WoS İndeksli Yayınlar Koleksiyonu / WoS Indexed Publications Collection

Permanent URI for this collectionhttps://hdl.handle.net/20.500.14517/18

Browse

Now showing 1 - 20 of 2803

Thiamine (Vitamin B1) Levels and Sensory Properties of Cooked Rice: Exploring Varietal Differences
(Sage Publications Ltd, 2025) Tireki, Suzan; Dagistan, Hazal; Cam, Dilaycan; Cakir, Cansel; Ozturk, Mehmet; Demirkoz, Asli Barla
The composition and cooking quality of rice varieties vary highly. Preparation and cooking techniques may lead to thiamine degradation. In this study, rice meals were prepared with three different rice varieties grown in T & uuml;rkiye (Baldo, Osmancik, and Cammeo), four different fats (sunflower oil, margarine, butter, and olive oil), and two different cooking methods (Salma and Roasted), and thiamine amount and sensory attributes were investigated. Thiamine levels of raw rice were measured as 0.399 +/- 0.0210, 0.299 +/- 0.0010, and 0.303 +/- 0.0070 mg/100 g in Baldo, Osmancik, and Cammeo, respectively. The thiamine amount varied between 0.038 +/- 0.0020 and 0.176 +/- 0.0020 mg/100 g in rice meals, and the highest value was observed in Osmancik rice cooked with margarine using the Roasted method. Descriptive sensory analysis (DSA) assessed 8 taste attributes. Salma-cooked samples had a higher intensity of starchy and butter taste than the Roasted-cooked ones. Roasted-cooked rice meals had a higher intensity of sweet, salty, margarine, olive oil, sunflower oil, and roasted taste than the Salma-cooked samples.
A Reliable Deep Neural Network Using the Radial Basis for the Spreading Virus in Computers with Kill Signals
(Elsevier, 2026) Sabir, Zulqurnain; Basbous, Bahaa; Souayeh, Basma; Umar, Muhammad; Salahshour, Soheil
Purpose: The purpose of this work is to provide a reliable neural network process for the spreading virus in computers with kill signals. The mathematical model shows susceptible, exposed, infected individuals to form the virus inactive, and kill signals classes. Method: A structure of deep neural network (DNN) is designed by using two different hidden layers having radial basis activation functions in both layers, optimization through the Bayesian regularization, twenty and thirty numbers of neurons in primary and secondary hidden layers for the spreading virus in computers with kill signals. The stochastic DNN framework is presented to solve the spreading virus in computers with kill signals by selecting the data for training as 70 %, and 15 %, 15 % for both validation and testing. Results: The accuracy of the scheme is observed through the overlapping of the solutions along with negligible absolute error for solving the model. The consistency of the solver is observed through the process of error histogram, regression, and state transition. Novelty: The proposed DNN structure having radial basis activation function has never been applied for the spreading virus in computers with kill signals.
Thermal and Stability Outcomes of Different Osteotomy Techniques and Implant Macrogeometries in Type IV Bone: An in Vitro Study
(MDPI, 2025) Canakci, F. Gulfesan; Cakir, Merve; Yalcin-Ulker, Gul Merve; Duygu, Gonca
Achieving reliable implant stability in type IV (D4) bone remains a clinical challenge due to its thin cortical plates and low trabecular density, which increase the risk of early failure. Novel osteotomy techniques such as osseodensification and the OsseoShaper have been proposed, yet their effects when combined with different implant macrogeometries are not fully understood. This in vitro study evaluated forty osteotomies in standardized polyurethane foam blocks simulating D4 bone density. Three site preparation protocols-conventional osteotomy, osseodensification, and OsseoShaper-were combined with two implant macrogeometries (parallel-walled conical and tri-oval tapered designs). Insertion torque (IT) was measured using a calibrated digital torque meter, and implant stability was assessed by resonance frequency analysis (ISQ). Intraoperative thermal changes were monitored with infrared thermography under constant irrigation. Statistically significant differences were observed among groups (p < 0.05). Osseodensification with parallel-walled implants achieved the highest stability, whereas osseodensification with tri-oval implants showed the lowest. Although osseodensification produced the greatest thermal increase, all values remained below the 47 degrees C osteonecrosis threshold. Within the study's limitations, both the osteotomy technique and implant macrogeometry significantly affected stability and thermal outcomes, with osseodensification plus parallel-walled implants providing the most predictable performance in D4 bone.
