Introduction

The ways in which healthcare professionals learn, both individually and collaboratively, have been transformed by technology-enhanced learning, particularly through the integration of artificial intelligence. Furthermore, AI has also changed how students are prepared for careers in healthcare, how research is conducted, and how performance is evaluated ( 1 - 7 ). Along with the educational challenges posed by using AI in the classroom, one challenge healthcare educators face is addressing ethical issues (e.g., data privacy, algorithm bias, and transparency). To successfully adopt AI, appropriate oversight and ethical guidelines are needed to support its ethical use by all stakeholders, minimize risks, and maximize potential benefits ( 8 , 9 ). Given the rapid pace of technological change, timely action is required to ensure that future healthcare professionals receive adequate education and training to use AI effectively ( 10 ). Therefore, it is essential that medical education programs and curricula adapt to the rapidly changing environments in which they operate.

The application of artificial intelligence in medical education has evolved rapidly through large language models (LLMs), which are systems capable of understanding and generating human-like natural language ( 11 ). LLMs are used to provide realistic clinical scenarios and constructive feedback to learners while creating individualized learning environments that accommodate each learner’s pace, style, and characteristics. Personalized learning emphasizes the development of learning processes that reflect individual learner characteristics, including abilities, preferences, prior knowledge, interests, and goals. It requires continuous and automated adaptation to ensure that learners’ knowledge and competency levels remain up to date ( 12 ).

Personalized learning originated in classical educational psychology and focused on tailoring instruction to each learner’s individual profile. Benjamin Bloom’s classic 1984 study, entitled the “2 Sigma Problem,” demonstrated that students who received individual, instructor-led tutoring performed, on average, two standard deviations higher than their non-tutored counterparts in traditional classroom settings. This work led to substantial subsequent research on adaptive instructional methods and the use of technology for instructional delivery. Barbara Means and other researchers have contributed to the body of research on personalized learning in medical education. Means demonstrated that blended learning and student-centered approaches were effective in digital learning environments ( 13 ). As additional empirical studies have been published, the positive impact of personalized learning on academic achievement, metacognitive awareness, and motivation has continued to be demonstrated ( 14 , 15 ). Taken together, this evidence points to the need to redesign medical education with the support of emerging AI capabilities.

Although active learning methods have been gradually adopted in medical education, educational systems still predominantly rely on a one-size-fits-all model. This can be problematic because medical students enter educational programs with varying levels of academic preparedness, learning styles, and prior knowledge. Because educational value added is closely related to learners’ background knowledge, these differences can result in learning gaps for some students and suboptimal experiences for others. This not only negatively affects the quality of the educational experience but also is in conflict with core principles of educational equity ( 16 ). Advances in AI, particularly those enabled by LLM-based technologies, present new opportunities for student-centered instructional adjustment and enhanced personalization. The purpose of this study is to identify the application of advanced learning technologies in creating personalized learning opportunities for medical students.

Research Contribution

Previous investigations examining the use of artificial intelligence to support individualized learning approaches among medical students have largely been conducted as independent and uncoordinated studies employing limited methodologies. Consequently, there is no single source that provides a comprehensive synthesis of the potential of AI-based personalization within medical education. Accordingly, the present study employs a meta-synthesis methodology to conduct a comprehensive and interpretive qualitative analysis of the existing literature on this topic. Following a systematic screening process and an exhaustive content analysis of the identified studies, this investigation delivers a multidimensional, contextually relevant, and action-oriented framework for the development and application of AI technologies to support effective personalized learning in medical education. By integrating disparate findings across the literature, this study contributes to a deeper understanding of the key concepts, emergent patterns, and practical recommendations for the design and implementation of adaptive medical education. As such, the findings provide a meaningful contribution to the interdisciplinary discourse at the intersection of medical education and educational technology.

Methods

This study was conducted in accordance with the four-phase framework proposed by Sandelowski and Barroso ( 17 , 18 ), encompassing a rigorous search strategy, critical appraisal of eligible studies, classification of the evidence, and synthesis of key findings. The combined use of systematic review and meta-synthesis provided a structured methodological approach to examine and formally integrate the findings of prior research ( 19 ), thereby generating cohesive insights grounded in accumulated empirical evidence.

Selection criteria

Articles that met the inclusion and exclusion criteria listed in Table 1 were analyzed for the meta-synthesis research.

Following the criteria for meta-synthesis outlined above, this study excluded non-primary research articles, conference abstracts, letters to the editor, books, and unpublished reports, as these sources do not meet the standards for rigorous integration. The search was further limited to English-language publications between 2010 and 2025, which may have resulted in the omission of relevant non-English studies. The exclusion of non-research and unpublished materials may also have omitted important perspectives typically included in grey literature.

