A Novel Software Engineering Approach Toward Using Machine Learning for Improving the Efficiency of Health Systems

Abstract

Recently, machine learning has become a hot research topic. Therefore, this study investigates the interaction between software engineering and machine learning within the context of health systems. We proposed a novel framework for health informatics: the framework and methodology of software engineering for machine learning in health informatics (SEMLHI). The SEMLHI framework includes four modules (software, machine learning, machine learning algorithms, and health informatics data) that organize the tasks in the framework using a SEMLHI methodology, thereby enabling researchers and developers to analyze health informatics software from an engineering perspective and providing developers with a new road map for designing health applications with system functions and software implementations. Our novel approach sheds light on its features and allows users to study and analyze the user requirements and determine both the function of objects related to the system and the machine learning algorithms that must be applied to the dataset. Our dataset used in this research consists of real data and was originally collected from a hospital run by the Palestine government covering the last three years. The SEMLHI methodology includes seven phases: designing, implementing, maintaining and defining workflows; structuring information; ensuring security and privacy; performance testing and evaluation; and releasing the software applications. INDEX TERMS Health dataset analysis, machine learning, methodology, software development management, software engineering.

A Secure Encrypted Classified Electronic Healthcare Data for Public Cloud Environment

“…Feature selection is an important process in healthcare data analysis. Based on the selected features, classification is performed so that individual risk and preventive measures can be provided [7,8]. Similar to feature selection, feature reduction is also an important process in healthcare data analysis [9].…”

A Supervised Learning Identification System for Prognosis of Breast Cancer

“…In general, black-box techniques are difficult to comprehend, but logistic regression demonstrates how they work. Logistic regressions are classified into three types: binary, multinomial, and ordinal [32]. Logistic regression originated as a classification technique rather than a linear regression.…”

Corrections to “A Novel Software Engineering Approach Toward Using Machine Learning for Improving the Efficiency of Health Systems“

“…The authors would also like to add another co-author, Dr. Oguz Ata, to improve our honesty and loyalty for above paper [1], as he worked with authors on analyzing the data, and later he tested the framework, furthermore, he participated with valuable feedback and comments.…”