Disease Diagnosis and Treatment Using Deep Learning Algorithms for the Healthcare System

Disease Diagnosis and Treatment Using Deep Learning Algorithms for the Healthcare System

Nirbhay Kumar Chaubey, Prisilla Jayanthi
Copyright: © 2020 |Pages: 16
DOI: 10.4018/978-1-7998-2101-4.ch007
OnDemand:
(Individual Chapters)
Available
$37.50
Current Special Offers
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

This chapter explicates deep learning algorithms for healthcare opportunities. Deep Learning is a group of neural network algorithms and learns from various levels of representation and abstraction to aid in the data interpretation. Since the datasets get bigger, computers become more powerful, and the training of the datasets (images or numeric) gets much easier and the results achieved using deep learning are better. In contrast to machine-learning algorithms that rely on large amounts of labelled data, human cognition can find structure in unlabeled data, a technique known as unsupervised learning. It was noted that using deep learning algorithms on the dataset will reduce the number of unnecessary biopsies in future. In this chapter, the authors study deep learning algorithms to diagnose diabetic retinopathy retinal images and training a convolution neural network (CNN) algorithm to identify object tumors from a large set of brain tumor images.
Chapter Preview
Top

Support Vector Machine (Svm) Decision Trees - Logistics Regression – Prediction Of Heart Disease

The rule-based framework provided by (Mythili, 2013) shown in the figure 1 for predicting the heart disease using SVM decision tree and Logistic Regression. The method is based on six modules involving preprocessing, training, testing with individual models, rules for application, and comparison of results for the prediction of heart disease. The database was collected from Cleveland Heart Disease Dataset UCI repository with 13 features such as age, sex, chest pain type, resting blood pressure, serum cholesterol in mg/dl, fasting blood sugar > 120 mg/dl, resting electrocardiographic result, maximum heart rate achieved, exercise induced angina, ST depression induced by exercise relative to rest, the slope of the peak exercise ST segment, and number of major vessels (0-3) colored by fluoroscopy. All these feature helps in predicting the disease and trains the required model, here regression model and decision tree model is built. The results proved that there is a need for more complex models and combinations among the models to improve the performance and to increase the accuracy of predicting the heart disease. The system will be known to be intelligent with huge amount of data fed into the system database.

Figure 1.

Framework for predicting heart disease (Mythili, 2013)

978-1-7998-2101-4.ch007.f01

Complete Chapter List

Search this Book:
Reset