Extracorporeal Membrane Oxygenation in Nontraumatic Critical Care: New Horizons in Resuscitation
(MRE Press, 2025) Akgun, Feride Sinem; Ozkaya, Bilgen; Ozturk, Goksu Afacan; Ozel, Betul Akbuga; Karcioglu, Ozgur
Extracorporeal membrane oxygenation (ECMO) transfers blood from a large vein into a machine, oxygenates it, and then reinfuses the oxygenated blood back to the patient through a large vascular catheter. Resuscitation procedures using ECMO are known as extracorporeal cardiopulmonary resuscitation or extracorporeal life support. On one hand, the procedure is suitable for selected patients with severe respiratory diseases, cardiogenic or septic shock, intoxications, thyrotoxicosis, trauma, or cardiac arrest. On the other hand, geriatric persons with multiple diseases, end-stage malignancies, those with cardiopulmonary diseases, or dementia are not candidates for extracorporeal cardiopulmonary resuscitation (ECPR). The potential indications for ECMO are still expanding, but well-designed, multicentric studies are needed to assess benefit and harms.
Observation of WZγ Production and Constraints on New Physics Scenarios in Proton-Proton Collisions at √s=13 Tev
(American Physical Society, 2025) Chekhovsky, V.; Hayrapetyan, A.; Makarenko, V.; Tumasyan, A.; Adam, W.; Andrejkovic, J. W.; Druzhkin, D.
A measurement of the WZ gamma triboson production cross section is presented. The analysis is based on a data sample of proton-proton collisions at a center-of-mass energy of root s = 13 TeV recorded with the CMS detector at the LHC, corresponding to an integrated luminosity of 138 fb(-1). The analysis focuses on the final state with three charged leptons, l(+/-)nu l(+)l(-), where l = e or mu, accompanied by an additional photon. The observed (expected) significance of the WZ gamma signal is 5.4 (3.8) standard deviations. The cross section is measured in a fiducial region, where events with an l originating from a tau lepton decay are excluded, to be 5.48 +/- 1.11 fb, which is compatible with the prediction of 3.69 +/- 0.24 fb at next-to-leading order in quantum chromodynamics. Exclusion limits are set on anomalous quartic gauge couplings and on the production cross sections of massive axionlike particles.
Comparison of Shear Bond Strength of Different Indirect Composites to New Generation Polyetheretherketone Materials
(Wroclaw Medical University, 2025) Seferli, Ziya; Degirmenci, Kubra; Saridag, Serkan
Background. The bond stability between polyetheretherketone (PEEK) materials and composites is a novel concern, and evidence regarding bond strength is limited. To date, no study has comprehensively evaluated the effect of different surface treatments, adhesive agents and composite materials on the shear bond strength (SBS) of various PEEK materials. Objectives. The aim of the study was to compare the SBS between 2 PEEK materials and different indirect composites applied after the application of various surface pre-treatment methods and adhesives. Material and methods. A total of 328 PEEK specimens (JUVORA (TM) (unfilled PEEK material, n = 164); BioHPP (compound containing 20% nanoceramic-filled PEEK, n = 164)) were divided into 4 groups according to the applied surface treatment: no treatment; air abrasion; acid etching; and acid etching + air abrasion. Subsequently, all specimens were conditioned with visio.link (VL) or Single Bond Universal (SBU). The specimens were veneered with crea.lign composite (CR) or SR Nexco composite (SR), and the bond strength values were measured. The co-variance analysis of variance (ANOVA) was performed to analyze the data (alpha = 0.003). Results. The SBS values for BioHPP specimens were significantly higher than the values for JUVORA (TM) specimens in the no treatment group (p < 0.001). The highest SBS values were detected between BioHPP and SR (22.54 +/- 1.22 MPa), and between JUVORA(TM )and SR (21.45 +/- 1.43 MPa) after acid etching and conditioning with VL (p < 0.001). The surface treatments, composites and adhesives affected the SBS between the composites and PEEK materials. Conclusions. Following air abrasion or acid etching of surfaces, conditioning with VL and aesthetic veneering with SR may bea more reliable clinical application than othersurface treatments and adhesive-composite combinations for PEEK.