Data Sources and Searches

An information specialist conducted a systematic literature review using multiple bibliographic databases. Databases searched included Medline, Embase, CINAHL, PsycINFO, PubMed, Web of Science, ScienceDirect, Wiley, SpringerLink, Taylor & Francis, and SAGE. The literature search included peer-reviewed articles published between January 1, 2010 and June 20, 2025, that provided evidence of the use of artificial intelligence in personalized learning. The bibliographic records were entered into an Excel spreadsheet and organized in a structured database to facilitate systematic screening and citation tracking. In this manner, the four members of the research team conducted an independent two-phase screening process (M-H, A-T, A-K, Z-F). The screening process consisted of an initial title and abstract review based on the predefined inclusion criteria of the study, followed by a full-text review for final inclusion. To enhance the quality of the review process, individuals outside of the original reviewer pool independently assessed each article during the quality appraisal stage. This approach provided an additional layer of verification and facilitated the resolution of differences in opinion regarding the article quality. Three major concept areas were used to develop a comprehensive systematic search strategy: medical education, learner-centered education, and artificial intelligence technology. Each concept area included multiple terms with similar meanings; therefore, terms within each area were combined using OR, while the three concept areas were combined using AND. To increase the precision of the search results, we applied limits to include only systematic review articles published between 2010 and 2025. The complete systematic search strategy, including medical subject headings and free-text terms, is presented in Table 2.

Further backward reference searching of the selected full texts was conducted to capture studies that may have been missed in the initial query. Any disagreements or discrepancies in the study selection were resolved through deliberation with a panel of three external reviewers, ensuring consensus and methodological transparency (Table 3). Figure 1 presents the PRISMA form in the article screening process.

Figure 1. The PRISMA form in the article screening process

Quality Appraisal

To evaluate the methodological quality of the included studies, we applied the Critical Appraisal Skills Programme (CASP). CASP provides a structured framework consisting of ten core questions designed to assess the credibility, relevance, and rigor of research. Building on the approach developed by Long, et al. ( 36 ) and previously employed in meta-synthesis studies ( 37 ), the CASP tool was operationalized into a 10-item checklist to enable a more granular and systematic appraisal. Two reviewers (A-K and M-H) independently conducted quality assessments using this expanded tool, with any discrepancies resolved through consultation with a third reviewer (A-T or Z-F). The ten CASP criteria included clarity of research aims; appropriateness of methodology; suitability of research design; adequacy of recruitment strategy; rigor of data collection; consideration of researcher–participant relationships; attention to ethical issues; rigor of data analysis; clarity of the findings; and overall value of the research. Each criterion was scored on a 0–4 scale (0 = not addressed; 4 = fully addressed), yielding a maximum possible score of 40 points. A cut-off score of 30 points (≥75%) was established as the minimum threshold for methodological adequacy; studies scoring below this threshold were excluded. As a result, all 16 included studies achieved scores above the threshold and were positively appraised ( 38 ).

Analysis and Synthesis

Data were interpreted using an inductively oriented text analytic approach ( 39 ). Following the extraction and transcription of relevant sections from each identified article, multiple read-throughs were conducted to ensure a general understanding of the primary messages of the articles. The entire article served as the unit of analysis; therefore, the data were deconstructed into condensed meaning units ranging from single words to complete sentences or paragraphs that represented thematically related concepts. Condensed meaning units were then consistently interpreted and assigned descriptive codes. Subsequently, codes with similar conceptual themes were compared and clustered to create preliminary sub-themes through thematic aggregation. This iterative process of coding and theme review continued until theme saturation was achieved, defined as the point at which no additional codes, categories, or insights emerged from the data. Theme saturation supported the validity and completeness of the analytical framework by confirming that the final set of themes represented the full scope of evidence across the included studies.

Data Validation

The reliability of the coding process was evaluated using an internal consistency test applied to all conceptual elements generated during analysis. A validation strategy was implemented through a repeated categorization process conducted independently by a single researcher at two separate time points. By comparing the results from both coding rounds, the consistency of code assignment was assessed. The codes that matched across both rounds were classified as “matched,” whereas those that did not were classified as “unmatched.” A ratio of matched to unmatched codes was subsequently calculated and used to determine test–retest reliability. In this study, the interval between coding rounds was 20 days, and both rounds were conducted by the same researcher. A total of 331 distinct codes were generated, of which 135 were assigned consistently across both coding rounds. Based on these results, a reliability coefficient of 81% was calculated, exceeding the accepted standard of 60% for test–retest reliability ( 40 ). Accordingly, the coding framework demonstrated stability and credibility, further supported by peer review and evaluation by external experts not affiliated with the research team ( 41 - 44 ).

$\mathrm{Test-Retest\; Reliability}=100\times \left(\frac{135\times 2}{331}\right)=\mathrm{81\%}$

Results

Through the meta-synthesis, we identified five main categories (and related sub-categories): (i) applications for faculty members; (ii) applications for students; (iii) applications in the learning process; (iv) curriculum development; and (v) development of student assessment mechanisms. Table 4 presents a detailed description of all main themes and subthemes.

