Mustafa Selim Sahin

Dr. Mustafa Selim Sahin

orcid

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
18 March 2026 in World Neurosurgery
Verified
Journal Article
Open access
DBDuygu Dolen burak
Duygu Dolen burak
MSMustafa Selim Sahin
Mustafa Selim Sahin
MEMerve Erguven
Merve Erguven
DSDuran Sahin
Duran Sahin
CGCafer Ikbal Gulsever
Cafer Ikbal Gulsever
OBOrhan Barlas
Orhan Barlas
Diagnosis and prognostication in idiopathic normal pressure hydrocephalus (iNPH) remain challenging. This study examined whether large language models (LLMs) can classify neuropsychological test (NPT) summaries for identifying iNPH and exploring postoperative shunt responsiveness. This retrospective study included 42 shunt-treated iNPH patients and 53 neurological controls. De-identified NPT summaries with a standardized structure across cognitive domains were evaluated using three LLMs (ChatGPT-5, Gemini 2.5 Flash, DeepSeek) under a standardized zero-shot prompt. Each model generated categorical outputs (iNPH-compatible) and confidence scores (0-100). For analysis, responses labeled "Compatible" were considered AI-positive, whereas "Incompatible" or "Indeterminate" responses were treated as AI-negative; outputs with low confidence were conservatively handled as negative classifications. Diagnostic performance was compared with established clinical diagnoses. Prognostic performance was assessed for postoperative improvement, defined as ΔiNPHGS ≥1. Diagnostic performance showed the highest accuracy for ChatGPT-5 (78%, 95% CI: 68.6-85.1), followed by Gemini 2.5 Flash (67%, 95% CI: 57.4-75.9) and DeepSeek (63%, 95% CI: 53.1-72.2). Corresponding AUC values were 0.84, 0.67, and 0.63. Calibration metrics for ChatGPT-5 included a Brier score of 0.18 and a calibration slope of 0.94. In the exploratory prognostic analysis conducted in the shunted iNPH cohort (n = 42), postoperative improvement occurred in 37 patients, whereas five patients were classified as non-responders. Prognostic accuracy was 83% (95% CI: 69.8-92.5) for ChatGPT-5, 67% (95% CI: 52.1-78.8) for DeepSeek, and 64% (95% CI: 49.8-76.9) for Gemini 2.5 Flash. Positive predictive values ranged from 87% to 96%. When model confidence was analyzed as a continuous predictor, AUC values for prognostic discrimination were 0.69 for DeepSeek, 0.56 for ChatGPT-5, and 0.53 for Gemini 2.5 Flash. Decision-curve analysis indicated varying net benefit patterns across threshold probabilities. LLMs applied to standardized NPT summaries demonstrated measurable diagnostic signal and exploratory prognostic potential in iNPH. These findings support the feasibility of NPT-based LLM assistance, but larger, balanced, externally validated cohorts are required before clinical implementation.

20 August 2026 in World Neurosurgery
Verified
Journal Article
Open access
DBDuygu Dolen burak
Duygu Dolen burak
MSMustafa Selim Sahin
Mustafa Selim Sahin
MEMerve Erguven
Merve Erguven
DSDuran Sahin
Duran Sahin
CGCafer Ikbal Gulsever
Cafer Ikbal Gulsever
OBOrhan Barlas
Orhan Barlas

31 August 2026 in Journal of Clinical Medicine
Verified
Journal Article
Open access
PDF file
DBDuygu Dolen burak
Duygu Dolen burak
MSMustafa Selim Sahin
Mustafa Selim Sahin
CGCafer Ikbal Gulsever
Cafer Ikbal Gulsever
DSDuran Sahin
Duran Sahin
TUTugrul Cem Unal
Tugrul Cem Unal
ASAltay Sencer
Altay Sencer
Background: The optimal sequencing of cranioplasty (CP) and ventriculoperitoneal shunt (VPS) placement after decompressive craniectomy (DC) remains uncertain, particularly in pediatric patients. This study aimed to describe our pediatric experience and provide an exploratory comparison of simultaneous and staged procedures. Methods: We retrospectively reviewed the medical records of 12 pediatric patients who developed hydrocephalus after DC at a single tertiary center between 2015 and 2023. Six patients underwent simultaneous CP and VPS placement, while six underwent staged VPS placement followed by CP. Demographic and clinical characteristics, surgical timing, postoperative complications, and neurological outcomes were compared. Given the small sample size, analyses were exploratory and primarily descriptive. Results: Postoperative complications occurred in two patients in each group (33.3%), including one wound infection and one shunt dysfunction requiring revision. No intracranial hemorrhage or deep shunt infection occurred in either group. The complication rate did not differ between the groups (p = 1.000). The median final Glasgow Outcome Scale score was 2.5 (range, 2–3) in the simultaneous group and 3.0 (range, 1–5) in the staged group (p = 0.457). Neurological outcomes likely reflected the severity of the underlying brain injury; however, the small sample size precluded adjustment for prognostic factors. Conclusion: In this small retrospective pediatric cohort, simultaneous CP and VPS placement was technically feasible, but postoperative complications occurred in one-third of patients. No significant differences were observed between the simultaneous and staged groups; however, these findings should not be interpreted as evidence of equivalence because of the limited sample size and non-randomized design. Surgical strategy should be individualized, and larger multicenter pediatric studies are needed.