Buyer Platform Experience (PLX) in Online Marketplaces: Scale Development, Validation, and the Salience of Variety, Competitive Pricing, Service Excellence, and Promotion
(Sage Publications Inc, 2025) Metehan, Celal Teoman; Sozer, Edin Guclu; Civelek, Mustafa Emre; Ertemel, Adnan Veysel; Mert, Ibrahim Sani
This study aims to develop and validate a Buyer Platform Experience (PLX) scale focused on the cognitive and relational dimensions of buyer experience in online marketplaces. Adopting a five-stage sequential scale-development methodology, the research combines qualitative interviews with quantitative analyses using SPSS and AMOS. A four-dimensional scale-variety, service excellence, competitive pricing, and promotional effectiveness-was validated through exploratory and confirmatory factor analyses. Structural equation modeling confirmed that variety, service excellence, and competitive pricing significantly predict customer loyalty. The findings isolate cognitive and relational aspects often overlooked in existing customer-experience frameworks and offer practical guidance for platform managers seeking to enhance buyer retention through service quality, price competitiveness, and variety management.
Cultural Adaptation and Turkish Version of the Late-Life Function and Disability Instrument for Individuals with Gonarthrosis: A Validity and Reliability Study
(Taylor & Francis Ltd, 2025) Hazar, Yagmur; Atici, Emine; Cagliyan Erdogan, Hande; Kardes, Kubra
Purposes: This study aimed to culturally adapt and assess the psychometric properties of the Turkish version of the Late-Life Function and Disability Instrument (LLFDI) in elderly individuals with knee osteoarthritis (gonarthrosis). Methods: A total of 240 participants aged >= 65 were selected using simple random sampling. The adaptation process followed Beaton's guidelines and included translation, back-translation, and pilot testing. Reliability was assessed using Cronbach's alpha, test-retest, and intraclass correlation coefficients (ICC), while validity was evaluated via exploratory and confirmatory factor analyses and convergent validity with established scales. As part of convergent validity, the relationship of the LLFDI with Barthe Index (BI) and the Functional Independent Measurement (FIM), the Lequesne Index were used. The correlations between the LLFDI scores and these scales were analyzed to determine the level of convergent validity. Results: The LLFDI demonstrated high internal consistency (Cronbach's alpha > 0.70) and test-retest reliability (ICC = 0.734-0.899). Construct validity was supported by Exploratory Factor Analysis (KMO = 0.838-0.952) and significant Bartlett's Test (p < 0.001), revealing appropriate factor structures. Convergent validity was confirmed through significant correlations with the BI, FIM, and Lequesne Index. Conclusion: This study demonstrated that the Turkish version of the LLFDI is a valid and reliable assessment tool for individuals diagnosed with gonarthrosis.
Constraints on Standard Model Effective Field Theory for a Higgs Boson Produced in Association with W or Z Bosons in the H → B(b)over-Bar Decay Channel in Proton-Proton Collisions at √s=13 Tev
(Springer, 2025) Chekhovsky, V.; Hayrapetyan, A.; Makarenko, V.; Tumasyan, A.; Adam, W.; Andrejkovic, J. W.; Lee, J.
