Cognitive Domain Assessment of Artificial Intelligence Chatbots: A Comparative Study Between ChatGPT and Gemini's Understanding of Anatomy Education.

Cognitive Domain Assessment of Artificial Intelligence Chatbots: A Comparative Study Between ChatGPT and Gemini’s Understanding of Anatomy Education - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Search Log in Primary site navigation Search Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice Med Sci Educ . 2025 Feb 15;35(3):1295–1304. doi: 10.1007/s40670-025-02303-0 Cognitive Domain Assessment of Artificial Intelligence Chatbots: A Comparative Study Between ChatGPT and Gemini’s Understanding of Anatomy Education Arthi Ganapathy 1 , Parul Kaushal 1, ✉ Author information Article notes Copyright and License information 1 Department of Anatomy, Teaching Block, All India Institute of Medical Sciences, New Delhi, 110029 India ✉ Corresponding author. Accepted 2025 Jan 27; Collection date 2025 Jun. © The Author(s) under exclusive licence to International Association of Medical Science Educators 2025. Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law. PMC Copyright notice PMCID: PMC12228882  PMID: 40625950 Abstract Purpose The integration of AI chatbots into education has gained traction, particularly in medical fields such as anatomy. This study aims to evaluate and compare the responses of ChatGPT 4o mini and Gemini across different cognitive domains of anatomy education. Methods A cross-sectional study was conducted to assess the responses of these two AI chatbots to a set of anatomy questions selected from the Manual on Competency-Based Undergraduate Curriculum. Questions were categorized into knowledge, comprehension and application levels of cognitive domain. Responses were scored against an answer key prepared by anatomy experts. Relevant comparative statistical analysis was performed. Results The overall performance of ChatGPT 4o mini (76.15%) was significantly superior to Gemini (72.84%). In application-level questions, ChatGPT 4o mini outperformed Gemini. Conversely, Gemini scored higher in comprehension-level questions (76.88% vs. 73.66%). Both chatbots exhibited factual inaccuracies and limitations in contextually accurate responses, particularly in application-level questions. Conclusion Both ChatGPT 4o mini and Gemini demonstrate potential as educational tools in anatomy, with strengths and limitations varying by cognitive domain. While AI chatbots can supplement traditional learning methods, they require ongoing refinement and validation. To ensure the responsible integration of AI into medical education, close attention must be devoted to faculty and student training, setting up relevant IT environment and ethical issues. Future research should focus on expanding question pools, incorporating user feedback and comparing with traditional educational approaches to enhance their effectiveness. Keywords: AI chatbots, Anatomy education, Cognitive domains, ChatGPT, Gemini, Educational technology, Gross anatomy Introduction Generation Z (Gen Z) has grown up in an era of rapid technological advancements, with digital devices and the Internet being an integral part of their lives. Exposure to technology has influenced their learning preferences, communication styles and expectations regarding education [ 1 ]. A survey by Salesforce on 4000 volunteers revealed that 70% of Gen Z used new generative artificial intelligence (AI) technologies as it saved time (46%), was easy to use (42%), taught new things (35%) and helped with complex concepts (32%). Gen Z students with technoholic mindset rely on digital technology for independent learning at their own pace [ 2 ]. Recently developed cognitive AI models like ChatGPT (chat generative pre-trained transformer: developed by Open AI) and Gemini (developed by Google AI) have become popular owing to their revolutionary capacity of retrieving and understanding unstructured data, to generate human like text, i.e. natural language conversations (NLC). A recent survey conducted amongst 6000 students across Swedish universities revealed a high level of awareness and usage of ChatGPT amongst the student population. More than half of the respondents expressed positive attitudes toward the integration of chatbots in education, revealing a growing acceptance of AI-driven technologies in academic settings [ 3 ]. These data suggest that students are not only receptive to emerging technologies but are also actively utilizing them to enhance their personal learning and knowledge acquisition. Both these AI models are still under development, and the potential of these platforms is being investigated in various fields, including medicine by exploring their knowledge, capacity to analyse clinical data, support diagnosis, provide therapeutic advice and assist in continuous training of healthcare professionals [ 4 ]. Anatomy is a keystone of medical education by offering a stage of knowledge suitable to all