9 January 2025 in Surgical Innovation
Verified
Journal Article
Closed access
MSMustafa Selim Sahin
Mustafa Selim Sahin
CGCafer Ikbal Gulsever
Cafer Ikbal Gulsever
MOMetehan Ozturk
Metehan Ozturk
DDDuygu Dolen
Duygu Dolen
IDIlyas Dolas
Ilyas Dolas
PSPulat Akin Sabanci
Pulat Akin Sabanci
Objective: The endoscopic transsphenoidal approach is commonly used for sellar and suprasellar pathologies. However, reaching above the diaphragma sella, especially for posterosuperior and retrocavernous orientation, still poses some challenges. We designed and developed a steerable tip suction cannula (STSC) that has distinct leverage for endoscopic resection of such pathologies. Methods: The entire suction cannula is made of stainless steel. The instrument consists of a handle, vacuum tube, suction tip bed, suction tip, finger knuckle, wire path, countersunk headpin, and isolated steel wire. The working principle of the product is to enable the surgeon to move the aspiration tip in the desired direction by steering the finger knuckle. Five cadaveric specimens fixed with the saturated salt solution were used to evaluate the instrument and obtain measurements. Results: In the straight position, the STSC aspirated 2.67% slower than the control aspirator tip and 13.24% faster than that at 30° angulation. Based on the CT measurements, the mean angulation of the instrument from the frontobasal axis was 38.3°. The mean distance from the frontobasal axis was 1.3 cm. The average angulation of the tip of the instrument in the cranium was 25.5°. Conclusion: The designed STSC might effectively resect large to giant pituitary adenomas, especially those with supradiaphragmatic extension. Its suction capability is comparable to that of conventional suction tubes.

4 September 2026 in World Neurosurgery
Verified
Journal Article
Open access
DBDuygu Dolen burak
Duygu Dolen burak
MSMustafa Selim Sahin
Mustafa Selim Sahin
MEMerve Erguven
Merve Erguven
DSDuran Sahin
Duran Sahin
CGCafer Ikbal Gulsever
Cafer Ikbal Gulsever
OBOrhan Barlas
Orhan Barlas

25 August 2025 in Operative Neurosurgery
Verified
Journal Article
Closed access
AAAydin Aydoseli
Aydin Aydoseli
CGCafer Ikbal Gulsever
Cafer Ikbal Gulsever
DSDuran Sahin
Duran Sahin
IUIrem Yaren Uysal
Irem Yaren Uysal
MSMustafa Selim Sahin
Mustafa Selim Sahin
TUTugrul Cem Unal
Tugrul Cem Unal
BACKGROUND AND OBJECTIVES: Effective reconstruction of skull base defects after endonasal endoscopic surgery is essential to prevent complications such as cerebrospinal fluid leakage and infection. Traditional methods may not always provide optimal outcomes for certain defects. This study evaluates the feasibility of using the Amplatzer Septal Occluder (ASO) as an adjunctive tool for skull base reconstruction in a cadaveric model. METHODS: Three fresh cadavers were used to simulate endonasal endoscopic approaches for resecting skull base pathologies. Defects measuring 8 × 8 mm and 10 × 10 mm were created in the sella turcica, tuberculum sellae, planum sphenoidale, and olfactory groove. The ASO was deployed using a standard delivery system, with the distal disc positioned intracranially and the proximal disc in the nasal cavity. High-resolution endoscopic imaging and scopic views were obtained to assess the anatomic fit and functional integrity of the reconstruction. RESULTS: The ASO provided stable coverage for all defect sites without compressing critical neurovascular structures. The device conformed well to the irregular contours of the skull base. Functional testing using barium liquid demonstrated minimal to no leakage into the nasal cavity, suggesting the potential for a watertight seal. Imaging highlighted the ASO's feasibility as a structural component in multilayer reconstruction strategies. CONCLUSION: The ASO demonstrates significant promise as a supplementary tool for skull base reconstruction in endonasal endoscopic surgery. Its incorporation into a multilayer closure strategy could improve defect stability and reduce postoperative complications. Future studies are necessary to explore device customization, bioresorbable materials, and injectable bioactive substances to evaluate outcomes before clinical implementation in humans.