Umut Yilmaz
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Umut Yilmaz has a total of 219 co-authors. Here are the top co-authors by number of shared publications:
Wolfgang Reith
Department of Neuroradiology, Saarland University Hospital
19 shared publications- Wolfgang Reith
Klinik für Diagnostische, und Interventionelle Neuroradiologie, Universitätsklinikum des Saarlandes, Homburg-Saar, Deutschland
13 shared publications Malvina Garner
Department of Neuroradiology, Saarland University Hospital
10 shared publicationsAlena Haußmann
Department of Neuroradiology, Saarland University Hospital
9 shared publicationsFrederik Fries
Department of Neuroradiology, Saarland University Hospital
8 shared publicationsW. Reith
Klinik für Diagnostische und Interventionelle Neuroradiologie, Universitätsklinikum des Saarlandes, Kirrberger Straße, 66424, Homburg-Saar, Deutschland
8 shared publications
Latest Publications
- Open Access
- 9 November 2026
- Published by Frontiers Media SA in Frontiers in Neurology
- Vol. 17, 1973423
Surface-based quantification of temporal-pole gray-white matter blurring in hippocampal sclerosis: an exploratory MRI profiling study
Objective: Temporal-pole gray-white matter blurring and adjacent superficial white-matter signal abnormality are recognized extrahippocampal Magnetic Resonance Imaging (MRI) features of temporal lobe epilepsy with hippocampal sclerosis (HS), but their spatial organization is rarely quantified. We tested whether automated FreeSurfer-anchored surface-normal MRI profiling can quantify distance-resolved anterior temporal gray-white boundary and superficial white-matter abnormalities on standard T1-weighted MRI. Materials and methods: We analyzed a scanner- and sequence-restricted public imaging database for the Epilepsy and Surgery subset comprising 41 patients with HS and 41 matched controls. N4-corrected T1-weighted images were reprocessed locally with FreeSurfer 8.2.0 and underwent focal manual boundary profiling and automated surface-normal sampling from cortex into white matter. Primary endpoints were affected-minus-unaffected relative white-matter contrasts from 1 to 2 mm and 2 to 3 mm at the white-matter ≥1.5-mm gate, standardized to hemisphere-specific controls. Morphology and sampling sensitivity analyses tested whether temporal-pole atrophy or profile geometry explained the primary findings. Results: Automated profiling was completed for all subjects. Controls showed right-dominant anterior temporal asymmetry, supporting hemisphere-specific standardization. In patients, affected-side relative white-matter contrast was reduced at 1–2 mm (mean delta z = −0.581, Bonferroni-adjusted p = 0.001) and 2–3 mm (mean delta z = −0.753, Bonferroni-adjusted p < 0.001). The strongest reduction localized to approximately +1 to +3 mm beyond the gray-white boundary. Morphology deltas did not correlate with Surface-LEAP effects, and primary effects remained significant after adjustment. Exploratory machine learning supported subject-level information but showed no significant classification gain over simple white-matter markers. Conclusion: Automated surface-based profiling provides an interpretable framework for quantifying a visually recognized anterior temporal MRI phenotype in HS. These exploratory findings should be interpreted as HS-versus-healthy-control abnormalities rather than disease-specific markers; disease-control and external validation are needed.