medical careers. This knowledge aids in patient examination, formation of diagnosis and efficient communication of medical conditions to the patient. Recently, newer teaching–learning methodologies are being introduced in teaching anatomy. The students are exploring the content available on the Internet and various e-learning platforms [ 5 ]. Much debate has arisen on the reliability of the content available online and the extent to which it caters to the core areas delineated by the medical councils of various countries. The main limitation of using these AI chatbots is that they have not yet been validated by subject experts, resulting in erroneous and inconsistent answers [ 6 ]. Yet, researchers have tried to validate the accuracy and adequacy of the information given by these chatbots. Various studies conducted on anatomy education have analysed the efficacy of these AI chatbots in answering multiple-choice questions (MCQs), generating MCQs and writing articles on anatomy topics [ 6 – 11 ]. As these chatbots are a type of cognitive computing that uses NLC and machine learning (ML) to understand and generate human-like text, the evaluation of their performance in cognitive domain is essential. Blooms cognitive taxonomy represents different domain levels giving a precise understanding of the level of knowledge a student has gathered [ 12 ]. Hence, the present study analysed the performance of two cognitive AI chatbots, ChatGPT 4o mini and Gemini through a validated set of anatomy questions focusing on various levels of cognitive domains such as knowledge, comprehension and application. Materials and Methods We conducted an observational cross-sectional study from 1st July 2024 to 25th August 2024, to assess the responses of two AI chatbots ChatGPT 4o mini and Gemini to gross anatomy questions. As no human subjects were involved, ethical approval was not required. Data Acquisition The questions were carefully selected from the Manual on Competency-Based Undergraduate Curriculum for the Indian Medical Graduate, which is implemented across all medical colleges in India under the guidelines of the National Medical Council (NMC). This manual provides standardized content and assessment guidelines for medical education in the country [ 13 ]. The selection criteria focused exclusively on questions classified as core competencies and “must know” areas (deemed essential for assessing the foundational knowledge in gross Anatomy), as emphasized in the manual. A total of 25 questions were selected. Among them, the distribution of questions for comprehension- and application-based was more, as compared to knowledge-based questions. This was in alignment with NMC curriculum for undergraduate anatomy, which emphasizes more on comprehension and application-based questions, besides including a foundation of knowledge-based questions. Table 1 presents the stems of the questions selected, the gross anatomy region they pertained to and the domain level they assessed. Table 1. Showing the gross anatomy questions administered to the AI chatbots, the topics covered and the cognitive domain level evaluated S. No Questions Topics Domain 1. Enumerate peculiarities of clavicle Upper limb Knowledge 2. Enumerate components of lacrimal apparatus Head and neck Knowledge 3. Enumerate the sites of portosystemic anastomosis Abdomen Knowledge 4. List components of nervous tissue General anatomy Knowledge 5. Enumerate parts of skeletal muscle General anatomy Knowledge 6. Describe parts, blood and nerve supply of long bone General anatomy Comprehension 7. Describe the structure of hyaline cartilage and its distribution in the body (any 2) Upper limb Comprehension 8. Describe parts of a neuron and classify them based on number of neurites, size & function Upper limb Comprehension 9. Describe factors maintaining medial longitudinal arch of the foot with its importance Upper limb Comprehension 10. Describe origin, course, relations and branches of a typical intercostal nerve Thorax Comprehension 11. Describe the formation, course, tributaries and termination of coronary sinus Thorax Comprehension 12. Describe movements of thumb and muscles involved Upper limb Comprehension 13. Describe the age changes in the mandible Head and neck Comprehension 14. Describe the plane regions and quadrants of abdomen Abdomen Comprehension 15. Describe the muscles of pelvic diaphragm Pelvis Comprehension 16. Explain functional difference between elastic, muscular arteries and arterioles General anatomy Application 17. Explain anatomical basis of injury to axillary nerve during intramuscular injections Upper limb Application 18. Describe the anatomical basis of venepuncture of cubital veins Upper limb Application 19. Explain the anatomical basis of locking and unlocking of the knee joint Lower limb Application 20. Explain surgical importance of deep facial vein Head and neck Application 21. Explain the anatomical basis of inguinal hernia Abdomen