Cardiovascular Disease Prediction

Cardiovascular Disease Prediction

How can a practitioner make a quick cardiovascular disease prediction?

Cardiovascular diseases (CVD) are a major cause of death. According to the World Health Organisation 17.91 million people die each year of CVD. Many of these deaths (which occur in 85% of cases as a consequence of a heart attack or a stroke) could be avoided with a quicker access to medical emergency care. Some could even be avoided with adequate drug treatment before the attack happens. But how can the medical staff know before the occurence of the heart attack that it’s time for these medical cares (drugs, hospitalisation)?

Technology and more specifically Artificial Intelligence (AI) can provide medical doctors, nurses, emergency room caregivers, general practitioners with new tools which help them better, quicker diagnose these diseases.

Problem to solve

  • Is it possible to predict if a patient is likely to have a heart attack knowing his/her age, gender, blood pressure, cholesterol level, maximum heart rate?
  • Can machine learning be used to anticipate a heart attack and put the patient under the relevant drugs before it happens? Or even drive the patient to the hospital right before the heart attack takes place?
  • How accurate is it to make such a prediction about heart attacks?
  • Is it even possible for a someone not trained in medicine in any way to identify a risk of heart attack?
  • Benefits of TADA
    for cardiovascular disease prediction

    Doctors, nurses, surgeons and medical staff in general are not Data Scientists. They usually do not have the required skills in machine learning nor in software coding to build predictive models. The data handled by these professionals are typically Small Data because they collect patient informations such as age, gender, blood pressure, cholesterol level, blood sugar level, electrocardiogram (ECG) results, average heart rate, maximum heart rate, for a few hundreds or thousands of patients. Small Data is the relevant category for data from several hundreds to at most thousands of data while Big Data is the term used for millions of data. Traditional machine learning algorithms work well with Big Data but do not perform with Small Data.

    MyDataModels allows medical health professionals to build predictive models from their Small Data automatically. These professionals do not need to undergo any kind of training to use our tools. Their raw data can be used directly without any form of preprocessing: no normalisation, nor outliers handling or feature engineering are required. The predictive results from processing these specific datasets can be obtained in a few clicks in less than a minute on a standard laptop. Thanks to its ease of use, TADA, MyDataModels software prediction application can be trained in the cloud or on any laptop. No specific equipment is required. Just bring in your Small Data and TADA will do the rest. It will generate a pertinent prediction function which can help you estimate the chance of a patient having a heart attack or a stroke.

    MyDataModels brings a self-service solution for those who have Small Data and no Data Scientists.

    TADA brings new possibilities for disease detection

    Heart attack and strokes are major causes of deaths today in the world, killing millions. The healthcare industry has collected and continues to collect large amounts of health related data. These data are stored and are left unused. They could easily be used on a daily basis by healthcare professionals in order to prevent a part of these CVD caused deaths. Patients could even in some cases be hospitalised before the stroke occurs.

    “Predictive model reached an 82% accuracy rate”

    MyDataModels has built such a heart attack prediction model using its engine TADA and has obtained an 82% accuracy rate. In other words, by entering a few past patient vital data into TADA, the engine has well predicted a high risk of heart attack in 82% of these past cases.

    Can you imagine how much TADA could help if used on future heart attacks?

    Doctors, nurses and health experts in general could easily use machine learning in a few clicks with their existing patient data as they are. This could help them identify a threat of heart attack or stroke which might get undetected today by those healthcare professionals not trained in heart diseases. It could save time. It could save money. But more importantly, it could save lives.