that was my experience just comparing Apple Health export using Stanford Labs data vs say FunctionHealth.

Im hoping to use LLMs to help with this process.

Will try to use LOINC dataset to standardize against