to have atrial fibrillation. In unadjusted and adjusted logistic regression analysis, smokers had significantly lower odds of having a good collateral status on DSA with an unadjusted odds ratio of 0.43 (95% CI: 0.24 - 0.75) and an adjusted odds ratio of 0.42 (95% CI: 0.23 - 0.76). DSA-based assessment demonstrated a significant association between positive smoking history and worse collateral status. These findings contradict prior CTA-based studies and the "smokers' paradox" hypothesis.
- 26 March 2026
- Published by American Society of Neuroradiology (ASNR) in American Journal of Neuroradiology
- Vol. 47 (11) , 2726-2732
Smokers with Large Vessel Occlusion Stroke Show Poor Leptomeningeal Collaterals
In ischemic stroke, the collateral circulation influences tissue survival, but the factors that affect collateral function remain understudied. Smoking is a major vascular risk factor, yet some studies have suggested a "smokers' paradox", with increased resilience of the brain's vascular network and better outcomes in patients with a positive smoking history. Here, we aimed to assess associations between smoking and leptomeningeal collateral status using DSA, the reference standard for collateral evaluation. We retrospectively analysed data from 921 patients with confirmed vessel occlusion from our multicentre stroke image and clinical data repository. Leptomeningeal collaterals were graded based on DSA. Smoking status was classified as never versus former or current smoker. Univariate and multivariate analyses were used to investigate the associations between smoking and collateral status, adjusting for relevant clinical features. 357 patients were included in the analysis (53% women), 19% were current or former smokers. Stroke patients with a positive smoking history were significantly younger and less likely to have atrial fibrillation. In unadjusted and adjusted logistic regression analysis, smokers had significantly lower odds of having a good collateral status on DSA with an unadjusted odds ratio of 0.43 (95% CI: 0.24 - 0.75) and an adjusted odds ratio of 0.42 (95% CI: 0.23 - 0.76). DSA-based assessment demonstrated a significant association between positive smoking history and worse collateral status. These findings contradict prior CTA-based studies and the "smokers' paradox" hypothesis.