Predict heart attack with good accuracy using Artificial Intelligence

  • Text

Heart Attack Prediction Using Artificial Intelligence

We come up with unique approach of using artificial intelligence to detect heart attack or cardiac arrest based on parameters such as pulse rate, blood pressure, heart rate and respiratory rate. Implemented this idea using effective Artificial Intelligence technique, that is Artificial Neural network with Feed Back Propagation algorithm.

Product Overview

The World Health Organization (WHO) has estimated that 12 million deaths occur worldwide, every year due to the heart disease. About 25% deaths in the age group of 25-69 year occur because of heart diseases. In urban areas, 32.8% deaths occur because of heart ailments, while this percentage in rural areas is 22.9. Over 80% of deaths in world are because of Heart disease. World Health Organization estimated by 2030, almost 23.6 million people will die due to Heart disease. Our research is to detect heart attack using Artificial Intelligence(AI) based on heartbeat, pulse rate and blood pressure in order to reduce death rate due to heart attack. There are several ways doctor can predict heart disease based on the parameters such as cholesterol levels, blood pressure, blood glucose levels and weight. We are trying to develop a method that will optimized to use less conventional equipment’s as much as possible but also maintain the accuracy of detection. We are trying to avoid conventional methods as they are time consuming but we are trying to detect the heart attack pain as soon as possible so that the patient can have enough time to react. Here we come up with new methodology where heart attack can be predicted based on parameters such as pulse rate, heartbeat and blood pressure. Our Heart condition can be measured by important vital signs such as heart beat and pulse rate. People whose pulse rate creped from under 70 beats per minute at the first reading to more than 85 beats per minute at the second measurement were in critical condition. People who started out pulse rates between 70 and 85 beats per minute were also at risk of heart-related death; if their heart rates rose beyond 85 beats per minute by the second reading, they had an 80% increased risk of dying from heart disease, compared with people whose heart rates remained stable. Experts say healthy adults can have pulse rates ranging from 60 to 100 beats per minute. Elite athletes typically have lower heart rates, around 40 beats per minute because of their better heart fitness. The main purpose of this paper is to build a heart rate prediction model, which is based on real-life heart rate. Here we use Artificial Intelligence (AI) algorithms to train data and search for patterns and can build internal guidelines. Here we will use sensor to check heart rate of a person based on the heart rate, system will predict heart attack based on parameters such as pulse rate, heart beat and blood pressure.

There are other vital signs which are standard in most medical settings:

1) Pulse rate     2) Respiratory rate      3) Blood pressure    

Pulse rate
"Pulse" defined as the rhythmic expansion and contraction of the arteries corresponding to each beat of heart. Therefore, pulse rate is the measurement of human heartbeats. Pulse rate can be measured either in the wrist of neck given by beats/min. The most prominent spots for measuring the pulses are wrist (Radial artery), neck (Carotid artery), inside of the elbow (Brachial artery), behind the knee (Popliteal artery) and ankle joint (Posterior tibial artery). Pulse rate is very helpful to determine the problems of human body but it is not used to diagnosis the problem. This rate is varying with ages and depends on physical and psychological effect on the body. If the rate of pulse is higher, it indicates the availability of abnormality in the body. It can also be caused by other reasons, e.g. anxiety, anger, excitement, emotion, asthma, heart disorder and so on.
Respiratory rate
Respiratory Rate, which a person takes the number of breaths within a certain amount of time or more. It also defined as the number of chest movements involving inspiration and expiration per unit time. Respiratory rate will increase if the demand of oxygen, due to illness, intensive physical activity is increased. The average respiratory rate for a healthy adult at rest is 12/60 Hz and it varies between 12-20 breaths/min. For the babies, young adults the rate is higher than adults rate.
The following table shows the pulse rate and respiratory rate for different ages.

The following table shows the pulse rate and respiratory rate for different ages
Blood pressure
Blood pressure is the most important vital sign for a human body. It is a force exerted by blood on the walls of arteries, veins and the chambers of the heart and it varies between a maximum pressure called systolic pressure and a minimum pressure called diastolic pressure. The average value of healthy adult is 120 mmHg during the systole and 80 mmHg during diastole. There have some factors, which affect the blood pressure of a healthy person; they are pumping rate, blood volume, viscosity etc.
  • Pulse Rate: Pulse rate is the measurement of human heartbeats which will be in the form of quantitative information.
  • Respiratory Rate: Respiratory rate is the quantitative information which defines number of chest movements involving inspiration and expiration per unit time.
  • Blood Pressure: Blood pressure is a force exerted by blood on the walls of arteries, veins and the chambers of the heart and it varies between a maximum pressure. In this study blood pressure is the quantitative information.
  • Predicts the probability of heart attack which will be in quantitative information.
  • Algorithm Used
  • We will use Multi-Layer Perceptron (MLP) and Backpropagation algorithm which will be used to train the data
  • However, we will concentrate on nets with units arranged in layers
  • Multi-Layer Perceptron Methodology works on non-linear separable data

  • Following are steps of algorithm
  • Initialize weights at random, choose a learning rate n
  • Until network is trained:
  • For each training example (input pattern and target outputs):
  • Present inputs for the first pattern to the input layer
  • Sum the weighted inputs to the next layer and calculate their activations using activation function formula
  • Present activations to the next layer, repeating (2) until the activations of the output layer are known
  • Compare output activations to the target values for the pattern and calculate deltas for the output
  • Propagate error backwards by using the output layer deltas to calculate the deltas for the previous layer
  • Use these deltas to calculate those of the previous layer, repeating until the first layer is reached
  • Calculate the weight changes for all weights and biases (treat biases as weights from a unit having an activation of 1)
  • If training by pattern, update all the weights and biases, else repeat the cycle for all patterns, summing the changes and applying at the end of the epoch
  • Conclusion

    This modern age is era of technology. There are numerous heart attack detection techniques which is expensive and time consuming. In our study we use various parameters to detect heart attack which will provide the accurate outcome. There are many researchers, who dedicated their whole life to find out the latest technology for medical applications. Such as Shnayder who denotes his life to find health supporting system for supporting people in chronic conditions. Our application is very efficient and suitable. By this system patients can easily take precaution for heart attack. Avoiding health risks can be more efficient than sustaining patients with chronic conditions that could have been avoided.

    Related Products