Psychotherapists in training lack a standardized and formalized method of patient interaction for the proper development of empathy, communication, and experience. Currently, the experience of a resident may include training with other residents where one patient acts as the patient and one the psychotherapist, or with the usage of a simulated patient, an actor, to practice with the resident. Both methods have shortcomings in availability, reliability, and the accuracy of the patient in replicating a real scenario. This project attempted to create virtual patients by utilizing online patient transcripts through the fine-tuning and transfer-learning of three modern Artificial Intelligence models, ChatGPT-4o, Lla Ma-3.1v-405B, and Gemini 1.5 Pro; alongside the miniature versions of these models as applicable. This included the creation of a website interface that can interact with the created models for evaluation, while also allowing an interactive format with a simplistic design. The accuracy of the models was independently evaluated through cosine similarities between data and model outputs to find semantic relations, and varied from 92.99% to 83.40%; with ChatGPT-4o Mini and Full having the highest fine-tuned and transfer-learning accuracies respectfully. Furthermore, a customizable model allowed user input for specific descriptions and mental illnesses. This highlights the potential for a model to be successfully representative of a patient which can be utilized to train residents easier than currently available methods. The need for further evaluation and continual training are at the forefront of current limitations.

References

• Accreditation Council for Graduate Medical Education, ACGME Competencies, 2024. [Online]. Available: https://www.acgme.org/programs-and-in stitutions/programs/common-program-require ments/
• New York State Licensed Professionals, NYS Mental Health Counseling, NYSED, 2024. [Online]. Available: https://www.acgme.org/programs-and-in stitutions/programs/common-program-requirements/
• H. S. Barrows, “An overview of the uses of standardized patients for teaching and evaluating clinical skills,” Acad. Med., vol. 68, no. 8, pp. 443–451, 1993.
• A. Lovink, M. Groenier, A. Niet, H. Miedama, and J. J. Rethans, “How simulated patients contribute to student learning in an authentic way: an interview study,” Adv. Simul., vol. 9, no. 4, pp. 1–10, 2024.
• A. J. Cleland, K. Abe, and J. J. Rethans, “The use of simulated patients in medical education: AMEE Guide No 42,” Med. Teach., vol. 31, no. 6, pp. 477–486, 2009.
• Body Interact, “Virtual Patients,” 2024. [Online]. Available: https://bodyinteract.com/virtual-pa tient-simulator/
• B. Zalewski, M. Guziak, and M. Walkiewicz, “Developing simulated and virtual patients in psycho logical assessment – Method, insights and recommendations,” PubMed, vol. 12, no. 1, pp. 455–461, 2023.
• IBM, “What is Machine Learning?,” 2024. [Online]. Available: https://www.ibm.com/topics/machine-learning
• A. M. Turing, “Computing machinery and intelligence,” Mind, vol. 59, no. 236, pp. 433–460, 1950.
• PyTorch, “ADAMW,” 2023. [Online]. Available: https://pytorch.org/docs/stable/generated/torch.op tim.AdamW.html
• C. Stryker and J. Holdsworth, “What is NLP?,” IBM, 2024. [Online]. Available: https://www.ibm.com/topics/natural-language-processing
• J. Lederer, “Activation functions in artificial neural networks: A systematic overview,” arXiv preprint arXiv:2101.09957, 2021.
• P. Baheti, “Activation Functions in Neural Networks: 12 Types & Use Cases,” v7, 2021. [Online]. Available: https://www.v7labs.com/blog/neural-networks-activation-functions
• Amazon Web Services, “What is a Neural Network?,” 2024. [Online]. Available: https://aws.amazon.com/what-is/neural-network/
• C. Stryker, “What is a recurrent neural network?,” IBM, 2024. [Online]. Available: https://www.ibm.com/topics/recurrent-neural-networks
• KiKaBeN, “Transformer’s Encoder-Decoder,” 2021. [Online]. Available: https://kikaben.com/transformers-encoder-decoder/
• J. L. Ba, J. R. Kiros, and G. E. Hinton, “Lay er normalization,” arXiv preprint arXiv:1607.0650, 2016.
• IBM, “What is Mixture of Experts?,” 2024. [Online]. Available: https://www.ibm.com/topics/mixture-of-experts
• A. Ahmed, H. Hayat, and D. Hayat, “Comparing GPT-4o, LLaMA 3.1, and Claude 3.5 Son net,” Walturn, 2024. [Online]. Available: https:// www.walturn.com/insights/comparing-gpt-4o-llama-3-1-and-claude-3-5-sonnet
• D. Haywood, “Gpt-4o tokens per second comparable to gpt-3.5-turbo. Data and analysis,” OpenAI Developer Forum, 2024. [Online]. Available: https://community.openai.com/t/gpt-4o-tokens-per second-comparable-to-gpt-3-5-turbo-data-and-analysis/768559