Estimation of histopathological types from breast MRI findings using a large language model

Key Points

• This research aims to evaluate the ability of a large language model to predict histopathological types of breast cancer based on MRI findings.
• Utilized a large language model (LLM) for analysis
• Collected MRI findings related to breast cancer
• Compared predictions to actual histopathological diagnoses
• LLM accurately inferred invasive ductal carcinoma cases
• Strong correlation between MRI findings and histopathological types
• Supports the potential of LLMs in medical imaging analysis