Application 22. Explain anatomical basis of Erb’s palsy Upper limb Application 23. Explain anatomical basis of varicose veins Lower limb Application 24. Explain the concept of “Peripheral heart” Lower limb Application 25. Explain the concept of nerve supply of joints and Hilton’s law General anatomy Application Open in a new tab Analysing Chatbots Responses Each question was administered simultaneously to the two AI chatbots by the authors (AG and PK). The questions selected from the Manual on Competency-Based Undergraduate Curriculum for the Indian Medical Graduate were prompted without any grammatical or language alteration in the respective chatbots. To reduce bias during statistical evaluation, the same timeline of question administration was maintained as these AI chatbots are dynamically evolving and continuously learning large language models. To ensure the chatbots were not influenced by any previous activity, the administration of the questions was done using newly created accounts. The responses given by the chatbots in their first attempt were copied in a file and then evaluated (data is shared as a supplementary file). Scoring System A detailed answer for each question was meticulously prepared by two experienced anatomists (AG and PK) by consulting standard anatomy textbooks [ 14 – 16 ]. The answer comprised of key points that must be included in the responses. For each key point in the answer, one mark was allotted; hence, the total score for one question was sum total of the marks, i.e. number of key points. Investigator triangulation was done to obtain the final score for each question. This was derived by calculating the mean of the scores assigned by the two anatomists, ensuring an objective and standardized evaluation. The responses of the chatbots were compared with the ideal answers prepared by the experts and also amongst themselves. Statistical Analyses The statistical analysis was conducted using GraphPad Prism version 10. Initially, the suitability of the data for normal distribution was assessed using Q-Q plots and the Shapiro–Wilk test. Since, the data was not normally distributed; hence, non-parametric methods were employed for further analysis. Descriptive statistics, including measures of central tendency, were presented as median and interquartile range (IQR), along with the estimation of confidence intervals to summarize the AI chatbots’ performance. For the comparison of responses between the ChatGPT and Gemini, the Wilcoxon signed-rank test was used, as it is appropriate for paired, non-normally distributed data. This test allowed for the evaluation of significant differences between the responses of the two AI chatbots. Additionally, the Friedman test was used to compare the AI chatbots’ responses with the expert evaluation scores, providing insights into how their performance aligned with the expected standards. The threshold for statistical significance was set at a p -value of < 0.05. Results Out of the 25 questions asked, 5 questions evaluated the knowledge domain, while 10 questions each evaluated the comprehension and the application domains. The scores obtained by ChatGPT and Gemini as compared to expert evaluation scores in response to all the questions are illustrated in Fig. 1 . The minimum and maximum scores obtained by the two AI chatbots, along with the median and interquartile range (IQR), are presented in Figs. 2 and 3 . Fig. 1. Open in a new tab Bar graph representing the scores recorded for responses of Gemini and ChatGPT against the expert evaluation scores Fig. 2. Open in a new tab Box and Whiskers plot showing comparative analysis of A overall scores obtained by ChatGPT and Gemini against the expert evaluation scores; B , C scores obtained in the knowledge domain questions by Gemini and ChatGPT, respectively, when compared to expert evaluation scores; D knowledge domain scores between Gemini and ChatGPT. Median and interquartile range along with the minimum and maximum scores are also represented. **** p < 0.0001 Fig. 3. Open in a new tab Box and Whiskers plot showing comparative analysis of A , B comprehension domain scores for Gemini and ChatGPT with expert evaluation scores, respectively; C comprehension domain values of inter AI chatbot comparison; D , E application domain scores for Gemini and ChatGPT with expert evaluation score; F application domain scores of inter AI chatbot comparison. Median and interquartile range along with the minimum and maximum scores are also represented. ** p < 0.001; C, comprehension; A, application When compared to the answers prepared by the expert anatomists, Gemini scored 72.84%, while ChatGPT4o mini scored 76.15%. Both the scores were significantly lower statistically ( p < 0.0001) as compared to the answers prepared by the expert anatomists (Fig. 2 A). The domain level breakdown of responses showed that Gemini had scored 76.08% in knowledge, 76.88% in comprehension and 53.57% in application levels. ChatGPT 