İstanbul Sağlık ve Teknoloji Üniversitesi Kurumsal Akademik Arşivi
DSpace@İSTÜN, Üniversite mensupları tarafından doğrudan ve dolaylı olarak yayınlanan; kitap, makale, tez, bildiri, rapor, araştırma verisi gibi tüm akademik kaynakları uluslararası standartlarda dijital ortamda depolar, Üniversitenin akademik performansını izlemeye aracılık eder, kaynakları uzun süreli saklar ve telif haklarına uygun olarak Açık Erişime sunar.
Güncel Gönderiler
Severity stratification changes what hallucination rates mean: An item-matched audit of five large language models in orthodontic decision support
(MDPI Publishing, 2026) Allaf, Ferdi; Özcan, Mustafa
Background/Objectives: Large language models (LLMs) are consulted for clinical decision support, and their reliability is judged by hallucination prevalence—a measure treating every unsupported element as equivalent, although a fabricated citation and a fabricated protocol differ in what a clinician acting on them would do. We tested whether prevalence tracks clinical risk. Methods: Five LLMs answered a 100-item orthodontic benchmark validated by three external orthodontists (content validity index 0.923). The reference standard was fixed by construction for the 45 items naming a non-existent entity, so any substantive elaboration is unsupported by design; citations were adjudicated against PubMed and CrossRef. Responses were coded with a seven-category taxonomy and assigned to clinical, operational, or epistemic severity tiers in a post hoc exploratory stratification, pre-specified rather than prospectively registered. Because all models answered the same items, comparisons used Cochran’s Q with pairwise McNemar tests and generalised estimating equations clustered on item. Results: Of 500 responses, 449 (89.8%) contained a hallucination but only 65 (13.0%; 95% CI 10.3–16.2) were clinically consequential: prevalence was roughly sevenfold greater than the rate of clinically consequential output as the authors defined it. Between-model differences were large for undifferentiated prevalence (73–100%; Q = 50.51, p < 0.001) and contracted at the clinical tier (10–16%; Q = 10.00, p = 0.040), where no pairwise contrast survived adjustment. Item-level clustering was far stronger for clinically consequential output than for undifferentiated prevalence (intra-class correlation 0.83 versus 0.07). Conclusions: Prevalence and clinically consequential error are not interchangeable and rank models differently. The stratification is exploratory and author-defined, and the benchmark stress-tests susceptibility to fabricated premises rather than surveying natural use.
Cost-effectiveness of universal and contingent nipt within a multi-tariff healthcare system: A full incremental decision-analytic cohort study
(Wiley, 2026) Bülbül, Mehmet; Çakır, Anıl Turhan; Aydın, Gerçek; Harma, Mehmet İbrahim; Öztürk, Azize Cemre; Barut, Aykut; Harma, Müge
Objective: To compare universal and contingent non‐invasive prenatal testing (NIPT) strategies within a multi‐tariff healthcare system using a full incremental cost‐effectiveness framework and to identify price levels at which universal NIPT becomes economically competitive with the best reflex strategy. Method: A deterministic decision‐analytic cohort model representing 1,000,000 pregnancies was developed. For each strategy, trisomy detection, false‐positive and missed cases, invasive procedures, procedure‐related fetal loss, and short‐term psycho logical quality‐adjusted life year (QALY) losses were estimated. Medical costs were derived from public, university, and private‐ sector tariffs. Incremental costs, QALY losses avoided, and cost‐effectiveness ratios were calculated across all strategies; CT‐based pairwise analyses were retained as secondary analyses. Price‐threshold analyses and probabilistic sensitivity analysis generated threshold estimates and acceptability curves. Results: Contingent NIPT strategies markedly reduced false‐positive results and invasive procedures, yielding the lowest short‐ term QALY losses. In full incremental analysis, Reflex300 was the only strategy on the frontier in private‐sector scenarios, whereas CT and Reflex300 remained on the frontier in SUT and university scenarios. uNIPT was not selected on the frontier and became economically competitive only at substantially reduced unit prices. Conclusions: Reimbursement structure strongly influenced cost‐effectiveness. Contingent NIPT was robust, whereas uNIPT required substantially lower tariff‐aligned pricing to become economically competitive.
Controversies in PCL reconstruction: Why does PCL reconstruction still underperform? Failure mechanisms, residual laxity, and a risk management framework
(SAGE Publications, 2026) Erden, Tunay; Tollefson, Luke V.; Wilebski, Benjiman J.; Sawyer, Ethan R.; Freetly, Taylor J.; LaPrade, Robert F.