A standard model effective field theory (SMEFT) analysis with dimension-six operators probing nonresonant new physics effects is performed in the Higgs-strahlung process, where the Higgs boson is produced in association with a W or Z boson, in proton-proton collisions at a center-of-mass energy of 13 TeV. The final states in which the W or Z boson decays leptonically and the Higgs boson decays to a pair of bottom quarks are considered. The analyzed data were collected by the CMS experiment between 2016 and 2018 and correspond to an integrated luminosity of 138 fb(-1). An approach designed to simultaneously optimize the sensitivity to Wilson coefficients of multiple SMEFT operators is employed. Likelihood scans as functions of the Wilson coefficients that carry SMEFT sensitivity in this final state are performed for different expansions in SMEFT. The results are consistent with the predictions of the standard model.
A High-Performance Neural Network Algorithm Using a Legendre Ensemble-Based Extreme Learning Machine for Solving Fractional Partial Differential Equations
(Elsevier, 2026) Isah, Ibrahim Onimisi; Senu, Norazak; Ahmadian, Ali
The recent advancement in the use of machine learning techniques across various fields has paved the way for innovative approaches to solving fractional partial differential equations (FPDEs), particularly those utilizing neural networks (NNs). These methods enable efficient representation of complete solutions, leveraging the universal approximation capabilities of neural networks. This study presents a neural network-based method that utilizes the ensemble extreme learning machine (EN-ELM) to efficiently solve FPDEs considered in the sense of the Caputo fractional derivative. The proposed approach incorporates Legendre polynomials to expand input features and employs the radial basis function as the activation function for hidden layer neurons. The EN-ELM framework, enhanced with cross-validation, ensures improved accuracy, stability, and reduced computational complexity. Numerical experiments are conducted to validate the approach, demonstrating its superior accuracy, execution time, and error minimization compared to some known methods. The results confirm the robustness and effectiveness of the proposed method for solving FPDEs.
A Numerical Method for the Solution of the Two-Dimensional Time-Fractional Cable Equation of Distributed-Order Involving Riesz Space-Fractional Operators
(Elsevier, 2026) Derakhshan, M. H.; Marasi, H. R.; Kumar, Pushpendra
this paper, we propose an efficient hybrid numerical approach to obtain approximate solutions to the two-dimensional time-fractional cable equation of distributed-order involving Riesz space-fractional operators. This combined numerical approach includes two numerical approaches in time and space directions. The weighted and shifted Gr & uuml;nwald difference numerical method is used to approximate the fractional problem in the time direction, and the fractional compact numerical method is applied in the direction of the space variable. The stability and convergence analyses are studied for the proposed numerical approach. To show the effectiveness of the presented numerical method, some numerical examples are given, and the numerical results for these examples are plotted. Also, this numerical approach is compared with other methods.
Bone-Preserving Osteotomy Strategies for Narrow Ridges: Comparative in Vitro Analysis of Densifying, Low-Speed Shaping, and Conventional Protocols
(MDPI, 2025) Canakci, Fatma Gulfesan; Cakir, Merve; Yalcin-Ulker, Gul Merve; Duygu, Gonca
Adequate ridge width and thermal control are critical for predictable implant site preparation. This in vitro study compared conventional osteotomy (CO), osseodensification (OD), and OsseoShaper (OS) protocols in standardized polyurethane foam blocks simulating narrow D2 ridges. A total of 18 osteotomies (n = 6 per group) were prepared under protocol-specific irrigation. Ridge width was measured at 3, 6, and 9 mm apical to the crest before and after osteotomy using a digital caliper, and expansion (Delta W) was calculated. Intraoperative thermal changes (Delta T) were recorded in real time with an infrared thermal camera. OD achieved consistent expansion at all depths (p < 0.05), while OS produced significant widening at the 3 and 6 mm levels; CO yielded only a minor but significant gain at the 3 mm level. At the intergroup level, OS showed significantly greater crestal expansion than CO at 3 mm (p = 0.006). All protocols generated comparable thermal changes (mean Delta T 7.4-8.2 degrees C), remaining well below the critical 47 degrees C threshold. Within the limitations of this in vitro model, OD and OS enhanced ridge expansion compared with CO, particularly at the crestal level, where expansion is most critical. All protocols maintained thermally safe profiles, supporting their clinical applicability.
