This PC application, developed in Python, is a cutting-edge tool designed to assist clinicians and researchers working with patients undergoing Deep Brain Stimulation (DBS) therapy using the Medtronic Percept™ PC neurostimulator. Tailored for continuous and retrospective monitoring, it streamlines the analysis of clinical and electrophysiological data, offering valuable insights into patient conditions and brain activity. This enhances the precision of clinical decision-making, particularly in optimizing electrostimulation settings.
The application significantly simplifies the extraction and interpretation of data from the complex JSON files generated by the Percept™ device. By eliminating the need for advanced programming skills in MATLAB or Python, it makes sophisticated data analysis more accessible. Key features include tools for efficient biomarker extraction (both periodic and aperiodic components), robust data visualization, and in-depth analytical capabilities.
Additionally, the application automates the creation of comprehensive PDF reports, exports LFP montages and home recording data to Excel, and provides an intuitive interface, making it an indispensable resource for both clinical practice and research.
- Download the zip file from here.
- Extract the contents of the ZIP folder (named "dist") to your preferred location on your computer.
- Right-click the file named "DBS_Data_Analyzer_ver_X" (currently version 2) and select Create Shortcut from the context menu.
- Move the shortcut to a convenient location on your computer.
- Launch the application by double-clicking the shortcut.
- Create a dedicated folder for a single patient and ensure it contains all the relevant JSON files associated with that patient.
- Launch the DBS Data Analyzer application. On the opening screen, click the Start button.
- Select Browse to navigate and choose the folder containing the patient's files.
- In the Patient Information section, enter the details to be saved in the clinical Excel file. During the first run on the folder, the Date of Surgery and Name fields are mandatory (names could be non-identified, i.e., patient’s number). However, in subsequent runs, the application will retain this information, and there will be no need to fill these fields again. Any data entered in the Comments and Medications fields will be included in a table presented on the last page of the PDF report which is generated by the application.
- Clinical questionnaires (Y-BOCS, CGI, HDRS, GAF) can be directly completed by the clinician on the application. In addition, the clinical can document the current clinical status of the patient comparing to their previous medical visit (Clinical Change, Side Effects, Medication Changes).
- The clinical Excel file will be generated after the first run on the patient's folder. In case there is previous clinical data that was not documented using the application, the clinical Excel allows the clinician to add past clinical scores by filling the Date column and the corresponding clinical score columns.
- In the Montage section, the frequency ranges and DBS contacts can be selected to be included in the montage figures in the PDF report. If needed, the frequency ranges can be manually adjusted with their labels by clicking the Change Manually button. This allows modifying both the numerical frequency ranges and their descriptive names (labels). If after running the application, outliers are noticed in the montage figures that hinder trend visualization, outliers removal can be done by checking the box labeled Remove outliers from montages plot. This option will automatically clean data points that deviate by more than 3 SD from the mean, improving data visualization.
- In the Home Recording section, recording frequency, recording contacts, and the hemisphere of interest can be selected to be included in the corresponding section of the PDF report. For figures displaying diurnal LFP patterns (mean 24-hour LFP) for the days recorded in each JSON file, the Include Circadian Plot field should be set to Yes. You can select a specific period of interest to focus on montages and home recordings plots. You can also manually choose the scale of the home recording plot by choosing Enable LFP Limits.
- Click the Generate Report button to run the application. Wait until the progress bar appears, indicating the application is running.
The process is complete once the progress bar reaches the end, signalling that the PDF report containing up to 8 sections (8/8) has been successfully generated and the next window will appear, Click OK.
10. Four output files will be created in the selected patient folder:
- PDF Report: A retrospective summary of the patient's electrophysiological and clinical state.
- Clinical Excel File: A file containing all accumulated clinical data for the patient. This file is updated during each run, ensuring there is always only one updated version in the folder.
- Montages Excel File: A file containing extracted features (both periodic and aperiodic) for each montage, hemisphere, and contact.
- Home Recording Excel File: A file including all LFP samples (recorded every 10 minutes within a defined frequency range) from all JSON files in the folder.
- The application tracks JSON files that have already been processed. During subsequent runs, the application processes only the new files in the folder that were not included in previous runs.
