Esra Akdeniz

Aims: The aim of this study is to systematically evaluate the performances of large language model-based generative Artificial Intelligence (Gen-AI) tools, Gemini and Copilot, in the generation and assessment of multiple-choice questions (MCQs) for use in medical education.Methods: A total of 335 MCQs were generated from two virtual patient cases using standardized prompts. Gen-AI tools selected the 56 best-quality items based on criteria encompassing the intended distributions regarding acceptable level of performance (ALP), Miller's competency pyramid (Miller) and Bloom's revised taxonomy (Bloom) levels, as well as alignment with learning objectives (LOs). Expert medical educators and current Gen-AI tools assessed these items based on the identification of misleading/confusing distractor(s) for borderline candidates -minimally competent examinees- (to calculate ALP values) and the identification of key(s), as well as Miller and Bloom levels, LO alignment, stem appropriateness, and technical item flaws. "AI-extended consensus" served as intersubjective consensus model (the gold standard). Generation performance was quantified by alignment with this consensus, and assessment performance by the degree to which Gen-AIs shifted or preserved Expert assessments. Analyses included ICC for reliability, Po/Cohen’s/Fleiss’ Kappa for categorical agreement, and inferential tests (Exact McNemar and Wilcoxon signed-rank) for detecting systematic bias and directional shifts.Results: Gen-AIs demonstrated markedly different performance patterns in assigning cognitive levels. For Miller, Gemini generated MCQs exhibited superior consistency with the intersubjective consensus (ICC(2,k)=0.82), whereas for Bloom, Copilot-generated MCQs demonstrated this superiority (ICC(2,k)=0.97). Both tools performed well in LO alignment and key identification, but their approaches to stem structure diverged substantially. Experts perceived the MCQs to be easier than the Gen-AIs claimed, and the current Gen-AI versions found them even easier than both the generating versions and the Experts did. In terms of assessment behaviour, Gen-AIs showed a systematic stringency tendency in Miller classifications, statistically significantly shifting Expert consensus from 'knows' to 'knows how' (p

This study investigates researcher variability in computational reproduction, an activity for which it is least expected. Eighty-five independent teams attempted numerical replication of results from an original study of policy preferences and immigration. Reproduction teams were randomly grouped into a ‘transparent group’ receiving original study and code or ‘opaque group’ receiving only a method and results description and no code. The transparent group mostly verified original results (95.7% same sign and p -value cutoff), while the opaque group had less success (89.3%). Second-decimal place exact numerical reproductions were less common (76.9 and 48.1%). Qualitative investigation of the workflows revealed many causes of error, including mistakes and procedural variations. When curating mistakes, we still find that only the transparent group was reliably successful. Our findings imply a need for transparency, but also more. Institutional checks and less subjective difficulty for researchers ‘doing reproduction’ would help, implying a need for better training. We also urge increased awareness of complexity in the research process and in ‘push button’ replications.

Objective: The purpose of this research is to evaluate e-PBL tutorials and compare these sessions with face-to-face PBL sessions. Patients and Methods: This research is a program evaluation study in which quantitative methods were used. In the research, four 90-minute e-PBL sessions held between April and June 2020 were evaluated. Sessions conducted online were realised using the sevenstep approach in groups of 13-14 students and a tutor. Video recordings were analysed with the thin slicing method. In addition, various quantitative data on evaluation were analysed using multiple tools and materials, including the end-of-program evaluation form. Friedman test and Mann-Whitney U test were used in the analyses of quantitative data. Results: Upon evaluating the analyses of the feedback received from the students about the structure, content and process of the program regarding the e-PBL tutorials, the students gave a positive opinion of 80% or more. In regards with the cases, the students had positive opinions of over 80% in terms of “motivation for learning and researching”, “daily life and its relation to their individual development”, “suitability to their levels of knowledge and skills”, “reinforcement of topics”. Support, guidance and feedback received from the tutor as a group and individually during online tutorials were statistically significantly higher than the face-to-face PBL tutorials (P<0.05). Conclusion: Research on the effectiveness of e-PBL tutorials, including ours, point out that e-PBL practices may constitute a viable alternative besides face-to face ones. However, for a sounder framing and better results, the subject should be studied in different aspects and more evidences be gathered in this area. These studies will provide evidence to educational institutions and practitioners on how to adapt and modify educational practices, including PBL.

Purpose We investigated the relationship between hearing aid or cochlear implant use and self-esteem, quality of life, hearing quality, and speech perception. Methods Of the 120 participants studied, 29 were cochlear implant users, 26 were hearing aid users, 33 were non-device users with hearing loss, and 32 had normal hearing. Each individual included in the study WHOQOL–BREF Quality-of-Life Scale, Rosenberg Self-Esteem and Speech, Spatial Perception and Qualities of Hearing Scale was applied. Results We found that the self-esteem level of the group with hearing loss, but without a device, was significantly lower than that of the group with a cochlear implant/hearing aid. There was a moderately statistically significant relationship between self-esteem, quality of life, speech perception, and hearing quality. Conclusions Using hearing aid and cochlear implants increases self-esteem among people with hearing impairments. Self-esteem is associated with quality of life, speech perception, and hearing quality.

