This model is a fine-tuned version of TinyLLaMA (1.1B parameters) for generating Subjective sections of SOAP notes in a healthcare context. It was fine-tuned using a synthetic dataset of 5,160 examples, with 4-bit quantization and LoRA (Low-Rank Adaptation) for efficiency.
This model is intended for generating Subjective SOAP notes in a solar-powered AI unit for rural healthcare settings. It can be deployed on low-power devices like the Raspberry Pi 5 after conversion to ONNX format.
This model was trained on the "Synthetic Healthcare Datasets for SOAP Note Generation" dataset, available at Zenodo:
https://doi.org/10.5281/zenodo.15399846. Please cite Schmidt Batista, Adans when using the dataset, as required by the CC BY 4.0 license.