The dataset created from the previously described classifications of extracted data was assembled into five overarching themes, yielding sixteen sub-themes and 192 key concepts that define multiple components of personalized learning in medical education.

The first overarching theme, “Applications for Faculty Members,” consists of five sub-themes with 41 key concepts. These concepts address topics such as aligning instructional delivery with varied learning styles, utilizing data to design instruction, and supporting faculty in managing instructional processes to enhance teaching effectiveness.

The second theme, “Applications for Students,” includes four sub-themes and the largest number of key concepts in this analysis, totaling 53. These concepts cover diverse aspects of student development, including individual learning trajectories, cognitive and non-cognitive skill development, self-regulation, and support for informed academic and career decision-making.

The third theme, “Applications in the Learning Process,” comprises two sub-themes with 28 key concepts. This theme highlights the transformative role of adaptive technologies, the use of realistic clinical case scenarios, and the creation of multisensory and responsive learning environments as essential elements in reshaping the learning process.

The fourth theme, “Curriculum Development,” includes two sub-themes and 25 key concepts. It emphasizes flexible content, modular and non-linear curriculum structures, and the involvement of learners as co-designers of the curriculum, promoting ownership and deeper engagement with the educational material.

The fifth theme, “Development of Student Assessment Mechanisms,” consists of three sub-themes and 45 key concepts. These focus on data-driven analytics, adaptive assessment strategies, context-aware assessment, and dynamic feedback mechanisms to improve the accuracy, efficiency, and equity of student evaluation.

Overall, this analytical framework demonstrates that individualized learning impacts all aspects of the educational system—from curriculum to pedagogy, assessment, and learner experience—creating innovative ways to advance medical education. These results are presented in Figure 2.

Figure 2. Visual summary of findings

The relative distribution of the main themes identified in this review is shown in Figure 2. The largest proportion (28%) is associated with Applications for Students, reflecting the importance of a student-centered approach to personalized learning, as demonstrated by adaptive learning pathways, cognitive-emotional development, and goal-driven self-regulation. The second largest share (23%) is attributed to Development of Student Assessment Mechanisms, reflecting the increasing use of data-driven, real-time, differentiated evaluation methods. Applications for Faculty Members (21%) highlight the pedagogical shift toward personalized instructional design and adaptive teaching practices supported by learner analytics. Applications related to the Learning Process (15%) and Curriculum Development (13%) represent smaller portions of the distribution but remain important, demonstrating the development of responsive learning environments and flexible curricular structures.

Overall, the distribution of these categories illustrates that personalized learning is not limited to learner autonomy; it also requires a systemic reconfiguration of instructional roles, assessment practices, and curricular design, factors with significant implications for transforming the medical education system.

Discussion

Since AI began to be integrated into medical education, there has been a shift toward more personalized, adaptive, and data-informed approaches to teaching and learning ( 45 , 46 ). Findings from the meta-synthesis reveal that AI applications are transforming medical education in multiple ways, including the provision of real-time learner analytics and feedback loops that enable instructors to respond more effectively to individual student needs, as well as supporting inclusive instructional practices designed to address learner diversity across instruction and assessment. In addition to changes in learners’ educational experiences, the data also indicates shifts in faculty instructional practices, curricular structures, and assessment mechanisms. Furthermore, synthesis of student-focused concepts within the meta-analysis highlights a growing recognition of the importance of enabling personalized learning to support students with diverse backgrounds, competencies, and aspirations in contemporary medical education. The primary aim of this study was to synthesize existing research on the application of artificial intelligence to personalized education in the training of physicians and to organize this body of literature into analytically coherent categories with clear practical implications. The findings are structured around five major domains: 1) faculty applications; 2) student applications; 3) learning process applications; 4) curriculum development applications; and 5) student assessment applications. Understanding the potential for transformative change in medical education requires consideration of both the distinct contributions of each domain and the interrelationships among them.

(i) Applications for Faculty Members

In the development of personalized approaches to learning, faculty roles have shifted from a primary focus on knowledge transmission to the design of personalized learning experiences for individual learners. Faculty are now able to tailor instructional resources, provide varied forms of feedback, and implement diverse instructional strategies by identifying learner diversity through performance data analysis and the development of adaptive teaching models. In addition, personalized technologies have helped reduce the time faculty spend on real-time administrative tasks and have enhanced data-driven decision-making through the more effective and efficient use of educational resources. As a result of these educational transformations, the quality of teaching has improved, and alternative modes of communication between instructors and learners have emerged that better address individual learner needs.

