ECG analysis software

Our client, a U.S. digital healthcare company that provides remote cardiac monitoring services to clinics and hospitals, needed a solution for an accurate automatic analysis of electrocardiogram tracing (ECG) obtained from mobile cardiac telemetry (MCT) wearable devices.

The client started to develop the ECG processing app but the product didn’t function as expected. EffectiveSoft joined the project to perform a code review. It turned out that the app needed significant improvements in the client-server part and more features had to be added for its proper work. Having considered our recommendations, the client delegated the app development to our team.

The team at EffectiveSoft refactored the code, improved the app architecture, and added new features. As a result, the client got a solution for automatic analysis of continuous ECGs during preclinical and clinical research studies.

• The application automatically eliminates 95% of noise in the ECG tracings due to the integration of the proprietary reading algorithms and patented technology for noise removal.
• Medical specialists can configure workflows that automatically upload large groups of ECG recordings into the app and process them. It takes about 30 seconds for the app to process a typical 24 hour ECG recording, and around 10-20 minutes for a specialist to analyze, review, edit, and prepare a report based on the automatically generated data.
• We implemented the multi-server architecture, where one server collects and analyzes raw data from ECG biosensors, and the other server stores the processed recordings for further analysis by medical specialists. The data exchange happens via WCF and RESTful APIs.
• The software supports many file formats including Dataquest, Ponemah, HL7, ISHNE, Matlab, and EDF. It also analyzes the ECG recordings by Holter monitors that are manufactured by IntriCon, TZ Medical, and Braemar. Results can be downloaded as a graphical or tabular PDF report and/or exported as .csv files for further analysis in Excel or other statistics software.
• The solution is designed to be 21 CFR part 11 compliant and leverages robust Windows account security administration.
• We also introduced the classification of beat types that shows the patient’s heart condition more comprehensively and improves the accuracy of ECG analysis reports.
• The application is cross-platform and works on Windows and Linux OSs.