About us      |      Background      |      Creators      |      Acknowledgments      |      FAQ


ABOUT US

Our team is comprised of clinical cardiologists, bioinformatics scientists, and PhD researchers from other fields. We are committed to researching the use of registry data to predict mortality and cardiac events in heart disease patients. In addition, we used air quality data in conjunction with cardiac factors to predict hospital admissions and cardiac deaths.

We hope with this unique calculator, we can help prevent avoidable deaths in the Malaysian population with heart attacks.


Why did we develop?

In the past 50 years, Malaysia has undergone a rapid evolution in terms of health and disease. Communicable diseases are no longer the leading cause of death, with heart disease taking the top spot for the past decade.

Unstable Angina (UA), Non-ST-Elevation Myocardial Infarction (NSTEMI), and ST-elevation Myocardial Infarction (STEMI) are the three types of heart attacks. STEMI patients have the highest mortality risk among these individuals. Nevertheless, current prediction tools are based on Western populations and may not fully reflect the Malaysian population's specific risk variables.

However, the existing predictive tools, primarily based on Western populations, may overlook the unique risk variables present within our Malaysian community. That's why our team has developed a region specific machine learning cardiovascular calculator. This tool accurately predicts ACS Risk and Mortality, leaning on insights from the NCVD cohort, along with Cardiovascular Risk Diseases (CVD) based on the REDISCOVER cohort while using a recent dataset from 2019. Uniquely, we have incorporated air quality factors into our risk model, offering a comprehensive understanding of environmental impact on cardiovascular health in Malaysia. With this novel step, we can give a more complete picture of the factors that affect cardiovascular health in Malaysia.

At the same time, we also wanted to find out about Cardiovascular Risk Diseases (CVD) by using the REDISCOVER cohort's valuable data. This comprehensive method makes sure that our predictions are comprehensive and complete, taking into account the unique health patterns and trends of our population.

The death rate for ACS patients fluctuates between 5 and 8 percent in the Western population. In Malaysia, however, the mortality rate from the NCVD registry is double at 10 to 12 percent. By giving an accurate risk calculation suited to the Malaysian population, our calculator can promote improved patient treatment and aid in the prevention of preventable fatalities.

Our population is different from the aspect of genetics, dietary, environment, risk factors and healthcare facility. Hence, we may have different risks contributing to the higher death rate that is unbeknown to us. A locally produced calculator helps to provide an accurate estimation to facilitate better patient care. We hope that our novel approach to risk assessment can enhance patient outcomes and inform public health policies and measures to lower the prevalence of cardiovascular disease among the Malaysian population.

In this revised version, we have added information about the various cohorts that the website can predict, as well as how incorporating air quality factors into the risk assessment model can assist in gaining a better understanding of the impact of environmental factors on cardiovascular health in Malaysia. In addition, we highlighted how a locally-produced calculator may provide an accurate risk calculation that is suited to the Malaysian community, resulting in improved patient care and fewer preventable fatalities.

  • Patients with ACS are at high risk of death in Malaysia.

  • Thrombolysis in Myocardial Infarction (TIMI) is used for 30 days mortality risk prediction in Malaysian hospital.

  • Derived mainly from Western Caucasian pool and may not reflect our diversity.

  • In Malaysia, patients presenting with STEMI are younger than the Western cohort.

  • These risk scores are not population-specific and may not be able to account nuances related to a specific region.

  • The machine learning calculator for STEMI may help a cardiologist, health care practitioners with evaluating risk/benefit of invasive and non-invasive procedures by knowing a patient’s baseline risk.

  • Air pollution is a significant risk factor for cardiac events and mortality.

  • To better understand the effect of environmental factors on cardiovascular health in Malaysia, our website incorporates air quality variables into the risk assessment model.

  • Better preventative and preparatory measures, thereby saving lives and helping to the accomplishment of the Sustainable Development Goals (SDGs).

  • Poor air quality may result in increased hospital admissions and cardiac mortality, highlighting the importance of addressing this issue in the context of cardiovascular health in Malaysia.

    • Back to Top ^

How it started?

    APRIL 2018
  • Preliminary study on the application of Machine Learning to predict mortality in Malaysian patients using sample dataset from UiTM NCVD ACS Registry.

    • APRIL 2019
  • Large scale study using National Cardiovascular Disease Registry (NCVD) to predict mortality after STEMI using machine learning algorithms and validated against the TIMI Risk score. Machine learning algorithm performed better than TIMI in the Malaysian population.

    • SEPTEMBER 2019
  • Developed an online STEMI mortality calculator.

    • MARCH 2020
  • Plan and development of a more advanced online ACS and CVD mortality calculator.

    • March 2022
  • Preliminary study on the application of Machine Learning encomprises Air Quality data and Cardiac Features.
  • Study on the mapping the prediction model on Map.

    • APRIL 2023
  • Developed an A.I. Predictor of Cardiac Events and Mortality for Malaysia.

    • JUNE 2023
  • Initiated system testing and validation to ensure the efficiency and accuracy of our tool.

Back to Top ^

BACKGROUND


MyHeart - STEMI

  • Retrospective data from the national cardiovascular disease database (NCVD-ACS) registry collected by the Ministry of Health Malaysia and National Heart Association Malaysia (NHAM) from 2006 – 2017 was utilized.

