SBIR Phase II: Development of Automated Post Operative Rhythm Identification Through Computerized Evaluation of Atrial Signals

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase II project is to build software to interpret a unique cardiac waveform available in many post-op heart patients, but which is not currently used by standard cardiac monitoring systems to detect arrhythmias. This software will alert the nurses and providers in the intensive care unit when it detects an arrhythmia using this unique waveform. By more quickly informing the care team of an arrhythmia, providers will be able to intervene more quickly. This Small Business Innovation Research (SBIR) Phase II project develops and validates machine learning algorithms that analyze a continuous atrial electrocardiogram waveform available in many patients after heart surgery. To accomplish this, this project will involve creating a machine learning model that automatically identifies key features of the cardiac waveform in real time, such as atrial and ventricular electrical activity, and uses those patterns to determine whether the patient’s heart rhythm is normal or abnormal. Prior Phase I research demonstrated technical feasibility of this approach using de-identified clinical data. Underlying the creation of this algorithm is the machine and human labeling of many hours of de-identified continuous cardiac waveform data from patients who have the unique atrial ECG signal available. This data will be processed to prepare it for automated interpretation. This project also involves the creation of a software interface to be used by care givers in the intensive care unit which will show the real-time labeled cardiac waveform and alert the team to potentially dangerous arrhythmias. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.