Tetzlaff, et al. ( 47 ) developed a framework to guide educators in implementing an innovative model that reconceptualizes personalized learning as a dynamic, multilayered, and data-driven process. They argue that for personalization to be effective, learners’ cognitive, motivational, and emotional characteristics must be continuously monitored in real time, with instructional adjustments made accordingly. As part of ongoing research related to personalized learning, Reber, et al. ( 48 ) proposed a framework for fostering students’ interest in mathematics and science. They describe how academic interest typically develops from situational interest into more stable, individualized interest when appropriate support is provided. To facilitate this transition, they identify three types of interventions: context personalization, example choice, and active personalization. Evidence indicates that these interventions can enhance situational interest and behavioral engagement among students with initially low levels of either and, in some cases, promote sustained individual interest.

Building on prior work in this area, Tang, et al. ( 49 ) developed a model of personalized STEM education centered on self-regulated learning, using analyses of learner behaviors on online platforms to generate customized learning recommendations. Data mining and machine learning algorithms are incorporated into this approach to support core self-regulatory processes, including planning, monitoring, and self-evaluation. These findings are consistent with a growing body of research on the personalization of education ( 20 , 21 , 24 , 25 , 27 - 34 ).

(ii) Applications for Students

Personalized learning through the creation of customized learning pathways is one of the most visible manifestations of personalization in education. These learning pathways are developed based on the learners’ educational background, cognitive style, personal interests, and career aspirations, and are intended to promote greater awareness of the learners’ own learning processes, self-regulation, and responsibility for learning. In addition to enhancing motivation and deeper comprehension of learning material, personalized learning supports the development of learners’ capacity to make informed decisions about their educational and career pathways. In this context, the use of real-time feedback, progress tracking, and individualized instruction has fostered learner-centered environments in which students are positioned at the core of the instructional process.

Wu, et al. ( 50 ) demonstrated that tailoring content, pace, and assessment to individual learner characteristics can substantially improve instructional effectiveness. Analyses based on behavioral and cognitive data illustrate how dynamic learning systems can accommodate the learners’ unique differences, including learning styles, prior knowledge, and preferences in responding to instructional materials. This approach is particularly relevant in medical education, where learners must master complex and specialized content, often of a clinical nature, which requires active engagement, strong self-regulation, and individualized learning trajectories. Similarly, the implementation of intelligent learning systems has been shown to increase student interaction, promote self-regulated learning, and improve academic performance ( 51 ). By aligning instructional content with students’ interests, experiences, and personal goals ( 52 ), Vorobyeva et al. demonstrated increases in engagement across motivational, cognitive, and behavioral dimensions ( 53 ). Finally, Ellikkal, et al. ( 53 ) illustrated how recommender systems can facilitate individualized learning experiences based on the learners’ cognitive and behavioral characteristics. Collectively, these findings are consistent with and support the broader literature on personalized learning ( 20 - 25 , 27 - 35 ).

(iii) Applications in the Learning Process

Recent advances in teaching and learning have necessitated new approaches within contemporary learning environments. Instructional design theory should emphasize instructional activities informed by data derived from both behavioral and cognitive assessments of learners. Cognitive load can be managed through the provision of individualized learning pathways. Creating opportunities for learners to engage in meaningful learning experiences, such as group activities, clinical simulations, and individualized projects, represents a critical component of this transformation. Learning within these environments is not solely cognitive in nature but also emotional, social, and experiential.

Alqahtani, et al. ( 54 ) suggest that the integration of artificial intelligence technologies into learning environments has the potential to transform instructional design and delivery. Adaptive learning pathways can be developed through the application of machine learning algorithms and behavioral data analysis. Moreover, adaptive learning technologies enable support to be provided to individual learners at their point of need. Through the implementation of such approaches, learners demonstrate increased engagement and self-regulation, while responsive and individualized learning environments are established. These developments also promote new forms of interaction between students and instructors that align with the learners’ styles, needs, and personal goals ( 55 ).

Furthermore, Fitria ( 56 ) reports that AI-supported systems can address a wide range of learner needs in the classroom and enhance the quality of learner–instructor interactions. By identifying learner behaviors, recognizing preferred learning modalities, and delivering targeted content, AI technologies can support more effective and efficient instruction. Collectively, these findings align with and support the broader literature on AI-enabled personalized learning ( 20 - 34 ).