- 9 November 2026
- Published by Frontiers Media SA in Frontiers in Neurology
- Vol. 17, 1973423
Open AccessSurface-based quantification of temporal-pole gray-white matter blurring in hippocampal sclerosis: an exploratory MRI profiling study
Objective: Temporal-pole gray-white matter blurring and adjacent superficial white-matter signal abnormality are recognized extrahippocampal Magnetic Resonance Imaging (MRI) features of temporal lobe epilepsy with hippocampal sclerosis (HS), but their spatial organization is rarely quantified. We tested whether automated FreeSurfer-anchored surface-normal MRI profiling can quantify distance-resolved anterior temporal gray-white boundary and superficial white-matter abnormalities on standard T1-weighted MRI. Materials and methods: We analyzed a scanner- and sequence-restricted public imaging database for the Epilepsy and Surgery subset comprising 41 patients with HS and 41 matched controls. N4-corrected T1-weighted images were reprocessed locally with FreeSurfer 8.2.0 and underwent focal manual boundary profiling and automated surface-normal sampling from cortex into white matter. Primary endpoints were affected-minus-unaffected relative white-matter contrasts from 1 to 2 mm and 2 to 3 mm at the white-matter ≥1.5-mm gate, standardized to hemisphere-specific controls. Morphology and sampling sensitivity analyses tested whether temporal-pole atrophy or profile geometry explained the primary findings. Results: Automated profiling was completed for all subjects. Controls showed right-dominant anterior temporal asymmetry, supporting hemisphere-specific standardization. In patients, affected-side relative white-matter contrast was reduced at 1–2 mm (mean delta z = −0.581, Bonferroni-adjusted p = 0.001) and 2–3 mm (mean delta z = −0.753, Bonferroni-adjusted p < 0.001). The strongest reduction localized to approximately +1 to +3 mm beyond the gray-white boundary. Morphology deltas did not correlate with Surface-LEAP effects, and primary effects remained significant after adjustment. Exploratory machine learning supported subject-level information but showed no significant classification gain over simple white-matter markers. Conclusion: Automated surface-based profiling provides an interpretable framework for quantifying a visually recognized anterior temporal MRI phenotype in HS. These exploratory findings should be interpreted as HS-versus-healthy-control abnormalities rather than disease-specific markers; disease-control and external validation are needed.
- Open Access
- 5 November 2026
- Published by Elsevier in World Neurosurgery
Longitudinal boundary-anchored CT profiling of the frontal horns after cerebrospinal fluid shunting in idiopathic intracranial hypertension: a pilot study
- 5 November 2026
- Published by Elsevier in World Neurosurgery
Open AccessLongitudinal boundary-anchored CT profiling of the frontal horns after cerebrospinal fluid shunting in idiopathic intracranial hypertension: a pilot study
- 22 October 2026
- Published by Georg Thieme Verlag KG in RöFo - Fortschritte auf dem Gebiet der Röntgenstrahlen und der bildgebenden Verfahren
Spatially resolved periventricular ADC profiling in normal pressure hydrocephalus: an exploratory pilot study
To explore whether boundary-anchored apparent diffusion coefficient (ADC) profiling can characterize periventricular diffusion patterns in normal pressure hydrocephalus (NPH). Seventeen patients with shunted NPH with baseline and postoperative follow-up ADC maps acquired at 1.5 T or 3 T and 17 age-, sex-, and scanner-matched controls were analyzed retrospectively. A short anchor was placed along the frontal horn ventricular boundary. ADC values were sampled along approximately orthogonal lines and summarized as amplitude- and shape-related profile metrics. Patient-control and longitudinal within-patient comparisons were analyzed with false-discovery-rate correction for multiple comparisons. At baseline, patients showed a higher reference ADC, inner-zone mean ADC, outer-zone mean ADC, and outer plateau ADC than controls. At follow-up, amplitude-related ADC differences persisted. Longitudinally, the profile multiphasicity index, inner peak ADC, and initial profile slope changed after shunting and remained significant after FDR correction, whereas most amplitude-related metrics remained stable. Boundary-anchored ADC profiling identified persistent periventricular ADC differences and selected postoperative profile changes in NPH. The findings are exploratory and require validation.
- 22 October 2026
- Published by Georg Thieme Verlag KG in RöFo - Fortschritte auf dem Gebiet der Röntgenstrahlen und der bildgebenden Verfahren
Spatially resolved periventricular ADC profiling in normal pressure hydrocephalus: an exploratory pilot study
To explore whether boundary-anchored apparent diffusion coefficient (ADC) profiling can characterize periventricular diffusion patterns in normal pressure hydrocephalus (NPH). Seventeen patients with shunted NPH with baseline and postoperative follow-up ADC maps acquired at 1.5 T or 3 T and 17 age-, sex-, and scanner-matched controls were analyzed retrospectively. A short anchor was placed along the frontal horn ventricular boundary. ADC values were sampled along approximately orthogonal lines and summarized as amplitude- and shape-related profile metrics. Patient-control and longitudinal within-patient comparisons were analyzed with false-discovery-rate correction for multiple comparisons. At baseline, patients showed a higher reference ADC, inner-zone mean ADC, outer-zone mean ADC, and outer plateau ADC than controls. At follow-up, amplitude-related ADC differences persisted. Longitudinally, the profile multiphasicity index, inner peak ADC, and initial profile slope changed after shunting and remained significant after FDR correction, whereas most amplitude-related metrics remained stable. Boundary-anchored ADC profiling identified persistent periventricular ADC differences and selected postoperative profile changes in NPH. The findings are exploratory and require validation.
