Bilal ERTUĞRUL

Dr. Bilal ERTUĞRUL

orcid

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

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

Representative Publications

View All Publications
3 May 2025 in Artificial Intelligence Review
Verified
Journal Article
Open access
PDF file
Mugahed A. Al-antariMugahed A. Al-antari
Mugahed A. Al-antari
SSSaied Salem
Saied Salem
MRMukhlis Raza
Mukhlis Raza
AEAhmed S. Elbadawy
Ahmed S. Elbadawy
EBErtan Bütün
Ertan Bütün
YGYeonghyeon Gu
Yeonghyeon Gu
Lumbar spinal stenosis (LSS) involves the narrowing of the spinal canal, leading to compression of the spinal cord and nerves in the lower back. Common causes include injuries, degenerative age-related changes, congenital conditions, and tumors, all of which contribute to back pain. Early diagnosis is critical for symptom management, preventing progression, and preserving quality of life. This study systematically reviews AI-based approaches for predicting LSS using MRI axial and sagittal imaging. The review focuses on various AI tasks: detection, segmentation, classification, hybrid approaches, spinal index measurements (SIM), and explainable AI frameworks. The aim is to highlight current knowledge, identify limitations in existing models, and propose future research directions. Following PRISMA guidelines and the PICO method (Population, Intervention, Comparison, Outcome), the review collects data from databases like PubMed, Web of Science, ScienceDirect, and IEEE Xplore (2005–2024). The Rayyan AI tool is used for duplicate removal and screening. The screening process includes an initial review of titles and abstracts, followed by full-text appraisal. The Meta Quality Appraisal Tool (MetaQAT) assesses the quality of selected articles. Of 1323 records, 97 duplicates were removed. After screening, 895 records were excluded, leaving 331 for full-text review. Among these, 184 articles were excluded for lacking AI relevance. Ultimately, 95 key articles (91 technical papers and 4 reviews) were identified for their contributions to AI-based LSS prediction. This review provides a comprehensive analysis of AI techniques in LSS prediction, guiding future research and advancing understanding in areas like explainable AI and large language models (LLMs).

6 September 2025
Verified
Conference Paper
Closed access
MRMukhlis Raza
Mukhlis Raza
JHJamil Hussain
Jamil Hussain
YGYeonghyeon Gu
Yeonghyeon Gu
BEBilal ERTUĞRUL
Bilal ERTUĞRUL
In Bo HanIn Bo Han
In Bo Han
Mugahed A. Al-antariMugahed A. Al-antari
Mugahed A. Al-antari
The automated creation of a medical knowledge graph (MKG) from unstructured clinical data is a significant outcome in healthcare decision support and retrieval-augmented generation (RAG) applications. However, large language models (LLMs) based on knowledge extraction inevitably introduce noise, hallucinations, and factual inconsistencies that compromise graph reliability and clinical utility. Existing refinement approaches require supervised labels or complex neural architectures, limiting their applicability in resource-constrained clinical environments. This study introduces Self-Refining Medical Knowledge Graph (SR-MKG), a novel two-stage framework that autonomously enhances LLM-generated medical knowledge graph (KG) quality through unsupervised topological analysis. The methodology comprises: (1) The first stage utilizes an LLM for high-recall triplet extraction from lumbar spine unstructured MRI radiological reports, and (2) an unsupervised self-refinement module implementing topological confidence scoring for each relationship based on graph structural support patterns. The SR-MKG framework evaluated with state-of-the-art LLMs Gemma3 and LLaVA, demonstrates substantial improvements in knowledge graph refinement across two thresholds (0.3 and 0.5). For Gemma3, the initial 1,524 triplets were refined to 843 and 372, reducing noise by 44.7 % and $\text{7 5. 5 \%}$, and boosting precision from $\text{4 4 \%}$ to $\text{9 2 - 9 4 \%}$ and recall from 54 % to $94-96 \%$. LLaVA started with 1,381 triplets, refining to 720 and 270, with noise reductions of 47.9 % and 80.4 %, while improving precision from 46 % to $76-90 \%$ and recall from $\text{5 8 \%}$ to $\text{8 8 - 9 4 \%}$. This work establishes a foundation for automated medical knowledge extraction that bridges the gap between raw clinical text and structured, actionable medical insights.

14 October 2026 in Clinics and Practice
Verified
Journal Article
Open access
PDF file
BEBilal ERTUĞRUL
Bilal ERTUĞRUL
HKHarun Kömürcü
Harun Kömürcü
Background: Chronic subdural hematoma (CSDH) recurs in 5–20% of patients despite surgical treatment, and readily accessible biomarkers associated with recurrence remain limited. This study evaluated perioperative hematological, biochemical, and metabolic biomarkers associated with CSDH recurrence, focusing on preoperative lactate and the lactate-to-albumin ratio (LAR). Methods: This retrospective single-center cohort study included 163 adults who underwent burr-hole evacuation for CSDH between January 2021 and December 2025. All patients underwent saline irrigation without placement of a postoperative subdural drain, in accordance with the institutional surgical protocol. Clinical, radiological, laboratory, and perioperative arterial blood gas data were analyzed. Recurrence was defined as symptomatic re-expansion requiring repeat surgery within six months. Receiver operating characteristic (ROC) analysis and multivariable logistic regression were used to assess discrimination and associations with recurrence. Results: Postoperative recurrence occurred in 22 patients (13.5%). Recurrence was associated with higher preoperative lactate, LAR, neutrophil-to-lymphocyte ratio, and systemic immune-inflammation index, as well as lower serum albumin. LAR showed moderate discrimination (AUC 0.758, 95% CI 0.652–0.864), while lactate had an AUC of 0.744. Increased LAR (OR 1.42 per 0.1-unit increase, 95% CI 1.15–1.76; p = 0.001), bilateral hematoma (OR 2.91, 95% CI 1.08–7.84; p = 0.035), and mixed-density hematoma (OR 2.64, 95% CI 1.01–6.89; p = 0.047) were independently associated with recurrence in the fitted model. The age × lactate interaction did not reach conventional statistical significance (p = 0.058) and was considered exploratory. Conclusions: Higher preoperative LAR was associated with CSDH recurrence and showed moderate discriminative performance in this cohort. These findings are exploratory and require independent validation before LAR or the combined model can be considered for clinical risk prediction.