(iv) Curriculum Development

Curriculum-level personalization involves aligning education with each learner’s unique needs, interests, and career aspirations to create curricula that effectively address those needs. This approach requires the design of educational programs informed by analyses of learning styles, learning performance, and other individual learner characteristics. Examples include the use of flexible digital content, customizable modules, and authentic, individualized projects embedded within the curriculum. Personalized curricula are typically developed using multifaceted, non-linear structures that provide learners with opportunities for gradual, varied, and intensive levels of learning. Collaboration in the curriculum design process, such as incorporating student input and co-developing course materials, actively engages learners and fosters a sense of ownership and cognitive engagement, thereby increasing the likelihood of long-term knowledge retention. According to Gerard, et al. ( 57 ), personalizing curriculum development is an effective strategy for addressing individual learner differences. They reported that well-designed curricula aligned with students’ learning styles, interests, prior knowledge, and academic objectives enhance engagement with the learning process, improve conceptual understanding, and foster a stronger sense of belonging. In a global survey study, Abbasi, et al. ( 58 ) argued that AI tools could assess learner-specific needs and provide real-time feedback, thereby supporting the development of critical thinking skills and facilitating more flexible, goal-directed curricula. Tavakoli, et al. ( 59 ) proposed an alternative approach to curriculum personalization through the integration of crowdsourcing and AI. Their four-level framework enables the continuous development and refinement of personalized curricula based on collective learner input and encompasses learning objectives, skills, topic content, and learning packages. Collectively, these findings support prior research on curriculum personalization ( 20 , 21 , 24 - 27 , 29 - 32 ).

(v) Development of Student Assessment Mechanisms

Assessment has traditionally been used as a tool to measure knowledge acquisition; however, it is increasingly being integrated as a core component of the learning process. The use of real-time graphical feedback and analysis of student performance data enables the linkage of evidence from individual learning outcomes to the development of personalized learning pathways based on prior performance. Examples of such assessment approaches include adaptive assessments, real-world assessment tasks grounded in authentic professional practice, and self-assessment models that allow students to evaluate their own performance. The application of artificial intelligence and machine learning enables more accurate identification of students’ strengths and weaknesses, thereby supporting the development of efficient and individualized learning pathways. Accordingly, assessment functions not only as a means of differentiation among learners but also as a mechanism for professional development, critical thinking, and enhanced decision-making. According to Hooda, et al. ( 60 ), machine learning technologies allow instructors to continuously monitor student performance throughout a course and to provide proactive support through early interventions for learners who may be struggling. Recent research suggests that AI has the potential to enhance the quality of higher education by improving learning experiences for students. Jani, et al. ( 61 ) demonstrated how intelligent algorithms could be effectively used to generate feedback, conduct assessments, and support formative evaluation processes. Their findings indicated that structured checklists combined with machine learning techniques enabled effective monitoring of student’s progress and facilitated the identification of areas in clinical performance requiring improvement. Similarly, Samarakou, et al. ( 62 - 64 ) examined the role of artificial intelligence in continuous assessment and learner support for engineering students using the StuDiAsE (Student Diagnosis, Assistance, Evaluation) system. Their results showed that AI systems could provide individualized feedback and assess student performance using both qualitative and quantitative indicators. These findings are consistent with previous studies on AI-supported assessment practices ( 20 , 21 , 23 - 34 ).

Limitations and Future Directions

The principal limitation of this investigation is its focus solely on English-language publications published between 2010 and 2025. As a result, relevant literature published in other languages or addressing different time periods may have been excluded. In addition, grey literature (e.g., technical reports, student theses) was not included, which may limit the identification of emerging perspectives or innovative applications. Another limitation is variability in how the concept of “personalized learning” and associated AI applications are defined and operationalized across the literature, including differences in definitions and measurement approaches. This variability makes direct comparison and synthesis of the findings across studies challenging. Future investigations should employ experimental and longitudinal research designs to more rigorously assess the effectiveness of AI in supporting personalized learning across diverse educational settings. Further research should also explore hybrid (human–machine) instructional approaches in clinical education, develop robust indicators for measuring equity in personalized learning environments, and apply machine learning techniques to examine academic and career pathways. Collectively, these directions would support the continued integration of AI in medical education and contribute to improving both the quality and effectiveness of training programs.

Conclusion

In this research, a meta-synthesis was conducted using a structured framework designed to identify, categorize, and consolidate previous evidence regarding the types of artificial intelligence applied to personalized learning in medical education. The findings indicate that AI-driven personalization, in combination with technologies such as large language models, virtual reality, and real-time learning analytics, enables alignment of educational processes with the learners’ cognitive styles, goals, and prior knowledge, while simultaneously redefining the role of faculty from transmitters of knowledge to adaptive designers of learning experiences. This evolving role creates opportunities to enhance learners’ self-regulation, motivation, and clinical reasoning, as well as to support evidence-informed educational and career development pathways. The results further demonstrate that personalization occurs across multiple levels, influencing curriculum design, assessment practices, and the configuration of educational technologies within diverse teaching and learning environments. Accordingly, educational institutions should employ data-mining techniques, adopt adaptive algorithms, and provide real-time feedback to learners to develop inclusive and equitable learning ecosystems. In addition to synthesizing findings from previously conducted studies and generating new insights through an innovative conceptual model of AI-enabled personalization, the conceptual and policy framework developed through this meta-synthesis may serve as a guide for the ethical and equitable governance of AI-enabled personalization. This framework includes recommendations for curriculum redesign, adaptive assessment strategies, and faculty development initiatives. Collectively, the findings of this study not only integrate existing research but also provide a roadmap for theory, policy, and practice with the potential to transform the future delivery of medical education.

