A

Research Paper

ON

“Incurable Disease Prediction System”

Submitted in the partial fulfillment of the requirement for the award of the

Degree
Of
Bachelor of Technology

In
Department of Computer Science & Engineering

SESSION: 2019 – 20

Submitted By:- Submitted To:-

Kuldeep Yadav Dr. Rohit Kumar Singhal
Mohit Bhargava Professor & HoD (CSE)
Maneesh Kumar
Pragya Narang
 Team Members:-
Name of Team Member Address of Team Member Email Address of Team Member Mob. No of Team Member

Kuldeep Yadav Alwar ky41666@gmail.com 8742084770

Mohit Bhargava Alwar

Maneesh Kumar Bharatpur

Pragya Narang Dausa

Abstract:-
Incurable diseases are the deadly diseases in the world. These has become the primary reason of death in
developing and developed both countries. The most effective way to reduce incurable diseases death is to detect
it earlier.The earlier detection of these is not easier process but if it is detected, it is curable. Many works have
been done in predicting these diseases; different data mining and machine learning approaches and algorithms
were adopted by different people. Each work has some limitations such as lack of intelligent prediction, and
inefficient in structure that motivated to take up this problem and to implement Machine Learning based
Incurable Diseases Prediction System. This system estimates the risk of the incurable diseases.The main aim of
this model is to provide the earlier warning to the users, and it is also cost efficient to the user.
Introduction:-
Health informatics is one of the greatest challenges and machine learning is one of the growing fields
in the computer science. Elec-tronic health records (EHR) information contains data of large number of
patients. The challenge with the huge data is how to analyze the data and infer the intelligence.
This is also known as Rapid Learning Health Care (RLHC) where a large amount of data that is
continuously growing needs to be analyzed to predict accu-rate model for prediction of health risks
and diseases. Challenges in representing patient data and analyzing it to produce meaningful
information are hurdle for predictive modelling using EHRs. Recent progress of machine learning
algorithms in designing and applying it on EHR dataset has shown promising results. However, majority
of techniques rely on labelled data for accurate prediction of results. We have proposed the incurable
diseases prediction system based on data mining and machine learning algorithms. Incurable diseases
prediction system estimates the risk of the incurable diseases.This system was validated by comparing
its predicted results with patient’s prior medical information and analyzed using Machine Learning
Algorithms. Detecting Incurable Diseases is still challenging for the doctors in the field of medicine.
Even now the actual reason and complete cure of these is not invented. Various tests are available for
predicting these diseases but detecting them in earlier stage is difficult, but earlier detection of cancer
is curable.

The goal of machine learning methods to learn and improve over the time can be used for predictions.
a) Observations:- Mining of electronic health records (EHRs) could yield insightful information that
can help to improve health risk predictions, thus improving the clinical decisions. A main objective
of precision medicine is to develop quantitative models for patients that can be used to predict
health status, so also help prevent disease or disability.

b) Contributions:- To summarize, this paper makes following contributions:-

• Classifies categories of machine learning methods.
• Classifies state-of-the art research in health risk prediction systems or disease prediction systems.
• Provides future research directions.
Architecture of Incurable Disease Prediction System:-
In this work, an architecture Machine Learning technique based Incurable Diseases prediction
system combining the prediction system with mining technology was used. In this model we have
used one of the classification algorithms.

Fig.02-Concept of Machine Learning Categorization

Fig.01-Architecure of Incurable Disease Prediction System

Process:-
This section describes the machine learning landscape.and basic concepts. Machine Learning methods facilitate development of the
intelligence into a machine, so that it can perform better in the future using the learned experience. Fig. 02 shows two ways to
categorize machine learning methods namely: Learning Categorization and Learning Problems Categorization. Machine learning
algorithms or techniques categorization using learning techniques contains five sub-fields namely: Supervised learning, Unsupervised
learning, Semi-supervised learning, Reinforcement learning, and Deep learning.

a) Supervised Learning: Supervised learning methods work on known expectations on the labelled input dataset. Methods
in this category aim at establishing a relationship between the input and output attributes of the labelled dataset. Labelled data helps
to build more reliable and accurate model,however, it is computationally expensive process.

b) Unsupervised Learning: These methods target to analyze the structure of data in the provided input unlabelled dataset and
build mapping between the input and output attributes, whereas output attributes are unknown before the analysis.

c) Semi-supervised Learning: These methods used both labelled and unlabelled datasets to develop models for intelligence inference.

d) Reinforcement Learning: These methods goal is to maximize the rewards from the result. That is reinforcement learning method
produces a sequence of decisions that help to acquire highest rewards.

e) Deep Learning: These methods focus on unifying artificial intelligence with machine learning. It works on common data to provide
meaningful insights. It works on input dataset that has less labelled data and solves problems classified under semi-supervised
learning to build complex neural network models.

Machine learning algorithms or techniques are also classified using learning problems as: Classification, Clustering, Optimization, and
Regression.
a) Classification: It is a grouping technique that depends on the given value of target and dataset. According to the provided target
value it quantifies and classifies the dataset.

b) Clustering: It is a technique that takes only dataset as an input and identifies interesting patterns to derive intelligence. As
compared to classification, in the clustering target value is not provided as an input or it is an unknown parameter.

c) Regression: In this technique intelligence or information is derived from the past learning experience. An equation is derived that
matches with most of the data points and the cases where data doesn’t fit with the curve, those data points we discard. This
technique is known as regression.

d) Optimization: It is a method to improve the performance of the system in terms of various attributes.
Result:-
The result of this researchpaper is that after this research we would be able to help the people before they are suffered by the dieseas
and we will be able to prevent the incurable diseases which causes a human being to death. We will predict Incurable Diseases with the
help of Machine Learning analyzing the previous reports of the patients. Thus, we would be able to give them treatment for the
incurable diseases by preventing them.

Conclusion :-
In this paper, we focus on exploiting machine learning methods and its applications in prediction of health risks and healthcare
systems.
The objectives of this research review paper are fourfold:
• serve as a guideline for researchers who are new to machine learning area and want to contribute to it,
• provide state-of-the-art survey of machine learning,
• application of machine learning techniques in the health prediction, and
• provides further research directions required into health prediction system using machine

References:-
•Our teachers helped in this a lot and give us some valuable suggestions to improve our predictions system.
•We also took the help of the google browser and youtube
•Last but not the least wikipedia was also helpful to us in determining the directions of this research papers.