- 21 September 2026
- Published by Georg Thieme Verlag KG in RöFo - Fortschritte auf dem Gebiet der Röntgenstrahlen und der bildgebenden Verfahren
Boundary-anchored thalamic ADC profiling shows spatially structured abnormalities in multiple sclerosis
Diffusion MRI is sensitive to tissue abnormalities in multiple sclerosis (MS), but conventional region-of-interest measures reduce spatially structured signal patterns to a single summary value. We evaluated boundary-anchored thalamic apparent diffusion coefficient (ADC) profiling in MS using line-based edge-anchored profiling (LEAP). In this retrospective matched-control study, age- and sex-matched controls (CTRL, n = 28) and participants with relapsing-remitting MS (RRMS, n = 14) or secondary progressive MS (SPMS, n = 14) were analyzed on the same 3T MRI scanner using the same diffusion protocol. LEAP was applied bilaterally at the medial thalamic surface adjacent to the third ventricle. Participant-averaged bilateral summary metrics and exploratory side-specific profile-shape metrics were derived from ADC curves. The clearest signal was observed in comparisons between MS and matched controls, with the largest number of differences seen for SPMS versus controls. Compared with controls, SPMS showed a higher ventricle-near mean ADC (p = 0.004), higher profile maximum ADC (p = 0.003), and a steeper early profile slope (p = 0.005). RRMS generally showed intermediate values. Direct RRMS-versus-SPMS comparisons of bilateral summary metrics were limited. Exploratory right-sided profile-shape metrics and left-right asymmetry measures suggested possible phenotype-related differences. Boundary-anchored thalamic ADC profiling detected spatially structured abnormalities in MS relative to matched controls, with the strongest deviation in SPMS. Exploratory shape-based findings should be considered hypothesis-generating and suggest additional profile-level information beyond bilateral summary values.
- 21 September 2026
- Published by Georg Thieme Verlag KG in RöFo - Fortschritte auf dem Gebiet der Röntgenstrahlen und der bildgebenden Verfahren
Boundary-anchored thalamic ADC profiling shows spatially structured abnormalities in multiple sclerosis
Diffusion MRI is sensitive to tissue abnormalities in multiple sclerosis (MS), but conventional region-of-interest measures reduce spatially structured signal patterns to a single summary value. We evaluated boundary-anchored thalamic apparent diffusion coefficient (ADC) profiling in MS using line-based edge-anchored profiling (LEAP). In this retrospective matched-control study, age- and sex-matched controls (CTRL, n = 28) and participants with relapsing-remitting MS (RRMS, n = 14) or secondary progressive MS (SPMS, n = 14) were analyzed on the same 3T MRI scanner using the same diffusion protocol. LEAP was applied bilaterally at the medial thalamic surface adjacent to the third ventricle. Participant-averaged bilateral summary metrics and exploratory side-specific profile-shape metrics were derived from ADC curves. The clearest signal was observed in comparisons between MS and matched controls, with the largest number of differences seen for SPMS versus controls. Compared with controls, SPMS showed a higher ventricle-near mean ADC (p = 0.004), higher profile maximum ADC (p = 0.003), and a steeper early profile slope (p = 0.005). RRMS generally showed intermediate values. Direct RRMS-versus-SPMS comparisons of bilateral summary metrics were limited. Exploratory right-sided profile-shape metrics and left-right asymmetry measures suggested possible phenotype-related differences. Boundary-anchored thalamic ADC profiling detected spatially structured abnormalities in MS relative to matched controls, with the strongest deviation in SPMS. Exploratory shape-based findings should be considered hypothesis-generating and suggest additional profile-level information beyond bilateral summary values.
