Back-Propagation Neural Network Versus Logistic Regression in Heart Disease Classification

Desai, Shrinivas D.; Chickerur, Satyadhyan; Narayankar, Prashant; Pudakalakatti, Neha R.; Sulegaon, Shreya

doi:10.1007/978-981-13-0680-8_13

Cited by 48 publications

(20 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Wearable health monitoring devices and monitoring systems have the following characteristics, light weight, removable, easy to carry, small size, intelligent display, wireless transmission, abnormal alarm, support for a long time, etc., and effectively apply the modern technology represented by sensor technology, wireless communication technology, chip technology, etc. In terms of the present stage, the scientific research personnel of wearable devices, as well as commercial companies, are given the focus on research and development work, also appearing on the market series of home health care equipment, such as household, particle size glucose meter, blood pressure, heart rate, and so on, devices have wireless data transmission function, based on the intelligent terminal, can realize the data real-time transmission and feedback [ 10 ]. Wearable devices not only meet the needs of users for medical and health monitoring, but also can not affect their daily life [ 11 ], so they are mostly in the form of chest bands, rings, clothing, watches, and so on.…”

Section: Literature Reviewmentioning

confidence: 99%

“…For this reason, it must be portable, so that users can participate anytime, anywhere. With the progress of society, portable smart devices have become widespread, mainly smart phones and tablet computers [ 10 ]. Therefore, it is most convenient to reflect the way that users receive platform services through mobile intelligent devices.…”

Section: Literature Reviewmentioning

confidence: 99%

See 1 more Smart Citation

Research on Smart Healthcare Services: Based on the Design of APP Health Service Platform

Jiang

Lei

2021

Journal of Healthcare Engineering

View full text Add to dashboard Cite

With the development of information technology, big data, and cloud computing, the concept of smart healthcare has gradually become more and more important. Compared with the traditional healthcare service, the new model, health service platform, is becoming increasingly popular and convenient. The use of wearable monitoring devices and some APPs is improving the health monitoring efficacy and effectiveness. To improve and facilitate the smart healthcare services, an effective and convenient app health service platform is needed urgently to serve the older and younger. Based on the above, this paper elaborates the principles of health service system and health information perception terminal design of the APP health service platform.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Section: Literature Reviewmentioning

confidence: 99%

Research on Smart Healthcare Services: Based on the Design of APP Health Service Platform

Jiang

Lei

2021

Journal of Healthcare Engineering

View full text Add to dashboard Cite

show abstract

“…Classification methods are the mainly extensively used algorithm in the healthcare field because it helps predict the status of patients by classifying patients' records and locating the class that matches the new patient records [7], [8]. This paper presents an introduction to the classification algorithm used in our comparative study of heart disease prediction models using Naïve Bayes (NB), KNN, SVM, LR, Decision Tree (DT), Random Forest (RF) classifier on healthcare dataset and diagnostic the patients' ailment from patients is inflated Becomes challenging when training and testing data is from a different domain [9], [10].…”

Section: Machine Learning Based Approachmentioning

confidence: 99%

“…LR [9], [17], [30] provides high accurateness plus chart illustration. In this algorithm, facts must be imported primary and then trained.…”

Section: Logistic Regression Classifier (Lr)mentioning

confidence: 99%

A Relative Research of Machine Learning Build Heart Disease Forecast Model

Agrawal¹,

Soni²,

Gupta³

2020

SKIT Research Journal

View full text Add to dashboard Cite

Nowadays, heart disease is the leading cause of death globally, with an estimated 610000 lives each year due to heart condition. One of the most common causes of heart disease is high blood pressure (HBP), fasting blood sugar (FBS), diabetes, cholesterol, Body mass Index (BMI), heart rate (HR). Diagnosis of heart disease is more prevalent nowadays; this involves a lot of accuracy and uncertainty due to the large-scale data decision-based on doctors may fail in some cases. Data mining is an intelligent diagnostic tool in healthcare. Thus, it is imperative to predict that each menace stage depends on age, sex, blood pressure, diabetes symptoms, what we can do for precaution by diagnosing the disease and proper treatment at the right moment. The purpose of the research work is to develop different predictive models using different forecasting measures and perform comparative analysis. In this work, we have used Cleveland and Statlog datasets with Naive Bayes (NB), K-Nearest Neighbors (KNN), and Logistic Regression (LR), Support vector machine (SVM), Decision Tree (DT), and Random Forest (RF) Classifier to develop various predictors. The experiment result shows that the random forest classifier gives better accuracy and results on both datasets when we perform a comparative analysis of them among all other classification models

show abstract

“…Kemudian terdapat juga beberapa metode kombinasi dalam mengklasifikasi penyakit jantung (clevelnad heart disease) yaitu algoritma kombinasi CSF-PSO-MLP mencapai akurasi 90% [12], algoirtma kombinasi GA-SVM-SMO-RBF mencapai akurasi 88% [13]. Metode backpropagation sendiri mencapai akurasi 85% [14]. Kemudian beberapa kombinasi algoritma backpropagation yaitu DE-Backpropagation mencapai akurasi 80%, algoritma PSO-Backpropagation mencapai akurasi 83%, algoritma DEGI-Backpropagation mencapai 87% [15].…”

unclassified

Peningkatan Akurasi Klasifikasi Backpropagation Menggunakan Artificial Bee Colony dan K-NN Pada Penyakit Jantung

Putra

Sukemi

Rini

2021

mib

View full text Add to dashboard Cite

Heart disease has ranked as the leading cause of death in the world, accounting for around 17.3 million deaths per year with some causes, as high blood pressure, diabetes, cholesterol fluctuation, fatigue, and some others which is collected on dataset. Heart disease dataset that was applied is cleveland heart disease with fourteen (14) data atribute. The method that was applied in this research was using Backpropagation algorithm on heart disease classifying, where will be combined Artificial Bee Colony and k-Nearest Neighbor algorithm for features or atribute choose due to this technique can increase classifier model accuracy which is produced as much as 94,23%.

show abstract

Back-Propagation Neural Network Versus Logistic Regression in Heart Disease Classification

Cited by 48 publications

References 12 publications

Research on Smart Healthcare Services: Based on the Design of APP Health Service Platform

Research on Smart Healthcare Services: Based on the Design of APP Health Service Platform

A Relative Research of Machine Learning Build Heart Disease Forecast Model

Peningkatan Akurasi Klasifikasi Backpropagation Menggunakan Artificial Bee Colony dan K-NN Pada Penyakit Jantung

Contact Info

Product

Resources

About