Acknowledgment

The authors wish to acknowledge their colleagues who contributed to the quality appraisal and consensus-building phases through their independent review of the documents and adjudication of discrepancies among authors. The authors express their sincere appreciation to all individuals involved in these phases, whose contributions provided valuable insights and supported the collaborative nature of the research process. The authors also acknowledge that an artificial intelligence (AI) language model was used to support the development of the manuscript structure and to enhance clarity of expression. All AI-assisted content was carefully reviewed, edited, and validated by the authors prior to submission to ensure that the ideas, interpretations, and conclusions presented in this manuscript are solely those of the authors.

Authors’ Contributions

M.H contributed to data collection. A.K performed data analysis. Z.F assessed the quality of the articles and wrote the initial draft. AT.M contributed to translation and final manuscript writing. A.K was the principal supervisor of the entire research process. All authors reviewed the final manuscript.

Conflicts of interest

The authors declare no conflicts of interest.

Declaration of AI Use

The authors used AI tools to translate and summarize some articles. All AI-generated outputs have been reviewed and approved by all authors. The authors assume full responsibility for the final content.

References

1. Keykha A, Fazlali B, Behravesh S, Farahmandpour Z. Integrating Artificial Intelligence in Medical Education: A Meta-Synthesis of Potentials and Pitfalls of ChatGPT. Journal of Advances in Medical Education & Professionalism. 2025; 13(3):155.
2. Keykha A, Mohammadi H, Darabi F, Hosseini SS. Identifying the Applications of Artificial Intelligence in the Assessment of Medical Students. Strides in Development of Medical Education. 2025; 22(1):e1512.
3. Keykha A, Imanipour M, Shahrokhi J, Amiri M. The Advantages and Challenges of Electronic Exams: A Qualitative Research based on Shannon Entropy Technique. Journal of Advances in Medical Education & Professionalism. 2025; 13(1):1.
4. Keykha A. Extraction and classification of smart university components to provide a conceptual framework: A meta-synthesis study. Sciences and Techniques of Information Management. 2022; 8(4):75-112.
5. Keykha A, Hojati M, Taghavi Monfared A, Shahrokhi J. Artificial Intelligence in Healthcare: Unveiling Ethical Challenges Through Meta-Synthesis of Evidence. Journal of Research and Health. 2025; 15(6):661-82.
6. Mahdi R, Keykha A, Kaliisa R, Darabi F. Exploring applications of artificial intelligence in enhancing the quality of medical education: a mixed methods research synthesis. Journal of Medical Education Development. 2025; 18(4):119-39.
7. Keykha A, Ahmadi G, Keramatfar A. Navigating the Knowledge Stream: Analyzing ChatGPT's Impact on Education Through Bibliometric. International Journal of Information Science and Management (IJISM). 2025; 24(1):1-27.
8. Mir MM, Mir GM, Raina NT, Mir SM, Mir SM, Miskeen E, et al. Application of artificial intelligence in medical education: current scenario and future perspectives. Journal of Advances in Medical Education & Professionalism. 2023; 11(3):133.
9. Mansourzadeh A, Rasouli S. The Future of Medical Education: A Review of the Opportunities and Challenges of Artificial Intelligence Integration. Medical Education Bulletin. 2024; 5(2):973-82.
10. Keykha A, Behravesh S, Ghaemi F. ChatGPT and Medical Research: A Meta-Synthesis of Opportunities and Challenges. Journal of Advances in Medical Education & Professionalism. 2024; 12(3):135.
11. Guo E, Ramchandani R, Park YJ, Gupta M. OSCEai: personalized interactive learning for undergraduate medical education. Canadian Medical Education Journal. 2025; 16(6):7-14.
12. Shemshack A, Spector JM. A systematic literature review of personalized learning terms. Smart Learning Environments. 2020; 7(1):33.
13. Ramazonovich MO. Personalized learning in pathophysiology: adapting education to student needs. The American Journal of Medical Sciences and Pharmaceutical Research. 2025; 7(3):32-8.
14. Major L, Francis GA, Tsapali M. The effectiveness of technology‐supported personalised learning in low‐and middle‐income countries: A meta‐analysis. British Journal of Educational Technology. 2021; 52(5):1935-64.
15. Ackermans K, Bakker M, van Loon AM, Kral M, Camp G. Young learners’ motivation, self-regulation and performance in personalized learning. Computers & Education. 2025; 226:105208.
16. Rajasekaran SK. Precision Education-The Future of Medical Education. Journal of Medical Education and Practice. 2024; 1(1):17-9.