Comment on 'What Is the Force on a Magnetic Dipole?’
(IOP Publishing Ltd, 2025) Kholmetskii, Alexander L.; Missevitch, Oleg, V; Yarman, Tolga
We identify errors in the paper by Franklin (2018 Eur. J. Phys. 39 3) and particularly invalidate his claim that the introduction of the hidden momentum of the magnetic dipole violates the conservation of total momentum. Also, we emphasize that the contribution of the 'hidden momentum' to the force on the magnetic dipole plays an important role in the implementation of the 'center-of energy' theorem of classical electrodynamics, and earlier attempt of Franklin to disprove this theorem are erroneous.
The Mediating Impact of Knowledge-Based HRM Practices on the Relationship Between Transformational Leadership and Open Innovation
(Emerald Group Publishing Ltd, 2025) Alameddine, Ashraf
PurposeThis paper aims to examine the mediating role of knowledge-based human resource management (KBHRM) practices in the relationship between transformational leadership (TL) and open innovation for Lebanese manufacturing companies. Blending the Knowledge-Based View (KBV), leadership theory and HRM, the present paper aims at explaining the influence of TL on inbound and outbound open innovation through knowledge-related practices.Design/methodology/approachA cross-sectional survey of 370 Lebanese manufacturing companies was conducted, and 255 usable questionnaires (69% response rate) were retrieved. A structured format questionnaire was utilized to quantify TL dimensions, KBHRM practices and open innovation activities. Reliability and validity tests, confirmatory factor analysis and structural equation modeling (AMOS) were conducted to test hypothesized relations. Mediation analysis was conducted using bootstrapping to identify indirect effects.FindingsThe results illustrate the presence of strong and positive correlations among TL, KBHRM practices and open innovation. TL emerged as a strong predictor of KBHRM practices (beta = 0.949), inbound innovation (beta = 0.674) and outbound innovation (beta = 0.428). KBHRM practices were positively influencing inbound (beta = 0.25) and outbound innovation (beta = 0.28). Mediation results further indicated that KBHRM practices had a significant mediating effect in projecting TL's impact to open innovation outcomes with large indirect effects.Originality/valueIt is a contribution to the Knowledge-Based View since it empirically verifies KBHRM as a dynamic capability that converts leadership into knowledge-based innovation outcomes. To the best of the authors' knowledge, it is the first scholarly work to explore KBHRM as a mediator of TL and open innovation in an emerging-economy context and offer theoretical contributions and managerial/practical implications.
Measurement of the W Boson Decay Branching Fraction Ratio B(W→cq)/B(W→qq¯′) in Proton-Proton Collisions at √s=13TeV
(Elsevier, 2025) Hayrapetyan, A.; Tumasyan, A.; Adam, W.; Andrejkovic, J. W.; Benato, L.; Bergauer, T.; Kim, J.
The most precise measurement to date of the W boson hadronic decay branching fraction ratio R-c(W) = B(W -> cq)/B(W -> q (q) over bar') is presented. The measurement is based on a sample of proton-proton collision data from the CERN LHC collected by the CMS experiment at a center-of-mass energy of 13TeV in 2016-2018 with an integrated luminosity of 138 fb(-1). The large cross section of top quark-antiquark production at the LHC offers a sizable high-purity sample of W bosons suitable for this measurement. Events with one charged lepton (electron or muon) and at least four jets, two tagged as bottom quark jets, are analyzed. Charm jets are tagged using the presence of a muon inside the jet. The result, R-c(W) = 0.489 +/- 0.020, is consistent with the standard model prediction and is twice as precise as the current world-average value.