4o mini gave a score of 78.26% in knowledge, 73.66% in comprehension and 83.92% in application domain. To estimate the cognitive level understanding of the AI chatbots, an extensive analysis of their responses to knowledge-, comprehension- and application-based questions was carried out. A comparison of the knowledge-based questions did not reveal any statistically significant difference between the responses of the AI chatbots and the expert evaluation scores (Fig. 2 B, C , D ). On the other hand, statistically significant ( p < 0.001) differences were noted in the comprehension (Fig. 3 A, B ) and application-level questions (Fig. 3 D, E ) amongst the AI chatbots and expert evaluation scores. The scores obtained in both these domains by Gemini and ChatGPT 4o mini were significantly lower statistically as compared to the expert evaluation ( p < 0.001). Gemini scored significantly higher statistically ( p < 0.001) in comprehension domain questions (Fig. 3 C) and ChatGPT 4o mini scored higher in application domain questions (Fig. 3 F). The accuracy of the responses was also analysed. Factually incorrect information was recorded in the response of both the AI chatbots for the question on describing origin, course, relations and branches of a typical intercostal nerve (Fig. 1 , question number 10). Both ChatGPT and Gemini elaborated the details of all 12 thoracic nerves though 3rd to 6th thoracic intercostal nerves are only considered as typical intercostal nerves (Fig. 4 ). Fig. 4. Open in a new tab Screenshot representations of factually incorrect responses generated by ChatGPT and Gemini For the question on: “Explain the anatomical basis of locking and unlocking of the knee joint” (Fig. 1 , question number 19), Gemini gave a response that quadriceps muscle is responsible for unlocking the knee joint but the answer is popliteus (Fig. 5 ). Gemini presented with links on the related topics for some of the questions which were not a feature of ChatGPT. Fig. 5. Open in a new tab Screenshot representation of factually incorrect response S No 19 generated by Gemini Discussion Incorporating conversational agents, like chatbots, into educational settings is an emerging practice that holds significant potential, though it has not yet become widespread [ 17 ]. Currently, chatbots can function as interactive tutors, providing personalized support to students [ 18 ]. They can act as advanced search engines, answering specific queries, offering real-time feedback and guiding learners through complex topics. This enables students to receive tailored assistance outside of traditional classroom settings, thereby fostering a more individualized learning experience. As the technology continues to evolve, chatbots are likely to become an integral part of educational strategies [ 19 , 20 ]. While AI-powered chatbots have the capability to analyse vast amounts of medical literature, they may not always accurately interpret, contextualize and apply the findings in a meaningful way. This limitation raises important questions about the validity and reliability of using AI chatbots across different academic subjects. Recognizing this, we aimed to systematically evaluate the performance of AI chatbots in addressing cognitive domain evaluation questions related to Anatomy. A comprehensive overview of previous studies that have evaluated AI chatbots in delivering anatomy knowledge across different formats is provided in Table 2 [ 7 – 11 , 21 ]. Table 2. Major studies and their notable findings in use of AI chatbots in anatomy education Investigator AI model Asked questions related to: Notable findings Totlis et al. (2023) [ 7 ] ChatGPT a. Anatomy knowledge b. Generate MCQs c. Anatomy research d. Anatomical variants e. Clinical anatomy • Provides detailed & organized descriptions of anatomy • Summaries chapters in bullet point format • Highlights the clinical significance of variants, but its replies are limited unless a variant is classified into types • Generates quizzes of varying difficulty levels, including MCQs and matching questions • Summarizes articles when presented with the text Ilgaz & Celik (2023) [ 8 ] ChatGPT and Google Bard a. Knowledge of anatomy by answering exam questions b. Generate MCQs c. Write articles on anatomy topics • No difference in problem-solving skills in anatomy amongst the two platforms • Google Bard better than ChatGPT in generating questions • Scientific writing ability of both chat boards was less detailed; however, ChatGPT article was more in line with academic norms Mogali (2023) [ 9 ] ChatGPT a. Fact-based questions b. Review of papers on particular topic • Results had major errors with respect to few questions • Inconsistent in providing accurate results • Detailed review was presented, but it was not restricted to the specific query Leng (2024) [ 10 ] ChatGPT a. Generating questions and PBL • Generates questions; however, few incorrect questions were also generated; hence, it was