17. Sandelowski M, Barroso J, Voils CI. Using qualitative metasummary to synthesize qualitative and quantitative descriptive findings. Research in nursing & health. 2007; 30(1):99-111.
18. Sandelowski M, Barroso J. Handbook for synthesizing qualitative research. USA: Springer Publishing Company; 2006.
19. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ. 2009; 6(7):e1000097.
20. Johnson KB, Wei WQ, Weeraratne D, Frisse ME, Misulis K, Rhee K, et al. Precision Medicine, AI, and the Future of Personalized Health Care. Clin Transl Sci. 2020; 14(1):86-93.
21. Obeagu EI, Tukur M, Asuma MN. Personalized Learning Plans: Catering to Individual Needs in Sickle Cell Disease Education. Elite Journal of Nursing and Health Science. 2024; 2(5):23-9.
22. Bahmani A, Cha K, Alavi A, Dixit A, Ross A, Park R, et al. Achieving Inclusive Healthcare through Integrating Education and Research with AI and Personalized Curricula. medRxiv. 2024; 5(1):4-7.
23. Tsekhmister Y, Konovalova T, Tsekhmister B. Using behavioral analytics to personalize learning experiences in digital medical education: a case study. Academia. 2023; 23(33):83-103.
24. Ali M, Wahab IB, Huri HZ, Yusoff MS. Personalised learning in higher education for health sciences: a scoping review protocol. Systematic Reviews. 2024; 13(1):99.
25. Raeisi E, Aazami MH, Solati K, Mohamadi O, Ahmady S. A modified student personalized learning approach to prompt academic acquisition in health sciences. Journal of Education and Health Promotion. 2019; 8(1):156.
26. Rojanasarot S, Milone A, Balestrieri R, Pittenger AL. Personalized learning in an online drugs and US health care system controversies course. American journal of pharmaceutical education. 2018; 82(8):6391.
27. Stambuk‐Castellano M, Carrera A, Tubbs RS, Alario‐Hoyos C, Verdú E, Iwanaga J, et al. Personalized strategies for academic success in learning anatomy: Exploring metacognitive and technological adaptation in medical students. Clinical Anatomy. 2024; 37(4):472-83.
28. Kelly L, Walters L, Rosenthal D. Community-based medical education: is success a result of meaningful personal learning experiences? Education for health. 2014; 27(1):47-50.
29. Abedi R, Nili Ahmadabadi MR, Taghiyareh F, Aliabadi K, Pourroustaei Ardakani S. The effects of personalized learning on achieving meaningful learning outcomes. Interdisciplinary Journal of Virtual Learning in Medical Sciences. 2021; 12(3):177-87.
30. Sadeqi-Arani Z, Vahidnia R, Nasrabadi EM. Benefits and Application of IoB in Educational Businesses: Smart, Sustainable, and Personalized Learning. Journal of Advances in Medical Education & Professionalism. 2025; 13(1):76.
31. Rabie RM. The Role of Artificial Intelligence and Personalized Education in Medical Curriculum: A Systematic Review of Applications and Challenges. Faculty of Education Journal Alexandria University. 2023; 33(4):365-84.
32. Sunmboye K, Strafford H, Noorestani S, Wilison-Pirie M. Exploring the influence of artificial intelligence integration on personalized learning: a cross-sectional study of undergraduate medical students in the United Kingdom. BMC Medical Education. 2025; 25(1):570.
33. Shen M, Shen Y, Liu F, Jin J. Prompts, privacy, and personalized learning: integrating AI into nursing education—a qualitative study. BMC nursing. 2025; 24(1):470.
34. Hu C, Li F, Wang S, Gao Z, Pan S, Qing M. The role of artificial intelligence in enhancing personalized learning pathways and clinical training in dental education. Cogent Education. 2025; 12(1):2490425.
35. Yovanoff M, Pepley D, Mirkin K, Moore J, Han D, Miller S. Personalized learning in medical education: designing a user interface for a dynamic haptic robotic trainer for central venous catheterization. Inproceedings of the human factors and ergonomics society annual meeting. SAGE Publications. 2017; 61(1):615-9.
36. Long HA, French DP, Brooks JM. Optimising the value of the critical appraisal skills programme (CASP) tool for quality appraisal in qualitative evidence synthesis. Research Methods in Medicine & Health Sciences. 2020; 1(1):31-42.
37. Barisone M, Bagnasco A, Hayter M, Rossi S, Aleo G, Zanini M, et al. Dermatological diseases, sexuality and intimate relationships: A qualitative meta‐synthesis. Journal of Clinical Nursing. 2020; 29(17-18):3136-53.
38. Boeije HR, van Wesel F, Alisic E. Making a difference: towards a method for weighing the evidence in a qualitative synthesis. Journal of Evaluation in Clinical Practice. 2011; 17(4):657-63.
39. Elo S, Kyngäs H. The qualitative content analysis process. Journal of advanced nursing. 2008; 62(1):107-15.