28 July 2026 in Balkan Journal of Electrical and Computer Engineering
Verified
Journal Article
Open access
BEBilal ERTUĞRUL
Bilal ERTUĞRUL
Background: Lumbar spinal stenosis (LSS) is a prevalent degenerative spinal disorder in neurosurgical practice, where magnetic resonance imaging (MRI) plays a central role in diagnosis and surgical planning. In recent years, artificial intelligence (AI), particularly deep learning–based models, has shown promising results in automated MRI analysis, including stenosis grading and segmentation. However, most existing AI systems rely heavily on predefined quantitative parameters, raising concerns regarding their clinical applicability.Objective: This study aims to critically evaluate the limitations of parameter-based AI models in the MRI assessment of LSS and to propose a conceptual framework for clinically aligned, explainable hybrid AI approaches.Methods: A narrative and conceptual analysis was conducted based on current literature and clinical experience. The limitations of parameter-based AI models were examined in terms of standardization, clinical decision-making complexity, clinicoradiological mismatch, multilevel disease evaluation, imposed parameterization, and labeling variability.Results: Parameter-based AI models demonstrate important limitations in reflecting real-world clinical decision-making. The lack of standardized thresholds, inability to integrate multidimensional clinical factors, and challenges in identifying symptomatic levels in multilevel disease reduce their clinical utility. Additionally, interobserver variability in labeling introduces noise that negatively affects model performance and generalizability.Conclusion: While parameter-based AI models contribute to the quantitative evaluation of LSS, they remain insufficient for comprehensive clinical decision support. Explainable hybrid AI approaches that integrate imaging data with clinical variables and structured knowledge representations may offer a more reliable and clinically meaningful framework for future applications.

7 June 2026 in Journal of Emergency Practice and Trauma
Verified
Journal Article
Closed access
PDF file
BEBilal ERTUĞRUL
Bilal ERTUĞRUL
IAİsMail Akçin
İsMail Akçin
Introduction: Traumatic basal ganglia hemorrhage is a rare phenomenon compared with spontaneous hypertensive intracerebral hemorrhage. We aimed to present a rare case of traumatic basal ganglia hemorrhage following a motor vehicle accident and discuss its clinical and radiological features. Case Presentation: A 59-year-old woman was admitted to the emergency department after a motor vehicle accident with confusion and left-sided weakness. Her Glasgow Coma Scale score was 12. Cranial computed tomography revealed a right basal ganglia hematoma measuring approximately 26 × 20 mm without midline shift or intraventricular hemorrhage. The patient was managed conservatively with antiedema treatment and close neurological follow-up. Follow-up imaging demonstrated gradual hematoma resolution, and the patient recovered without neurological deficit. Conclusion: Traumatic basal ganglia hemorrhage is an uncommon form of traumatic intracerebral hemorrhage. Careful differentiation from spontaneous hypertensive hemorrhage is essential. Conservative management may provide favorable outcomes in patients without significant mass effect or neurological deterioration.

31 August 2025 in Sinir Sistemi Cerrahisi Dergisi
Verified
Journal Article
Open access
PDF file
BEBilal ERTUĞRUL
Bilal ERTUĞRUL
IAİsMail Akçin
İsMail Akçin
BABekir Akgün
Bekir Akgün
MKMetin Kaplan
Metin Kaplan
Giriş: Tethered spinal kord sendromu (TSKS), konus medullarisin olması gerekenden daha aşağıda sonlanmasıyla nörolojik, ürolojik ve ortopedik semptomlara neden olan bir omurilik malformasyonudur. Sıklıkla çocukluk yaş grubunda görülür. Yapılan çalışmalarda intraoperatif nöromonitörizasyon (İN) kullanımının olası komplikasyonları azalttığı ve daha güvenli cerrahi sağladığı gösterilmiştir ancak hala tartışmalı bir konudur. Bu çalışmada kliniğimizde ameliyat edilen hastalarda İN kullanımını retrospektif olarak değerlendirdik. Yöntem: 2016 ve 2023 yılları arasında kliniğimizde ameliyat edilen 31 hasta İN kullanımına göre karşılaştırıldı. Hastaların yaşı, cinsiyeti, semptomları ve ameliyat öncesi nörolojik muayeneleri incelendi. Ameliyat sırasında 15 hastada İN kullanılırken, 16 hastada İN kullanılmadı. Bulgular: Hastaların nörolojik bulguları ameliyat sonrasında değerlendirildi. Ameliyat sonrası İN kullanmayan sadece 1 hastada ameliyat sonrası defisit gözlenirken, diğer 30 hastada ameliyat sonrası ek nörolojik defisit gözlenmedi. Sonuç: Literatürde yasal sorunları önlemek ve komplikasyonları azaltmak için TSKS hastalarında İN kullanımı önerilse de, ameliyat merkezinin deneyimine dayanarak, İN kullanılmamasının ameliyat sonrası komplikasyon oranını etkilemediği görülmüştür.