- 9 September 2026
- Published by Springer Nature in Die Radiologie
- Vol. 66 (11) , 842-849
Intramedulläre Tumoren des Rückenmarks
- 9 September 2026
- Published by Springer Nature in Die Radiologie
- Vol. 66 (11) , 842-849
Intramedulläre Tumoren des Rückenmarks
- 11 August 2026
- Published by Wolters Kluwer Health in Neurology
- Vol. 107 (3) , e218270
Functional Outcomes and Safety of Endovascular Thrombectomy in Pediatric Acute Ischemic Stroke With Cardiac Abnormalities
Background and ObjectivesCardiac abnormalities are among the most common and complex risk factors of pediatric acute ischemic stroke (AIS). Although endovascular treatment (EVT) is increasingly use...
- 11 August 2026
- Published by Wolters Kluwer Health in Neurology
- Vol. 107 (3) , e218270
Functional Outcomes and Safety of Endovascular Thrombectomy in Pediatric Acute Ischemic Stroke With Cardiac Abnormalities
Background and ObjectivesCardiac abnormalities are among the most common and complex risk factors of pediatric acute ischemic stroke (AIS). Although endovascular treatment (EVT) is increasingly use...
- Open Access
- 21 July 2026
- Published by Springer Nature in Neurosurgical Review
- Vol. 49 (1) , 487
Middle meningeal artery arteriovenous shunting after surgical evacuation of chronic subdural hematoma: a retrospective angiographic cohort study
Arteriovenous shunting involving the middle meningeal artery (MMA) has been reported after cranial trauma, neurosurgical procedures, and endovascular interventions. However, its angiographic frequency and characteristics after chronic subdural hematoma (cSDH) surgery remain poorly defined. This study aimed to assess the frequency and angiographic features of MMA arteriovenous shunting in patients undergoing MMA embolization after surgical evacuation of cSDH. We retrospectively analyzed consecutive patients who underwent selective and superselective MMA angiography during MMA embolization after surgical evacuation of cSDH between January 2020 and March 2023. Angiograms were reviewed for angiographic evidence of arteriovenous shunting, shunt morphology, flow characteristics, venous drainage, relationship to embolization, and angiographic closure after embolization. Surgical technique was classified as burr-hole craniotomy or formal craniotomy. Thirty patients met the inclusion criteria. Angiographic evidence of MMA arteriovenous shunting was identified in 11 patients (36.7%). Shunting was significantly more frequent after craniotomy than after burr-hole craniotomy (90.9% vs. 5.3%, Fisher’s exact test, p < 0.001). Most shunts were detectable before microcatheter advancement, while superselective angiography improved visualization of shunt morphology and venous drainage. All shunts demonstrated low-flow angiographic characteristics without cortical venous reflux. Post-embolization control angiography demonstrated disappearance of the angiographic shunt in all affected patients. Low-flow arteriovenous shunting involving the MMA is a frequent angiographic finding in patients undergoing MMA embolization after cSDH surgery, particularly following craniotomy. These findings provide systematic angiographic characterization of a previously underrecognized postoperative vascular phenomenon. The clinical relevance of such shunting remains uncertain and warrants prospective investigation.
- 21 July 2026
- Published by Springer Nature in Neurosurgical Review
- Vol. 49 (1) , 487
Open AccessMiddle meningeal artery arteriovenous shunting after surgical evacuation of chronic subdural hematoma: a retrospective angiographic cohort study
Arteriovenous shunting involving the middle meningeal artery (MMA) has been reported after cranial trauma, neurosurgical procedures, and endovascular interventions. However, its angiographic frequency and characteristics after chronic subdural hematoma (cSDH) surgery remain poorly defined. This study aimed to assess the frequency and angiographic features of MMA arteriovenous shunting in patients undergoing MMA embolization after surgical evacuation of cSDH. We retrospectively analyzed consecutive patients who underwent selective and superselective MMA angiography during MMA embolization after surgical evacuation of cSDH between January 2020 and March 2023. Angiograms were reviewed for angiographic evidence of arteriovenous shunting, shunt morphology, flow characteristics, venous drainage, relationship to embolization, and angiographic closure after embolization. Surgical technique was classified as burr-hole craniotomy or formal craniotomy. Thirty patients met the inclusion criteria. Angiographic evidence of MMA arteriovenous shunting was identified in 11 patients (36.7%). Shunting was significantly more frequent after craniotomy than after burr-hole craniotomy (90.9% vs. 5.3%, Fisher’s exact test, p < 0.001). Most shunts were detectable before microcatheter advancement, while superselective angiography improved visualization of shunt morphology and venous drainage. All shunts demonstrated low-flow angiographic characteristics without cortical venous reflux. Post-embolization control angiography demonstrated disappearance of the angiographic shunt in all affected patients. Low-flow arteriovenous shunting involving the MMA is a frequent angiographic finding in patients undergoing MMA embolization after cSDH surgery, particularly following craniotomy. These findings provide systematic angiographic characterization of a previously underrecognized postoperative vascular phenomenon. The clinical relevance of such shunting remains uncertain and warrants prospective investigation.
