Machine Intelligence to Predict Health Issues

  • Text

Study on Machine Intelligence to Predict Health Issues

We develop system to reduce health risk using Artificial Intelligence. Effective Artificial Neural Network technique used to predict multiple health related issues. In this study we researched on Multi-Layer Perceptron (MLP) and Backpropagation algorithm to work on non-separable data.

Product Overview

Here we come with an idea to use Artificial Intelligence to detect health issues in human body based on certain variables. This System can definitely assist physicians to make better clinical decisions or even replace human judgement in certain functional areas. This system can be utilized to predict various health issues such as heart disease, diabetes, head-ache, depression and anxiety. The main objective of this system is to predict multiple disease. We had done deep research to predict multiple heath issues with different health parameters. The unique idea is to predict multiple health issues by using one product. In Artificial Intelligence, it is necessary to learn features from large volume of health-related data. We apply sophisticated Artificial Intelligence algorithm to work on large data and to obtain insights to assist clinical practice. The system extracts useful information from a large patient data to assist in making real-time inferences for health issue alert. Here we researched artificial intelligence techniques that extracts information from unstructured data generated from medical reports. In this study, data need to be trained that are generated from clinical reports such as screening, diagnosis, treatment assignment etc. Data Analytical algorithms used to extract pattern from data. Input data will be patient medical report or disease specific data such as EP test, Physical examination results, clinical symptoms, medication and so on. There are many Artificial Intelligence algorithm here we use Artificial Neural Network methodology. In this study we implemented Artificial Neural Network to predict health risk with minimal error. In neural network, the associations between the outcome and the input variables are depicted through multiple hidden layer combinations of prespecified functionals. The goal is to estimate the weights through input and outcome data so that the average error between the outcome and their predictions is minimized. Artificial Intelligence system helps physicians to make clinical decision points.


  • To extract information from unstructured data
  • To predict health risk with minimal error
  • To predict multiple disease
  • To make real-time inferences
  • To learn features from large volume of data

  • Clinical test variables such as Hematology, Biochemistry, Serology, All Microscopy Tests etc. which will be in quantitative information
  • Hematology: Diagnosis related to blood and its component such as blood cells, hemoglobin, blood proteins, bone marrow, platelets, blood vessels, spleen, and the mechanism of coagulation.
  • Biochemistry: Biochemical tests performed on body fluids and tissues, to support diagnosis, treatment and monitoring of disease. Biochemical test such as Blood Sugar AC, S Cholesterol, Calcium, S Protein etc.
  • Serology: Serology refers to the diagnostic identification of antibodies that are typically formed in response to an infection. Serology test such as Mantoux Test, L E Cell test, Widal Test, STO etc.
  • Microscopy Tests: Microscopy Tests is identification of parasite or its antigens/products in the blood of the patient
  • Predicts multiple health related issues which will be in qualitative 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

    Artificial Intelligence is expanding role in modern medicine. Artificial Intelligence has huge potential to transform medical diagnosis. Implementing Artificial Intelligence in medicine has to face other challenges such as cost, resources and data. In this study we require various health related data to be analyzed to develop Artificial Intelligence System to predict health issues. It is necessary to survey the major disease types to deploy Artificial Intelligence System. Artificial Intelligence to predict health issues requires real life medical data for making better decisions