1.Improving the existing solution with a modern AI-based solution to detect and classify dangerous arrhythmia will make our software as a medical device more convenient as a consumer-based device with improved accuracy as a diagnostic tool, like a wrist-based blood pressure monitor.
2.Developing new business models for end-user self-diagnosis and real-time monitoring of outpatients.
1.Develop training datasets
2.Develop tokens for heartbeat annotations and ECG digital samples
3.Building a heart-based language foundation model
4.Apply transformers to analyze sequences of time-series data to detect and classify dangerous arrhythmia
1.Training datasets covering a variety of arrhythmia, considering the quality of measured signals, the divergence of population, age, and gender.
2.Binning strategy to build tokens for heartbeat annotations and ECG digital samples
3.Heart-based language model
4.Fine-tuned model to classify dangerous arrhythmia
1.Providing better healthcare, by enabling a medical device that will alert on the detection of dangerous arrhythmia and protect against severe heart damage.
2.Become first on the market with an independent device to offer automatic recognition of abnormality and real-time alerting.
2.Change the way healthcare is delivered with sophisticated AI-based technology that puts the power back into the hands of patients.
2.The luxury of free normal life activities, while outpatients heart is monitored and looked after.