Purpose To investigate development of radiation pneumonitis (RP) in relation to pulmonary function, dosimetric factors, and transforming growth factor beta-1 (TGFβ1) expression in irradiated breast cancer patients. Methods A total of 49 breast cancer patients who received post-operative radiotherapy (RT) were evaluated in terms of pulmonary function tests (PFTs), quality of life (QoL), development of RP, dosimetric factors, cytokine levels, and lung high-resolution computed tomography (HRCT) before and after RT. ROC analysis was performed for performance of dosimetric factors in predicting RP, while frequencies of single nucleotide polymorphisms (SNPs) genotyped for TGFβ1 (rs11466345 and rs1800470) were also evaluated. Results All cases with RP (10.2%) recovered clinically at the end of third post-RT month. PFT and HRCT parameters were similar before and after RT overall, as well as by RP and the radiation field subgroups. ROC analysis revealed the significant role of the ipsilateral V5 (cutoff value of 45.9%, p = 0.039), V10 (29.4%, p = 0.015), V20 (23%, p = 0.017), and MLD (1200 cGy, p = 0.030) in predicting RP. Higher post-RT TGFβ1 levels (p = 0.037) were noted overall and in patients with RP. Patient and control groups were similar in terms of frequencies of SNPs genotyped for TGFβ1 (rs11466345 and rs1800470). EORTC QLQ-C30 and QLQ-BR-23 scores were similar in patients with vs. without RP. Conclusion Our findings revealed significant role of dosimetric factors including MLD, V20 as well as the low dose-volume metrics in predicting the risk of RP among breast cancer patients who received post-operative RT. Implementation of RT, extent of radiation field or the presence of RP had no significant impact on PFTs.

Identification of wood species with fast, reliable and non-destructive methods is highly important for forestry and wood-related industries. Near-infrared spectra of anatomically similar pine species (Pinus sylvestris L. and Pinus nigra J.F. Arnold) were taken and analysed by partial least squared discriminant analysis (PLS-DA) for comparing the efficiency of preprocessing methods. Raw data were subjected to multiple scatter correction (MSC), standard normal variate (SNV), Savitzky–Golay for derivatives (1st and 2nd Dr) and smoothing (Sm) and combination of these preprocessing methods (1st Dr, 1st Dr + SNV, 1st Dr + MSC, Sm + 1st Dr and Sm + 2nd Dr). The success of the models was determined by the accuracies of test sets that did not participate in the calibration phase. In this study, it was determined that not all the preprocessing methods improve the model performance. Smoothing with 1st derivatives (Sm + 1st Dr) enhanced 14.3% improvement and have the best performance (95%) for classification of pine species. For understanding modelled relationship, mean spectra and selectivity ratio were used. It was found that discrimination was held by the differences at their absorption, and the most important variables for wood classification were noted around 4000–7000 cm⁻¹. GRAPHICAL ABSTRACT

Background: Metabolic syndrome (MS) is a condition that consists of several disorders, and the individual impact of these disorders on metabolic dysfunction-associated fatty liver disease (MAFLD) is still not clear in a combined diagnosis of MS. In this study, we aimed to investigate the effect of MS on advanced fibrosis in patients with MAFLD. Methods: We recruited the patients from our gastroenterology out-patient clinic who were being followed up for MAFLD. MAFLD was diagnosed with liver biopsy in all patients. The frequency of MS and other metabolic parameters were also compared between groups with advanced fibrosis and groups in which fibrosis was not as advanced. Results: In total, we enrolled 424 biopsy-proven MAFLD patients to the study. In univariate analysis, individuals with greater age, body mass index (BMI), higher aspartate transaminase (AST), MS, impaired fasting glucose, hypertension, enlarged waist circumference (WC), diabetes mellitus (DM), and women had significantly increased risk for fibrosis. In multivariate analysis, it was found that DM, greater age, higher BMI, and increased AST were seen more commonly in MAFLD patients with advanced fibrosis Conclusion: Greater age, a higher BMI, higher AST and a diagnosis of diabetes were more commonly associated with advanced fibrosis. However, DM was found to be the strongest predictive factor of advanced fibrosis in our cohort (OR: 2.495). Multivariate analyses did not indicate a significantly common occurrence of MS in the advanced fibrosis group, despite its important role in MAFLD pathophysiology.

Aim: We can define clinical reasoning as the ability to think and decide in clinical processes, which is non-reductive, multidimensional and free from prejudices. Clinical reasoning is a basic competence that a physician should have. In this article, theories about clinical reasoning, stages of clinical reasoning, training and evaluation methods, and the relationship between clinical reasoning and medical malpractice are discussed and related evidence is presented.Methods: The article was prepared by the traditional review methodResults: Non-analytical (system 1), analytical (system 2) thinking and dual process theories are accepted among the theories of how clinical reasoning is realized. Non-analytical thinking is an intuitive thinking that takes place fast and does not require much mental effort. Analytical thinking (system 2 thinking) is established by establishing a cause-effect relationship. In clinical reasoning, the stages of gathering information, forming hypothesis and testing hypothesis constitute a general framework. For clinical reasoning education, frameworks / models for this competence should be determined in educational activities, assessment and evaluation from the first years, and their use should be encouraged. Medical errors and inadequate clinical reasoning processes are related. For this reason, it is important to consider the clinical reasoning insufficiency areas in educational approaches. Clinical reasoning assessments should cover all stages of reasoning. Numerous methods and question / inquiry frameworks have been identified in the evaluation of clinical reasoning. In these, the basic principles are to ask questions that require reasoning with guides structured in accordance with the clinical reasoning stages.Conclusions: Clinical reasoning ability is the basic competence that a physician should have in order to ensure patient safety by reducing medical errors, to reduce unnecessary work and procedures in health, to prevent patient physician disputes and therefore violent events. Learning and assessment activities should be planned using models suitable for clinical reasoning terms and stages.