A Comparative Study of Biochemical, Antimicrobial Effects and Phytochemical Composition Analysis of Glycyrrhiza Glabra L. Varieties Root Extracts
(Marmara University, Faculty of Pharmacy, 2025) Gulmez, Gizem; Sen, Ali; Servi, Huseyin; Barak, Timur Hakan; Tekin, Fetullah; Marzi, Mahdi; Sener, Azize
Plants are the significant global interest as alternative treatment sources with their biologically active compounds. This study compares the chemical composition and the antioxidant, antidiabetic, and antimicrobial properties of ethanol extracts of G. glabra L. two different varieties from different regions. The phytochemical compositions was determined using GC-MS. Additionaly, total phenolic (TPC), flavonoid (TFC) and triterpene (TTC) contents were determined. Glycyrrhizic acid contents were analysed by HPLC. G. glabra var. glandulifera (GF1) showed the highest antioxidant activity. All extracts had strong antidiabetic effects, besides GF1 showing the highest effect. The MIC values was determined against 8 bacterial and 1 yeast strain and values ranged from 2.500 to 0.500; 2.500 to 0.714; 2.500 to 0.714 for G. glabra var. glabra (GB), GF1, G. glabra var. glandulifera (GF2) respectively. Phytochemical studies have shown that TPC was 100.60 +/- 5.06, 127.90 +/- 0.30, 69.01 +/- 0.30 mg GAE /g extract; TFC was 80.07 +/- 0.15, 25.35 +/- 0.0, 16.58 +/- 0.31 mg KE/g and TTC was 217.30 +/- 6.05,172.40 +/- 2.17, 126.30 +/- 4.50 mg OE/g extract for GB, GF1, GF2, respectively. GF1 in particular has the highest glycyrrhizic acid content. This study will contribute to the creation of new treatment strategies and potential therapeutic agents in addition to the use of G. glabra L. in traditional treatments. Our study is also a preliminary study for future studies.
Managing Post-Disaster Primary Health Care: Experiences of Public Health Professionals After the 2023 Kahramanmaraş Earthquake
(Frontiers Media SA, 2025) Sahan, Ceyda; Bilgin, Ahmet Can; Kilic, Bulent; Bulut, Pelin; Mert, Esra; Kantas, Secil Nur; Erbaydar, Tugrul
Objectives To explore the post-earthquake experiences and perspectives of public health specialists, with a particular focus on the delivery of public health services following the 2023 Kahramanmara & scedil; earthquake.Methods This qualitative study employed a phenomenological approach, aiming to capture lived experiences and contextual understanding of the situation. In-depth interviews were conducted with 15 public health specialists. Participants were selected through purposive sampling. Data were collected via in-depth interviews and analyzed thematically using multiple coding by the research team.Results Public health specialists reported experiencing stress, burnout, and housing difficulties due to increased workloads and suboptimal working conditions, despite also expressing a sense of professional fulfillment in crisis management. Key challenges included problems in temporary housing, infectious disease control, and insufficient coordination in primary healthcare services and community-based approaches. Poor coordination, vague job roles, and inadequate training and resources were cited as significant barriers to effective disaster response.Conclusion The main challenges following the earthquake included increased workloads and unclear role definitions. These findings underscore the importance of establishing pre-defined job descriptions and clear lines of authority to improve disaster preparedness and response in the health sector.
Electrochemical and Optical Biosensors for Periodontitis Detection
(Elsevier, 2026) Zarasvandi, Amir Masoud; Gerayeli, Mohammad; Abasi, Mina; Goodarzi, Mehdi; Ganjehzadeh, Shabnam; Solhmirzaei, Raheleh; Milani, Haniyeh Alavi
Periodontitis is a common chronic inflammatory disease that frequently results in tooth loss and systemic complications. Traditional diagnostic methods have limited sensitivity and often fail to detect early disease activity. Biosensors have emerged as promising tools for the early and accurate detection of specific biomarkers in saliva and gingival crevicular fluid. This review highlights recent advancements in electrochemical, optical, lab-on-a-chip technologies, and nanosensors for periodontal diagnosis. These innovations provide rapid, noninvasive, point-of-care capabilities, enabling improved monitoring and personalized treatment. While challenges remain in clinical translation, biosensors have significant potential to transform periodontal diagnostics and enhance patient outcomes. Clinical trial number: not applicable.
