Every year 2.1 million patients die due to heart ailments in India. We have just 1 Cardiac Specialist for every 3 lakh patients. Although secondary and tertiary health care providers are engaged doing ECGs and several other vital collections to analyze heart health conditions, but none of the existing solutions can identify high risk cases from such reports and respond in time.
To address this gap, iKure combining Big Data, AI and machine learning works on ECG signal anomaly detection & Patient risk model using stank ranking & deep learning that can assist cardiologists offer preventive care much faster in the human life cycle. In association with IBM data science, iKure’s integrated platform provide seamless automation and interpretation of patient’s data to enable instant feedback for high risk cases and offers high quality and cost effective continuity of cardiac care both for rural and semi-urban population.
Timely interventions can reduce the time span between critical and stable heart health conditions. However, failure of such gets magnified typically in last mile populations that constitute over 16 million cardiac patients out of total 30 million people. It has been observed that over the year from 2000 to 2015, the age-standardized rate of mortality (per 100,000 person-years) due to coronary heart diseases increased among rural men by 40% and for female it rose to 56%, whereas decline was registered among urban residents. Cardiac care in rural India is limited and inaccessible due to major shortfalls like shortage of doctors, diagnostic centers, equipped facilities, coupled with lack of awareness.
Cardiac diseases can be addressed through effective treatment plans, constant screening, and monitoring. Artificial Intelligence(AI) and Machine-learning(ML) can provide life-saving care in general and CardioVascularDiseases(CVD) specifically as high risk conditions can be identified early, impending strokes and heart attacks spotted in advance. Studies have established that AI is set to revolutionize cardiac care. This is particularly relevant in today’s context, as India is home to 16% of the global population, 25% of the world’s Coronary Heart Disease (CHD) burden, 120 million hypertensives, and a large number of individuals with Rheumatic heart disease (RHD). CVD will be the leading cause of morbidity and mortality in the country by 2020. Researchers have started meshing up Big Data with AIand machine learning algorithm to predict the anomalies quickly, cheaply and accurately without using the invasive methods.However, implementation of such advanced technology in primary healthcare delivery is still in its nascent stage in last mile communities. iKureusingAI and ML into their primary healthcare delivery, has taken cardiac care to advanced level of treatment.
The Cardiac Risk Model: The initiative is jointly done by iKure and the IBM Data +AI elite team that work on ECG signal anomaly detection and Patient risk model using stack ranking & deep learning. By leveraging iKure’s cloud based platform, it brings the power of AI with respect to patient’s clinical data stored in an AWS, MySQL data store to develop high risk pattern and can predict cardiac arrest in patients in much earlier stage.
Data is knowledge and when big Data from EKG signals coupled with a large volume of patient data from IBM cloud, cardiologists will get empowered to deal with cardiac problems through risk ranking of cardiac patients. It could also classify the EKG data as normal or with anomalies and present a prioritized list of patient cases to the call center physician for review based on potential acuteness via a cloud-based web application. It will help them to treat patients based on acuteness. In that way resources are utilized in the best possible way. Proof of concept is already developed that demonstrated that the platform can bring the power of AI to iKure’s data stored in an AWS, MySQL. The pilot project is showcased at IBM’s technology conference, Think 2019 in San Francisco. Accuracy of this model is already proved using patient clinical and demographic variables and physician feedback with the added benefits of rapid model development, publication and iteration.
Analysis of arrhythmia anomaly to predict cardiac arrest
Conclusion: The analytics help screening of vital parameters including Heart Rate monitoring, a disorder in application sequence, arrhythmias, an electric axis of heart, myocardial ischemia and infarction and cardiac rhyme diagnosis vital for the prevention of early cardiac arrest. Leveraging iKure’s integrated model together with machine learning algorithm and Artificial Intelligence will enable specialist to have a more logical approach to diagnosis and treatment of cardiac arrhythmia. The output of such effort will enhance continuity of cardiac care even in last mile communities.
Showing 1 of 27