suggested that the questions be verified • Writes cases to assist teaching of PBL; however, these questions need some refinement by the teachers Pal & Sankarasubbu (2024) [ 11 ] ChatGPT-4 & Gemini Pro a. Knowledge of Anatomy and other medical subjects • Gemini Pro scored lower (76.9%) than ChatGPT-4 (81.8%) in anatomy and neuroanatomy • Low performance in cardiology, dermatology, forensic medicine as well; suggested: need for improvement in reliability and trustworthiness Collins et al. (2024) [ 21 ] AnatomyGPT and ChatGPT a. Anatomy National Board of Medical Examiners (NBME) anatomy questions • AnatomyGPT achieved high scores on NBME sample items for gross anatomy, embryology, histology and neuroscience and scored comparably to ChatGPT • AnatomyGPT provided several citations in the responses that it generated, while ChatGPT did not • Both GPTs provided rationales for all sample items Open in a new tab In some studies, ChatGPT was found to outperform other chatbots, while in other cases, the opposite was true. However, most of these studies focused on multiple-choice questions (MCQs) when conducting objective comparative statistical analyses. The evaluation of chatbots’ performance in responding to descriptive-type questions especially in the context of anatomy was not assessed in detail. This is a significant lacuna, as descriptive answers are critical in assessing different levels of knowledge and understanding in anatomy, a subject that demands not only factual recall but also the ability to explain complex concepts. A recent study developed a custom interactive chatbot, Anatbuddy, and compared it with ChatGPT 3.5. They compared the responses to 18 questions from the thoracic anatomy with respect to factual accuracy, relevance, completeness, coherence and fluency on a 5-point Likert scale and reported Anatbuddy to be more reliable in factual accuracy. No significant difference was noted in any other parameter considered [ 22 ]. In our study, we conducted a comparative analysis of the knowledge, comprehension and application-level competencies (Bloom’s taxonomy) of AI chatbots, comparing it with meticulously prepared answers with specific key points and marking them for factual accuracy and completeness. For the knowledge-based questions, which are primarily fact-based and require enumeration, both AI chatbots performed well, with their responses meeting the expected standards. However, a significant difference emerged when it came to descriptive and comprehension-based questions. In these areas, both chatbots’ performances fell short of expectations. ChatGPT provided better responses for analytical-type questions, highlighting its strength in more complex reasoning tasks. These findings align with the results of previous research by Ranjan and co-workers [ 23 ] which demonstrated that ChatGPT outperformed other chatbots in applied microbiology questions. However, for descriptive sections like general microbiology, Gemini was found to perform better. Similarly, Hussain and co-workers [ 24 ] evaluated the performance of ChatGPT and Google Bard on internal medicine MCQs. They found that ChatGPT excelled in C3-level cognition, particularly in problem-solving MCQs, whereas both chatbots showed similar performance in C2-level descriptive questions. Similar to findings from previous studies [ 9 , 10 ], we observed that while some of the chatbots’ responses included key points, they were excessively detailed. This level of elaboration will make it challenging for students to fully grasp the essential information. Some factual errors were also observed (Fig. 4 ). The study presented with certain limitations. AI models like ChatGPT and Gemini are still evolving, and their performance may vary over time and with updates; hence, in a temporal context, this may potentially affect the consistency of results. Additionally, we analysed only two AI tools, which limits the scope of our comparison. Precise drafting of the prompt plays a vital role in generating best responses from the chatbots; however, in this study, the stems were used as per the CBME curriculum; as the students would use it. This also highlights the need for sensitisation of students on effective use of the available AI tools. Finally, the free version of ChatGPT was utilized, which could have limited the extent of the responses. Conclusion and Future Prospects This study provides a comparative analysis of the cognitive domain assessment of two AI chatbots, ChatGPT 4o mini and Gemini, in the context of anatomy education. While both chatbots demonstrated strengths in delivering factual and knowledge-based responses, significant differences were observed in their performance across various cognitive levels. ChatGPT 4o mini excelled in application-level questions, whereas Gemini performed better in comprehension-type questions. Despite these strengths, both chatbots exhibited limitations in delivering accurate and contextually appropriate