Evaluating the strengths and weaknesses of large language models in answering neurophysiology questions.

Large language models (LLMs), like ChatGPT, Google's Bard, and Anthropic's Claude, showcase remarkable natural language processing capabilities. Evaluating their proficiency in specialized domains such as neurophysiology is crucial in understanding their utility in research, education, and clinical applications. This study aims to assess and compare the effectiveness of Large Language Models (LLMs) in answering neurophysiology questions in both English and Persian (Farsi) covering a range of topics and cognitive levels. Twenty questions covering four topics (general, sensory system, motor system, and integrative) and two cognitive levels (lower-order and higher-order) were posed to the LLMs. Physiologists scored the essay-style answers on a scale of 0-5 points. Statistical analysis compared the scores across different levels such as model, language, topic, and cognitive levels. Performing qualitative analysis identified reasoning gaps. In general, the models demonstrated good performance (mean score = 3.87/5), with no significant difference between language or cognitive levels. The performance was the strongest in the motor system (mean = 4.41) while the weakest was observed in integrative topics (mean = 3.35). Detailed qualitative analysis uncovered deficiencies in reasoning, discerning priorities, and knowledge integrating. This study offers valuable insights into LLMs' capabilities and limitations in the field of neurophysiology. The models demonstrate proficiency in general questions but face challenges in advanced reasoning and knowledge integration. Targeted training could address gaps in knowledge and causal reasoning. As LLMs evolve, rigorous domain-specific assessments will be crucial for evaluating advancements in their performance.

Prediction of Diabetes Using Data Mining and Machine Learning Algorithms: A Cross-Sectional Study.

OBJECTIVES: This study aimed to develop a model to predict fasting blood glucose status using machine learning and data mining, since the early diagnosis and treatment of diabetes can improve outcomes and quality of life. METHODS: This crosssectional study analyzed data from 3376 adults over 30 years old at 16 comprehensive health service centers in Tehran, Iran who participated in a diabetes screening program. The dataset was balanced using random sampling and the synthetic minority over-sampling technique (SMOTE). The dataset was split into training set (80%) and test set (20%). Shapley values were calculated to select the most important features. Noise analysis was performed by adding Gaussian noise to the numerical features to evaluate the robustness of feature importance. Five different machine learning algorithms, including CatBoost, random forest, XGBoost, logistic regression, and an artificial neural network, were used to model the dataset. Accuracy, sensitivity, specificity, accuracy, the F1-score, and the area under the curve were used to evaluate the model. RESULTS: Age, waist-to-hip ratio, body mass index, and systolic blood pressure were the most important factors for predicting fasting blood glucose status. Though the models achieved similar predictive ability, the CatBoost model performed slightly better overall with 0.737 area under the curve (AUC). CONCLUSIONS: A gradient boosted decision tree model accurately identified the most important risk factors related to diabetes. Age, waist-to-hip ratio, body mass index, and systolic blood pressure were the most important risk factors for diabetes, respectively. This model can support planning for diabetes management and prevention.

Ontology Engineering for Gastric Dystemperament in Persian Medicine.

OBJECTIVE: Developing an ontology can help collecting and sharing information in traditional medicine including Persian medicine in a well-defined format. The present study aimed to develop an ontology for gastric dystemperament in the Persian medicine. METHODS: This was a mixed-methods study conducted in 2019. The first stage was related to providing an ontology requirements specification document. In the second stage, important terms, concepts, and their relationships were identified via literature review and expert panels. Then, the results derived from the second stage were refined and validated using the Delphi method in three rounds. Finally, in the fourth stage, the ontology was evaluated in terms of consistency and coherence. RESULTS: In this study, 241 concepts related to different types of gastric dystemperament, diagnostic criteria, and treatments in the Persian medicine were identified through literature review and expert panels, and 12 new concepts were suggested during the Delphi study. In total, after performing three rounds of the Delphi study, 233 concepts were identified. Finally, an ontology was developed with 71 classes, and the results of the evaluation study revealed that the ontology was consistent and coherent. CONCLUSION: In this study, an ontology was created for gastric dystemperament in the Persian medicine. This ontology can be used for designing future systems, such as case-based reasoning and expert systems. Moreover, the use of other evaluation methods is suggested to construct a more complete and precise ontology.

The Effects of WW2/WW3 Domains of Smurf2 Molecule on CD4+CD25+/CD4+ Proportion in Spleen of 4T1 Tumor Bearing BALB/c Mice.

Background: TGF-ß has long been considered as the main inducer of Tregs in tumor microenvironment and is the reason for the aberrant number of Tregs in tumor-bearing individuals. Recently, it has been suggested that the enzyme arginase I is able to mediate the induction of Tregs in a TGF-ß-independent fashion. The recombinant WW2/WW3 domains from smad ubiquitination regulatory factor 2 molecule was demonstrated to increase TGF-ß signaling while reducing arginase I gene expression. In this study, we aimed to examine the effects of this recombinant protein on CD4+CD25+/CD4+ proportion in the spleen of 4T1 mammary carcinoma-bearing BALB/c mice. Methods: Flow cytometry was used to evaluate CD4+CD25+ spleen cell populations of the tumor-bearing mice that received WW2/WW3 protein treatment and those of the control group. Results: The results indicated a significant rise in CD4+CD25+/CD4+ ratio, along with an average increase in tumor mass of the subjects that underwent protein treatment. Conclusion: It can be inferred that the heightened CD4+CD25+/CD4+ proportion in the spleen of protein-treated tumor-bearing mice can be the result of the increased TGF-ß signaling despite the reduced arginase I expression.

Development and Evaluation of Ontologies in Traditional Medicine: A Review Study.

BACKGROUND: Development of ontologies in traditional medicine can be a foundation for other applications of informatics in this field. Despite the importance of the development of ontologies in traditional medicine, there are few review studies in this area. This study aims to review different methods for ontology development and evaluation in traditional medicine. METHODS: This review study was performed in 2019. To find related papers, six databases including Scopus, Web of Science, PubMed, Embase, IEEE Xplore, and SpringerLink were searched. Initially, 761 articles were identified. After applying inclusion and exclusion criteria, 22 articles were selected to review different methods for ontology development and evaluation in traditional medicine. RESULTS: Five different methods were used for ontology development in traditional medicine, namely conventional, customized, semiautomatic, upper-level, and large-scale methods. The results showed that ontology evaluation was only considered in 32% of the studies. The common methods used for ontology evaluation were competency questions, expert-based evaluation, and automatic detection of inconsistency errors. CONCLUSION: Development of ontologies is of high importance for organizing knowledge in traditional medicine, as this branch of medicine is often not documented or is documented in local languages. The results of this study can help ontology developers to be familiar with the common methods of ontology development and evaluation in traditional medicine and use them for future research.