A Fast and Sensitive Method for the Analysis of 4-Methylimidazole in Tea Extract: Stir-Bar Sorptive Extraction Combined with GC-MS
(Springer, 2025) Yasa, Murat; Sarikaya, Ebru; Ertug, Zeynep Balta; Dogan, Nuray; Demirkoz, Asli Barla
A novel and efficient analytical method was developed for the determination of 4-methylimidazole (4-MeI) in tea extracts using stir-bar sorptive extraction (SBSE) coupled with gas chromatography-mass spectrometry (GC-MS). Extraction parameters were optimized using ethylene glycol- and silicone-coated stir bars. The ethylene glycol-coated stir bar was used due to its higher extraction efficiency. The method demonstrated excellent linearity over the range of 2 mu g/mL and 100 mu g/mL (R2 = 0.9982), with a limit of detection (LOD) of 1.19 mu g/mL and a limit of quantification (LOQ) of 3.97 mu g/mL. Method precision was confirmed with intra- and inter-day relative standard deviations of less than 1.6% and recoveries exceeding 102.26 +/- 3.82%. The combined and expanded uncertainties were calculated as 0.0187 and 0.0373, respectively. Compared with liquid chromatography mass spectrometry (LC-MS/MS) and Quick, Easy, Cheap, Effective, Rugged and Safe-based (QuEChERS) methods, the developed SBSE-GC/MS protocol offers reduced matrix interference, simplified sample preparation, and comparable sensitivity. Real sample analyses validated the method's applicability for routine monitoring of 4-MeI in commercial tea-based products. This approach offers a fast, reliable, and eco-friendly alternative for monitoring food safety and detecting thermally generated contaminants.
Balancing Accuracy and Readability: Comparative Evaluation of AI Chatbots for Patient Education on Rotator Cuff Tears
(MDPI, 2025) Koluman, Ali Can; Ciftci, Mehmet Utku; Ciftci, Ebru Aloglu; Cakmur, Basar Burak; Ziroglu, Nezih
Background/Objectives: Rotator cuff (RC) tears are a leading cause of shoulder pain and disability. Artificial intelligence (AI)-based chatbots are increasingly applied in healthcare for diagnostic support and patient education, but the reliability, quality, and readability of their outputs remain uncertain. International guidelines (AMA, NIH, European health communication frameworks) recommend that patient materials be written at a 6th-8th grade reading level, yet most online and AI-generated content exceeds this threshold. Methods: We compared responses from three AI chatbots-ChatGPT-4o (OpenAI), Gemini 1.5 Flash (Google), and DeepSeek-V3 (Deepseek AI)-to 20 frequently asked patient questions about RC tears. Four orthopedic surgeons independently rated reliability and usefulness (7-point Likert) and overall quality (5-point Global Quality Scale). Readability was assessed using six validated indices. Statistical analysis included Kruskal-Wallis and ANOVA with Bonferroni correction; inter-rater agreement was measured using intraclass correlation coefficients (ICCs). Results: Inter-rater reliability was good to excellent (ICC 0.726-0.900). Gemini 1.5 Flash achieved the highest reliability and quality, ChatGPT-4o performed comparably but slightly lower in diagnostic content, and DeepSeek-V3 consistently scored lowest in reliability and quality but produced the most readable text (FKGL approximate to 6.5, within the 6th-8th grade target). None of the models reached a Flesch Reading Ease (FRE) score above 60, indicating that even the most readable outputs remained more complex than plain-language standards. Conclusions: Gemini 1.5 Flash and ChatGPT-4o generated more accurate and higher-quality responses, whereas DeepSeek-V3 provided more accessible content. No single model fully balanced accuracy and readability. Clinical Implications: Hybrid use of AI platforms-leveraging high-accuracy models alongside more readable outputs, with clinician oversight-may optimize patient education by ensuring both accuracy and accessibility. Future work should assess real-world comprehension and address the legal, ethical, and generalizability challenges of AI-driven patient education.

Browse

Recent Submissions