responses. Factual inaccuracies and inconsistent responses highlight the necessity for further refinement of these AI tools. The observations of this study and a meticulous review of previous published manuscript all point to one significant finding that while AI can provide a foundation of knowledge, the use of the AI chatbots can only augment the classical teaching–learning methods and cannot replace them. The human components like years of educational experience and the unique understanding of a teacher to individual student needs remain unparalleled. Based on the study’s findings, chatbots can be used by educators for generating relevant information quickly which saves time. Previous studies have shown that they can be used for quickly generating multiple-choice questions (MCQs), case studies and quizzes tailored to specific topics [ 24 – 26 ]. The suggestions from the present study are consistent with the previous research that emphasize meticulous scrutiny of the content generated by chatbots before delivering it to the students. To ensure the responsible integration of AI into medical education, close attention must be devoted to faculty and student training, setting up relevant IT environment and ethical issues. Establishing a culture of academic integrity and providing students with the tools they need to assess AI critically and use it responsibly while balancing it with their own distinct perspectives are important. Future studies could assess a broader range of questions covering additional topics and complexities in anatomy to provide a more comprehensive evaluation of chatbot performance. Conducting longitudinal studies to assess the impact of AI chatbots over time and their adaptability to changes in educational content and technology would provide insights into their long-term effectiveness. Further development of customized chatbots should be done along with subject experts to make the experience of using AI chatbots for student learning more effective. Author Contribution Arthi Ganapathy: Conceptualization; data curation; formal analysis; investigation; methodology; project administration; resources; software supervision; validation; visualization; writing, original draft; writing, review and editing. Parul Kaushal: Conceptualization; data curation; formal analysis; investigation; methodology; project administration; resources; software; supervision; validation; visualization; writing, original draft; writing, review and editing. Data Availability All data analysed during the study are included in this manuscript. The complete original datasets generated during the current study are available in the form of supplementary material. Declarations Declaration of Generative AI and AI-Assisted Technologies in the Writing Process During the preparation of this work, the authors used ChatGPT 4o mini and Gemini in order to analyse their responses to the questions asked and also to assist in write up. After using this tool/service, the authors reviewed and edited the content as needed and take full responsibility for the content of the published article. Ethical Approval As no human subjects were involved in the study, ethical approval was not required. Competing Interests The authors declare no competing interests. Footnotes Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References 1. Seemiller C, Grace M. Generation Z: educating and engaging the next generation of students. About Campus. 2017;22(3):21–6. 10.1002/abc.21293. [ Google Scholar ] 2. Iftode D. Generation Z and learning styles. 2020. 10.2139/ssrn.3518722. 3. Stohr C, Ou AW, Malmström H. Perceptions and usage of AI chatbots among students in higher education across genders, academic levels and fields of study. Comp Educ: Art Intell. 2024;7:100259. 10.1016/j.caeai.2024.100259. [ Google Scholar ] 4. Peng C, Yang X, Chen A, et al. A study of generative large language model for medical research and healthcare. Npj Digit Med. 2023;6:210. 10.1038/s41746-023-00958-w. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 5. Leung BC, Williams M, Horton C, Cosker TD. Modernising anatomy teaching: which resources do students rely on? J Med Educ Curric Dev. 2020. 10.1177/2382120520955156. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 6. Omar R, Mangukiya O, Kalnis P, Mansour E. ChatGPT versus traditional question answering for knowledge graphs: current status and future directions towards knowledge graph chatbots. Available from: https://arxiv.org/abs/2302.06466 . 7. Totlis T, Natsis K, Filos D, Ediaroglou V, Mantzou N, Duparc F, Piagkou M. The potential role of ChatGPT and artificial intelligence in anatomy education: a conversation with ChatGPT. Surg Radiol Anat. 2023;45(10):1321–9. 10.1007/s00276-023-03229-1. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 8. Ilgaz HB, Çelik Z. The significance of artificial intelligence platforms in anatomy education: an experience with ChatGPT and Google Bard. Cureus. 2023. 