40. Keykha A, Ezati M, Salehi M. Entrepreneur university model design: Qualitative approach (Case study: University of Tehran). Iranian Journal of Engineering Education. 2019; 21(83):51-77.
41. Keykha A, Ezati M, Khodayari Z. Identification of the barriers and factors affecting the quality of higher education in Allameh Tabataba’i university from the viewpoints of faculty members. Quality in Higher Education. 2022; 28(3):326-44.
42. Keykha A, Towfighi J. Redefining the role of faculty members in higher education policy: a qualitative study using thematic analysis. Iranian Journal of Public Policy. 2021; 7(3):55-76.
43. Keykha A, Ezati M. Identifying factors hindering ecosystem development, entrepreneurship, entrepreneurial university. Innovation Management Journal. 2021; 10(2):55-97.
44. Keykha A. Analysis of causes of unemployment of graduates in higher education. Journal of Teaching in Marine Sciences. 2022; 9(1):21-39.
45. Jafari F, Keykha A, Taheriankalati A, Monfared AT. The role of AI in shaping medical education: insights from an umbrella review of review studies. Journal of Advances in Medical Education & Professionalism. 2025; 13(4):270.
46. Keykha A, Mohammadi F, Taghavi Monfared A, Taheriankalati A. Unblocking innovation: a meta-synthesis of blockchain applications in medical education, research, and healthcare. Strides in Development of Medical Education. 2025; 22(1):1587.
47. Tetzlaff L, Schmiedek F, Brod G. Developing personalized education: A dynamic framework. Educational Psychology Review. 2021; 33:863-82.
48. Reber R, Canning EA, Harackiewicz JM. Personalized education to increase interest. Current directions in psychological science. 2018; 27(6):449-54.
49. Tang Y, Liang J, Hare R, Wang FY. A personalized learning system for parallel intelligent education. IEEE Transactions on Computational Social Systems. 2020; 7(2):352-61.
50. Wu S, Cao Y, Cui J, Li R, Qian H, Jiang B, et al. A comprehensive exploration of personalized learning in smart education: From student modeling to personalized recommendations. Front Comput Sci. 2026; 1:1-61.
51. Naseer F, Khan MN, Tahir M, Addas A, Aejaz SH. Integrating deep learning techniques for personalized learning pathways in higher education. Heliyon. 2024; 10(11):e32628.
52. Vorobyeva KI, Belous S, Savchenko NV, Smirnova LM, Nikitina SA, Zhdanov SP. Personalized Learning through AI: Pedagogical Approaches and Critical Insights. Contemporary Educational Technology. 2025; 17(2):ep574.
53. Ellikkal A, Rajamohan S. AI-enabled personalized learning: empowering management students for improving engagement and academic performance. Vilakshan-XIMB Journal of Management. 2025; 22(1):28-44.
54. Alqahtani T, Badreldin HA, Alrashed M, Alshaya AI, Alghamdi SS, Bin Saleh K, et al. The emergent role of artificial intelligence, natural learning processing, and large language models in higher education and research. Research in social and administrative pharmacy. 2023; 19(8):1236-42.
55. Son JB, Ružić NK, Philpott A. Artificial intelligence technologies and applications for language learning and teaching. Journal of China Computer-Assisted Language Learning. 2025; 5(1):94-112.
56. Fitria TN. Artificial intelligence (AI) in education: Using AI tools for teaching and learning process. Indonesia: Institut Teknologi Bisnis AAS; 2021.
57. Gerard L, Bradford A, Linn MC. Supporting teachers to customize curriculum for self-directed learning. Journal of Science Education and Technology. 2022; 31(5):660-79.
58. Abbasi BN, Wu Y, Luo Z. Exploring the impact of artificial intelligence on curriculum development in global higher education institutions. Education and Information Technologies. 2025; 30(1):547-81.
59. Tavakoli M, Faraji A, Molavi M, Mol S, Kismihók G.Paper presented at: ; 2022.
60. Hooda M, Rana C, Dahiya O, Rizwan A, Hossain MS. Artificial intelligence for assessment and feedback to enhance student success in higher education. Mathematical Problems in Engineering. 2022; 2022(1):5215722.
61. Jani KH, Jones KA, Jones GW, Amiel J, Barron B, Elhadad N. Machine learning to extract communication and history-taking skills in OSCE transcripts. Med Educ. 2020; 54:1159-70.
62. Samarakou M, Fylladitakis ED, Karolidis D, Früh WG, Hatziapostolou A, Athinaios SS, et al. Evaluation of an intelligent open learning system for engineering education. Knowl Manag E-Learning Int J. 2016; 8:496-513.
63. Farahmandpour Z, Voelkel R. Teacher Turnover Factors and School-Level Influences: A Meta-Analysis of the Literature. Education Sciences. 2025; 15(2):219.
64. Farahmandpour Z, Voelkel Jr RH. Unraveling the causes of teacher turnover: A meta-analysis of global literature. Leadership and Policy in Schools. 2025; 1:1-7.