  • The registry collects data on a standardized set of clinical, demographic, and procedural variables, along with outcomes, for consecutive patients treated at participating institutions.

  • The calculator includes in hospital, 30-day, and 1-year all-cause mortality.

  • All patients from the ACS registry without exclusion were used; patients who received reperfusion (fibrinolysis, primary PCI (PPCI), angiography demonstrating spontaneous reperfusion, or urgent coronary artery bypass grafting (CABG)) for STEMI.

  • This score also included patients with left bundle branch block.

  • Support Vector Machine was used as the primary algorithm for risk score stratification of the STEMI patients.


    • MyHeart - Air Pollution and Cardiovascular
    • Retrospective data from the national cardiovascular disease database (NCVD-ACS) registry collected by the Ministry of Health Malaysia and National Heart Association Malaysia (NHAM) from 2006 – 2017 was utilized.

    • Air Quality data accquire from Department of Environment (DOE), Malaysia from 2005--2017 was combined with NCVD-ACS data.

    • Comprises 5 main features in Air Pollution and Cardiovascular section.

    • Visualize the geolocation of the hospital together with predicted hospital admission and cardiac death rate.

    • A.I. Calculator that predicts the mortality probability of Warded patients and Emergency patients based on Cardiac and Air Quality features.

      • MyHeart - Acute Coronary Syndrome (ACS)
      • Retrospective data from the national cardiovascular disease database (NCVD-ACS) registry collected by the Ministry of Health Malaysia and National Heart Association Malaysia (NHAM) from 2006 – 2019 was utilized.

      • The registry collects data on a standardized set of clinical, demographic, and procedural variables, along with outcomes, for consecutive patients treated at participating institutions.

      • The calculator includes in hospital, 30-day, and 1-year all-cause mortality.

      • Able to predict the ACS mortality probability of Warded patients and Emergency patients.

      • Support Vector Machine was used as the primary algorithm for risk score stratification of the ACS patients.


          • MyHeart - Cardiovascular Diseases (CVD)
        • The REDISCOVER (Responding to Increasing Cardiovascular Disease Prevalence) was utilized for 10-Year CVD risk prediction.

        • The study includes participants aged 18 and up from eight different states across Malaysia with a follow-up for every three years (2007 – Ongoing).

        • Outcomes are risk of developing CVD event in 10 years (Myocardial Infarction, Stroke, Heart Failure)

        • Logistic Regression was used as the primary algorithm.

        Back to Top ^

CREATORS

  • Prof. Sazzli Kasim - Director Hospital Al-Sultan Abdullah UiTM (HUITM). Consultant Cardiologist at UiTM Specialist Center.

  • Dr. Sorayya Malek - Senior Lecturer at Bioinformatics unit, Institute of Biological Sciences, Faculty of Science University of Malaya.

  • Dr. Khairul Shafiq Ibrahim - Lecturer, Faculty of Medicine, UiTM Selangor Branch, Sungai Buloh Campus, Sungai Buloh.

  • Dr. Firdaus Aziz - Postgraduate student at Bioinformatics unit, Institute of Biological Sciences, Faculty of Science University of Malaya. Lecturer, Universiti Kebangsaan Malaysia (UKM) (current)

  • Ms. Putri Nur Fatin - Postgraduate student at Bioinformatics unit, Institute of Biological Sciences, Faculty of Science University of Malaya.

  • Ms. Song Cheen - Postgraduate student at Bioinformatics unit, Institute of Biological Sciences, Faculty of Science University of Malaya.

  • Ms. Nurulain Ibrahim - Postgraduate student at Faculty of Medicine, UiTM Selangor Branch, Sungai Buloh Campus, Sungai Buloh.

    • Back to Top ^

ACKNOWLEDGEMENTS

    We would like to acknowledge the contributions of several organizations that have been integral to our project's success:

    Logo1

    We extend our sincere gratitude to the Minister of Science Technology and Innovation, Malaysia. Your support has been crucial to the advancement of our study. Thank you for your commitment to scientific progress and innovation.

    Logo2

    We sincerely appreciate the National Heart Association of Malaysia for providing the invaluable study dataset and contributing to the development of our cardiac calculator. Your collaboration has been instrumental in the progression of our research. Thank you for your ongoing support.

    Logo3

    We express our sincere thanks to the Department of Environment, Malaysia, for supplying the essential air quality data. This information has significantly contributed to our study and tool development. Your support is highly appreciated.

    Logo4

    We gratefully acknowledge the support from Hospital UiTM. Their assistance and cooperation have been essential in the progress of our research. We thank them for their invaluable contribution.

    We respectfully thank the REDISCOVER Study Committee at the Centre for Translation Research and Epidemiology (CenTRE), UiTM, for supplying the crucial study dataset and helping to create our cardiac calculator. Your contributions have been critical to the advancement of our research. We appreciate your continued assistance.

    Back to Top ^

    FAQ

    For more information about our model, including details about its development and the factors it takes into account, please refer to the FAQ section below.

  • MyHeart ACS FAQ

  • MyHeart CVD FAQ

  • MyHeart Air FAQ

    • Back to Top ^