Background: Posterior cruciate ligament (PCL) reconstruction (PCLR) remains among the least standardized procedures in knee ligament surgery. Despite substantial technical advances, objective residual posterior laxity and functional deficits persist at rates far exceeding those observed after anterior cruciate ligament reconstruction. Purpose: To critically evaluate the principal controversies in PCLR through the lens of failure mechanisms, residual instability, and individualized risk management. Study Design: Narrative review. Methods: A comprehensive search of MEDLINE/PubMed, Embase, and the Cochrane Library was conducted from database inception through February 27, 2026. Search terms included PCLR, PCL failure, residual laxity, posterolateral corner (PLC), tibial slope, allinside PCL, tibial inlay, double-bundle PCL, and suture augmentation. Published consensus statements and relevant knee ligament surgery guidance documents from the American Academy of Orthopaedic Surgeons, the International Society of Arthroscopy, Knee Surgery and Orthopaedic Sports Medicine, and the Orthopaedic Research Society were additionally reviewed. Biomechanical investigations, cadaveric studies, prospective cohort studies, systematic reviews, and meta-analyses were prioritized; case series and expert opinion were included when appropriate. Results: Eight major controversies were identified: (1) single- versus double-bundle reconstruction; (2) transtibial versus tibial inlay technique; (3) autograft versus allograft selection; (4) the clinical significance of residual posterior laxity; (5) management of concomitant PLC deficiency; (6) rehabilitation and hamstring reintroduction timing; (7) the role of posterior tibial slope correction; and (8) suture augmentation. Across all domains, biomechanical advantages of technically superior approaches have failed to consistently translate into superior patient-reported outcomes. Failure to address PLC insufficiency, decreased posterior tibial slope, and graft-loading forces during rehabilitation represents the most common and preventable constellation of failure risk factors. A standardized definition of PCLR failure does not currently exist in the published literature. Conclusion: Successful PCLR depends less on identification of a universally superior technique and more on individualized management of biomechanical risk factors. A practical risk stratification framework is proposed to guide surgical decision-making. Future progress requires standardized failure definitions, objective instability assessment protocols, and criterion-based rehabilitation pathways.
In silico prediction of protein targets for benzofuranone ring-containing agents and molecular docking analysis
(Bentham Science Publishers, 2026) Ata, Fatma Kübra; Çalışkan, Zerrin; Özdemir Özgentürk, Nehir
Introduction: Benzofuranone and its derivatives, which serve as drug precursors and intermediates due to their biological activity, are of great importance in the pharmaceutical industry. These heterocyclic molecules, composed of benzene and furan chains, possess antifungal, antiinflammatory, antidepressant, and anticancer properties. Given their pharmaceutical importance, mapping benzofuranone targets is a crucial step toward elucidating their molecular mechanisms of action and predicting potential side effects or cross-reactivity. Methods: In this study, we evaluated the in silico pharmacokinetic properties and potential protein targets of 15 derivatives of 2(3H)-Benzofuranone. Chemical structures were constructed and optimized using Spartan software. Next, SwissADME and ProTox servers were used to analyze pharmacokinetic features, while SwissTargetPrediction and STRING/Reactome were used to predict protein targets. In the final stage, the binding affinities of benzofuranone-hybrid molecules for Homo sapiens CDK1 and PLK1 proteins were calculated using molecular docking. Results: Ligands M and N were classified as class IV toxicity and met all Lipinski, Ghose, Veber, Egan, and Muegge filter conditions. The estimated LD50 values (rat acute oral toxicity) of these ligands were found to be 2000 mg/kg. Protein target prediction showed that the M ligand interacts with CDK1 and PLK1 proteins responsible for microtubule stability and spindle fiber organization. Molecular docking results predicted favorable binding of ligand M within the CDK1 and PLK1 binding sites, with mean binding energies of -8.2 kcal/mol and -7.7 kcal/mol, respectively. Discussion: These results were comparable to those obtained for the reference CDK1 inhibitor Abemaciclib (-10 kcal/mol) and the PLK1 inhibitor Volasertib (-8.2 kcal/mol). Furthermore, structural comparison with the reference inhibitor Abemaciclib predicted a partially overlapping binding pattern in the CDK1 binding domain (Lys33, Ile10, and Leu135) of ligand M. Conclusion: In our study, we identify ligand M as a promising starting scaffold for derivatizing benzofuranone- based compounds. In silico information provided in this study may serve as a basis for future researchers to design, synthesize, and evaluate new derivatives of this ligand with improved biological properties.
Synthesis, moleculer modeling, and carbonic anhydrase inhibitor effects of novel 1-(1-benzyl-2-oxoindolin-3- ylidene)-4-(3-sulfamoylphenyl)thiosemicarbazides
(Wiley, 2026) Apaydın, Çağla Begüm; Cinek, Tuğçe; Nadaroğlu, Hayrunnisa; Karalı, Nilgün
In this study, we describe the synthesis of new 1-(1-benzyl-2-oxoindolin-3-ylidene)-4-(3-sulfamoylphenyl)thiosemicarbazides and evaluate their inhibitory activity against hCAs (6a–l). The inhibitory effects of compounds 6a–l were examined in vitro against hCA I, II, IX, and XII isoforms. Acetazolamide (AAZ) was employed as the standard reference compound in this study. The compounds 6a–l inhibited the tumor-associated hCA IX and XII isoenzymes at nanomolar levels, and showed weak inhibitory effects on the off-target hCA I and II isoenzymes. Selectivity index (SI) values of compounds 6a–l were determined to be significantly high for hCA IX and XII over hCA I and II compared to AAZ. Compound 6k (R1 = Cl, R2 = 3-Cl) showed the strongest inhibitory effect (Ki = 25.23 nM) and high selectivity (SI = 225, 185) against hCA IX over hCA I and II. Compound 6f (R1 = CH3, R2 = 4-Cl) was found to have the highest inhibitory effect (Ki = 36.65 nM) and high selectivity (SI = 377 and 192, respectively) against hCA XII over hCA I and II. Molecular docking and dynamic simulation analyses were conducted to explore how compounds 6f and 6k interact with active sites of hCA XII and IX enzymes.
