Raw data transformation of laboratory results
Problem definition:
For a clinical study, it should be possible to analyze all pre- and post-operative laboratory data for a large number of surgical patients.
Due to the lack of an interface, the laboratory data could not be provided in standardized formats via the network. This had to be done manually with a data carrier in an Excel/CSV format that could not be evaluated automatically.
Therefore, the task was to convert the measured values of surgical and post-operative laboratory examinations from a proprietary format into an analyzable database format.
Situation before:
No alternative solution available except massive manual work.
Situation afterwards:
My system was able to parse all the file formats from the designated folder, then migrate them to a folder "already processed" and match the data. This was challenging because the number of measurement series was always different and the delivered measurement values always differed (depending on the doctor's evaluation requirements).
I was now able to extract these measured values with their data from the data supplied without a regular structure and save them in a table in an analyzable form.
Simple Menu
Flexible formulas for processing the data could be stored for defined fields
Detail of the laboratory data: Patient ID at the top left; all other data is always flexible in terms of quantity and structure
Next occurrence of the patient ID defines a new chunk
Next occurrence of the patient ID defines a new chunk
Content of the folder with files to be imported (Before & after surgery) – a bit work in progress
Uploaded data with isolated and identified values