- Open Access
- 23 June 2026
- Published by Elsevier in European Journal of Radiology
- Vol. 203, 113033
Cortical edge-anchored ADC profiling in idiopathic intracranial hypertension: A longitudinal matched-control pilot study
- 23 June 2026
- Published by Elsevier in European Journal of Radiology
- Vol. 203, 113033
Open AccessCortical edge-anchored ADC profiling in idiopathic intracranial hypertension: A longitudinal matched-control pilot study
- Open Access
- 5 June 2026
- Published by Springer Nature in Clinical Neuroradiology
In-Stent Stenosis After Surface-Modified Versus Uncoated Flow Diversion: a Randomized Trial (Derivo 2 Heal vs Derivo 2)
Background and Purpose: Surface-modified flow diverters have been introduced to reduce thrombogenicity and potentially influence vascular healing. However, their impact on in-stent stenosis and aneurysm occlusion remains uncertain. We compared a coated flow diverter (Derivo 2 heal) with the corresponding uncoated device (Derivo 2) regarding in-stent stenosis development and aneurysm occlusion in a randomized clinical setting. Materials and Methods: In this single-center randomized, open-label trial, 21 patients with unruptured internal carotid artery aneurysms were treated with either a coated ( n = 10) or uncoated ( n = 11) flow diverter. All patients received dual antiplatelet therapy and underwent digital subtraction angiography at 3 and 9 months. In-stent stenosis was assessed using a standardized ratio-based method. Device characteristics and oversizing were recorded. The presumed etiology of in-stent stenosis was categorized as intimal hyperplasia or stent deformity based on morphologic appearance. Results: Mean in-stent stenosis at 3 and 9 months did not differ between groups. Complete aneurysm occlusion at 9 months was lower in the coated group (40.0% vs 81.8%; two-sided Fisher’s exact p = 0.080). Oversizing was not associated with in-stent stenosis but was numerically greater in patients with complete occlusion. Stent deformity was associated with higher in-stent stenosis than presumed intimal hyperplasia (3 months: 40.8% vs 21.7%, p = 0.007; 9 months: 30.1% vs 13.3%, p = 0.022). Conclusions: In this randomized single-center study, the surface-modified Derivo 2 heal did not demonstrate reduced in-stent stenosis compared with the uncoated Derivo 2, while complete occlusion at 9 months was lower in the coated-device group. These findings warrant confirmation in larger comparative studies.
- 5 June 2026
- Published by Springer Nature in Clinical Neuroradiology
Open AccessIn-Stent Stenosis After Surface-Modified Versus Uncoated Flow Diversion: a Randomized Trial (Derivo 2 Heal vs Derivo 2)
Background and Purpose: Surface-modified flow diverters have been introduced to reduce thrombogenicity and potentially influence vascular healing. However, their impact on in-stent stenosis and aneurysm occlusion remains uncertain. We compared a coated flow diverter (Derivo 2 heal) with the corresponding uncoated device (Derivo 2) regarding in-stent stenosis development and aneurysm occlusion in a randomized clinical setting. Materials and Methods: In this single-center randomized, open-label trial, 21 patients with unruptured internal carotid artery aneurysms were treated with either a coated ( n = 10) or uncoated ( n = 11) flow diverter. All patients received dual antiplatelet therapy and underwent digital subtraction angiography at 3 and 9 months. In-stent stenosis was assessed using a standardized ratio-based method. Device characteristics and oversizing were recorded. The presumed etiology of in-stent stenosis was categorized as intimal hyperplasia or stent deformity based on morphologic appearance. Results: Mean in-stent stenosis at 3 and 9 months did not differ between groups. Complete aneurysm occlusion at 9 months was lower in the coated group (40.0% vs 81.8%; two-sided Fisher’s exact p = 0.080). Oversizing was not associated with in-stent stenosis but was numerically greater in patients with complete occlusion. Stent deformity was associated with higher in-stent stenosis than presumed intimal hyperplasia (3 months: 40.8% vs 21.7%, p = 0.007; 9 months: 30.1% vs 13.3%, p = 0.022). Conclusions: In this randomized single-center study, the surface-modified Derivo 2 heal did not demonstrate reduced in-stent stenosis compared with the uncoated Derivo 2, while complete occlusion at 9 months was lower in the coated-device group. These findings warrant confirmation in larger comparative studies.
