Date of Award



Lily Laiho

Abstract Patient Care Network is a St. Jude Medical product that is used for remote patient management. The basic concept of is to allow the physician to view patient device follow-up information as well as general patient and device information on a web application. The system also interfaces with the patient and will send them notification if they miss a follow-up. All device information will be collected automatically while the patient is sleeping. This information is sent through a telephone line to a server to process a report package and display the collected information on the web application. The verification team ensures that all reports generated by the servers are processed and outputted correctly. There are currently 296 device parameters supported by, and the manual extraction and comparison of the expected parameter values takes several hours for each patient follow-up session. Currently there are 250 patient follow-up sessions used for verification testing. Each new release will continue to create additional patient follow-up sessions. releases are approximately 6 months apart, and each new release adds approximately 30-50 new patient follow-up records to support the new devices. In order to meet aggressive project deadlines, while ensuring that the system is processing and outputting patient follow-up data correctly, it is necessary to come up with an automated process to verify the contents of the processed data is correct. This will save a tremendous amount of time as well as improve on the quality of the verification process by eliminating human error and rework. It is critical for patient safety that the patient device follow-up information is processed and outputted correctly.

In this thesis an automated process was developed to verify the correct content of the server generated reports for each patient follow-up session. This process leveraged different tools and scripting languages to achieve automation. TDE (Test Development Environment) tool was used to extract the device parameters from the patient follow-up sessions. The TDE script was written to extracts the desired parameter values from the patient follow-up session and automatically populates parameters in a device parameters spreadsheet. Once all the device parameter values are extracted in the spreadsheet, they are passed through a set of mapping rules, which form the expected values. The mapping rules were implemented as VBA (Visual Basic for Application) macros, one macro for each report. The VBA macros write the expected values back to the spreadsheet to form an “expected values spreadsheet”. The patient follow-up session is then sent to the server to process, which generates a processed patient follow-up session that contains a reports package in .zip format. A perl script was then written to compare the parameter values in the generated reports with the corresponding expected values from the expected values spreadsheet. The perl script generates a comparison report displaying the discrepancies between the actual and the expected values.