10.7759/cureus.45301. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 9. Mogali SR. Initial impressions of ChatGPT for anatomy education. Anat Sci Educ. 2019;17(2):444–7. 10.1002/ase.2261. [ DOI ] [ PubMed ] [ Google Scholar ] 10. Leng L. Challenge, integration, and change: ChatGPT and future anatomical education. Med Educ Online. 2024;29(1):2304973. 10.1080/10872981.2024.2304973. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 11. Pal A, Sankarasubbu M. Gemini goes to med school: exploring the capabilities of multimodal large language models on medical challenge problems & hallucinations. Available from: https://arxiv.org/abs/2402.07023 . 12. Anderson LW, Krathwohl DR, Airasian PW, et al. A taxonomy for learning, teaching, and assessing: a revision of Bloom’s taxonomy of educational objectives. 1st ed. New York: Longman; 2001. [ Google Scholar ] 13. AETCOM. Attitude, ethics and communication. Competencies for the Indian Medical Graduate 2018. https://www.nmc.org.in/information-desk/for-colleges/ug-curriculum/ . Accessed 2 Jul 2024. 14. Standring S. Gray’s Anatomy: the anatomical basis of clinical practice. 42nd ed. London: Churchill Livingstone; 2021. [ Google Scholar ] 15. Mehta V. Snell’s clinical anatomy by regions. 2nd South Asian ed. New Delhi: Wolters Kluwer; 2023. 16. Moore KL, Dalley AF, Agur A. Clinically oriented anatomy. 8th ed. Philadelphia: Lippincott Williams and Wilkins; 2017. 17. Okonkwo CW, Ade-Ibijola A. Chatbots applications in education: a systematic review. Comp Educ: Art Intell. 2021;2:100033. 10.1016/j.caeai.2021.100033. [ Google Scholar ] 18. Labadze L, Grigolia M, Machaidze L. Role of AI chatbots in education: systematic literature review. Int J Educ Technol High Educ. 2023;20:56. 10.1186/s41239-023-00426-1. [ Google Scholar ] 19. Sáiz-Manzanares MS, Marticorena-Sánchez R, Martín-Antón LJ, Díez IG, Almeida L. Perceived satisfaction of university students with the use of chatbots as a tool for self-regulated learning. Heliyon. 2023;9(1):e12843. 10.1016/j.heliyon.2023.e12843. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 20. Ghorashi N, Ismail A, Ghosh P, Sidawy A, Javan R. AI-powered chatbots in medical education: potential applications and implications. Cureus. 2023. 10.7759/cureus.43271. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 21. Collins BR, Black EW, Rarey KE. Introducing AnatomyGPT: a customized artificial intelligence application for anatomical sciences education. Clin Anat. 2024;37(6):661–9. 10.1002/ca.24178. [ DOI ] [ PubMed ] [ Google Scholar ] 22. Arun G, Perumal V, Urias FPJB, Ler YE, Tan BWT, Vallabhajosyula R, Tan E, Ng O, Ng KB, Mogali SR. ChatGPT versus a customized AI chatbot (Anatbuddy) for anatomy education: a comparative pilot study. Anat Sci Educ. 2024;17(7):1396–405. 10.1002/ase.2502. [ DOI ] [ PubMed ] [ Google Scholar ] 23. Ranjan J, Ahmad A, Subudhi M, Kumar A. Assessment of artificial intelligence platforms with regard to medical microbiology knowledge: an analysis of ChatGPT and Gemini. Cureus. 2024. 10.7759/cureus.60675. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 24. Husain S, Shahid S, Ansari Z, Ayoob T, Hussain A, Mujahid R. To compare the efficiency of ChatGPT and bard in medical education: an analysis of MCQ-based learning and assessment. Ann Abbasi Shaheed Hosp Karachi & KMDC; 2024. 10.58397/ashkmdc.v29i1.782. 25. Kıyak YS, Emekli E. ChatGPT prompts for generating multiple-choice questions in medical education and evidence on their validity: a literature review. Postgrad Med J. 2024;100(1189):858–65. 10.1093/postmj/qgae065. [ DOI ] [ PubMed ] [ Google Scholar ] 26. Al Shuraiqi S, Aal Abdulsalam A, Masters K, Zidoum H, AlZaabi A. Automatic generation of medical case-based multiple-choice questions (MCQs): a review of methodologies, applications, evaluation, and future directions. BDCC. 2024;8(10):139. 10.3390/bdcc8100139. [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Data Availability Statement All data analysed during the study are included in this manuscript. The complete original datasets generated during the current study are available in the form of supplementary material. Articles from Medical Science Educator are provided here courtesy of Springer ACTIONS View on publisher site PDF (1.5 MB) Cite Collections Permalink RESOURCES Similar articles Cited by other articles Links to NCBI Databases Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top

Related documents

• Harnessing ChatGPT for Medical Education: A Narrative Review.#223037
• AI in the Classroom: Observing Preclinical Students' Use of ChatGPT During Case‐Based Learning at a UK Medical School#1053080
• Performance of ChatGPT on USMLE: Potential for AI-Assisted Medical Education Using Large Language Models#199562
• Performance of ChatGPT on USMLE: Potential for AI-assisted medical education using large language models#199308
• Performance of ChatGPT on USMLE: Potential for AI-assisted medical education using large language models#16568