- Open Access
- 21 May 2026
- Published by Elsevier in SSRN Electronic Journal
Atlas-anchored distance-resolved profiling of deep brain nuclei in Parkinson disease: a proof-of-concept MRI study
Background: Atlas-based MRI analysis of deep nuclei in Parkinson disease (PD) often reduces each target to a single mean value and may miss boundary-localized or row-wise heterogeneity effects.Methods: We analyzed public T1-weighted MRI from 20 patients with PD and 20 controls using atlas-anchored line profiling of the red nucleus (RN), an AHEAD-derived nigral region (SN_region), and the subthalamic nucleus (STN). We extracted conventional ROI summaries, distance-resolved mean profiles, boundary-dispersion metrics, and left-right asymmetry features. Primary models adjusted for age, sex, and mask-quality covariates. Because the STN is anatomically small, STN heatmap-bank heterogeneity was recomputed in a target-size-aware boundary-adjacent window; the same strategy was applied to the RN as a consistency check.Results: The pipeline yielded complete bilateral measurements across all subjects and targets. The most robust adjusted finding was higher right nigral-region standard deviation at the boundary in PD (adjusted q = 0.027; Cohen d = 0.87). Conventional ROI means showed broad raw-signal differences, whereas normalized mean profiles showed limited group separation. Target-size-aware analysis identified exploratory STN boundary-heterogeneity and asymmetry candidates (best adjusted q = 0.083), but no STN feature survived broad FDR correction. The corresponding RN reanalysis produced only modest adjusted candidates.Conclusions: Atlas-anchored profiling yielded spatially explicit boundary features beyond conventional ROI means. The strongest adjusted signal localized to the right nigral boundary, while compact-nucleus heatmap-bank analysis provided exploratory, anatomically coherent STN candidates requiring validation.
- 21 May 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessAtlas-anchored distance-resolved profiling of deep brain nuclei in Parkinson disease: a proof-of-concept MRI study
Background: Atlas-based MRI analysis of deep nuclei in Parkinson disease (PD) often reduces each target to a single mean value and may miss boundary-localized or row-wise heterogeneity effects.Methods: We analyzed public T1-weighted MRI from 20 patients with PD and 20 controls using atlas-anchored line profiling of the red nucleus (RN), an AHEAD-derived nigral region (SN_region), and the subthalamic nucleus (STN). We extracted conventional ROI summaries, distance-resolved mean profiles, boundary-dispersion metrics, and left-right asymmetry features. Primary models adjusted for age, sex, and mask-quality covariates. Because the STN is anatomically small, STN heatmap-bank heterogeneity was recomputed in a target-size-aware boundary-adjacent window; the same strategy was applied to the RN as a consistency check.Results: The pipeline yielded complete bilateral measurements across all subjects and targets. The most robust adjusted finding was higher right nigral-region standard deviation at the boundary in PD (adjusted q = 0.027; Cohen d = 0.87). Conventional ROI means showed broad raw-signal differences, whereas normalized mean profiles showed limited group separation. Target-size-aware analysis identified exploratory STN boundary-heterogeneity and asymmetry candidates (best adjusted q = 0.083), but no STN feature survived broad FDR correction. The corresponding RN reanalysis produced only modest adjusted candidates.Conclusions: Atlas-anchored profiling yielded spatially explicit boundary features beyond conventional ROI means. The strongest adjusted signal localized to the right nigral boundary, while compact-nucleus heatmap-bank analysis provided exploratory, anatomically coherent STN candidates requiring validation.