A machine learning model for improving healthcare services on cloud computing environment

doi:10.1016/j.measurement.2018.01.022

Measurement

Volume 119, April 2018, Pages 117-128

https://doi.org/10.1016/j.measurement.2018.01.022 Get rights and content

Abstract

Recently, cloud computing gained an important role in healthcare services (HCS) due to its ability to improve the HCS performance. However, the optimal selection of virtual machines (VMs) to process a medical request represents a big challenge. Optimal selection of VMs performs a significant enhancement of the performance through reducing the execution time of medical requests (tasks) coming from stakeholders (patients, doctors, etc.) and maximizing utilization of cloud resources. For that, this paper proposes a new model for HCS based on cloud environment using Parallel Particle Swarm Optimization (PPSO) to optimize the VMs selection. In addition, a new model for chronic kidney disease (CKD) diagnosis and prediction is proposed to measure the performance of our VMs model. The prediction model of CKD is implemented using two consecutive techniques, which are linear regression (LR) and neural network (NN). LR is used to determine critical factors that influence on CKD. NN is used to predict of CKD. The results show that, the proposed model outperforms the state-of-the art models in total execution time the rate of 50%. In addition, the system efficiency regarding real-time data retrieval is greatly improved by 5.2%. In addition, the accuracy of hybrid intelligent model in predicting of CKD is 97.8%. The proposed model is superior to most of the referred models in the related works by 64%.

Introduction

In recent years, cloud computing gained a great attention in HCS applications due to its ability to provide different medical services over the internet. Cloud computing allows applications to provide infrastructure services to big numbers of stakeholders with assorted and dynamically changing requirements [1]. Technically, cloud is composed of datacentres, hosts, VMs, resources, etc. Datacentres are containing a big number of resources and list of different applications. Hosts are composed of several VMs to store and regain several medical resources to stakeholders. Cloud computing uses the virtualization technique which permits to share a single physical instance of a resource or an application among various stakeholders and enterprises [2]. It does this by allocating a logical name to a physical storage and providing a pointer to that physical resource when requested.

Virtualization consists of hardware virtualization, operating system (OS) virtualization, server virtualization and storage virtualization in cloud computing environment. Hardware virtualization is mainly done for the host platforms, because controlling VMs is much easier than controlling a physical host [3]. Operating System Virtualization is mainly used for experimenting different applications on different platforms of OS [4]. Server virtualization can be divided into several physical hosts on the request basis. Storage virtualization is created to recovery purposes. Virtualization plays a very significant role in the cloud computing, stakeholders’ share the medical data in the clouds like medical applications etc. [5].

Currently, many healthcare applications that are used for diseases diagnosis or prediction does not support real time use which enable the stockholders to access them anytime and anywhere [6], [7], [8], [9], [10], [11], [12]. However, the time delay represents a big challenge for the most of stakeholders in HCS applications that run the medical requests on a cloud computing environment. In this paper, a new model for diseases diagnosis and prediction is proposed for CKD. According to the recent health services applications [5], [8], [11], surveys showed that CKD is one of the most serious diseases facing the world where the latest statistics are recorded 2.5–11.2% across Europe, Asia, Australia and North America are suffering from CKD. The United States of America has 27 million people and 50 thousand people in Egypt suffering from CKD. In addition, most available CKD diagnosis and prediction applications are based on traditional statistical methods which may lead to less accurate results.

Accordingly, the contribution of this paper is two-fold. First, a VMs optimization model is proposed using PPSO algorithm to improve the performance of HCS applications in a cloud computing environment. Second, a CKD diagnosis and prediction model is proposed to reduce the execution time of CKD prediction requests processing and speeding up reply to CKD prediction requests coming from stakeholders, and maximizing utilization of cloud resources. The proposed CKD model is implemented using as a hybrid schema composed of LR and NN. LR is used to determine critical factors of CKD. Then, NN is used for CKD prediction.

The reset of the paper is arranged as follows: Section 2 introduces the basics and the background information related to the algorithms used to develop our proposed model. The recent related work is discussed at Section 3. Section 4 describes the proposed cloud computing optimization model for VMs. Section 5 explains in details the proposed hybrid algorithms for CKD detection. Section 6 discusses the experimental results. Finally, Section 7 presents the conclusion and the future work.

Section snippets

Parallel particle swarm optimization

PSO has particles which perform elect solutions of the problem, each particle seeking for most favorable solution in the search space, each particle or candidate solution has a position and velocity. A particle updates its velocity and position based on its inertia, own experience and gained knowledge from other particles in the swarm, aiming to detect the best solution of the problem. The particles update its position and velocity according to the following Eqs. (1), (2) [13], [14]: $V_{1}^{k + 1} = {wV}_{1}^{k} + c$

Related work

Through related work, many studies were done on applying and using different optimization algorithms to determine optimal VMs on cloud environment. Previous work also introduces a set of studies based diagnosis of CKD on cloud environment, as follows:

Fang et al. [16], introduced a new framework to find the optimal VMs placement of cloud environment based on ant colony optimization (ACO). This study tries to detect VMs placement of datacentre in order to maximize quality of VMs, minimize energy

The proposed could computing based optimization model for VMs

This section describes the architecture of the proposed cloud computing model for HCS. It consists of four components are stakeholders’ devices, stakeholders’ requests (tasks), cloud broker and network administrator as shown below in Fig. 1. The communication devices services are responsible for implementing different network communication management between stakeholders and the cloud.

Stakeholders use a variety of devices (PC, Laptop, Smartphone, Tablet, Digital sensors, etc.) to send a variety

CKD diagnosis using LR analysis

This section introduces a LR analysis model for CKD diagnosis. LR analysis is used to specify the critical factors that affect CKD behavior. Fig. 3 shows the flowchart of the proposed algorithm. LR provides the mechanism for regression statistics such as the mean absolute error (MAE), the root mean squared error (RMSE), the relative absolute error (RAE), the relative squared error (RSE) and the coefficient of determination (CD). MAE is a quantity used to measure how close predictions are to the

Experimental results

This section discusses the experimental results of our proposed model. The model is implemented using two different tools. The first tool is CloudSim package which is used to implement the proposed PPSO for VMs optimization [38]. While the second one is the Windows Azure which is used to implement the hybrid LR and NN model [39].

Conclusion and future work

The stakeholders are facing a big challenge to during their interaction with HCS applications due to the limited resource and the time consumption. By determining the optimal VMs on cloud computing, HCS applications will be able to reduce the execution time. This paper proposes a new model for HCS in a cloud environment using PPSO to determine optimal selection of VMs. In addition, a hybrid model for predicting CKD based on cloud environment is proposed. The proposed CKD prediction model is

References (39)

L. Chen et al.
MTAD: A Multitarget Heuristic Algorithm for Virtual Machine Placement
Int J Distrib Sensor Netw
(2015)
T.P. Shabeera et al.
Optimizing VM allocation and data placement for data-intensive applications in cloud using ACO metaheuristic algorithm
Eng Sci Technol Int J
(2017)
J.W. Bos et al.
Private predictive analysis on encrypted medical data
J Biomed Inform
(2014)
A. Singh et al.
Cluster based bee algorithm for virtual machine placement in cloud data centre
JATIT
(2013)
R. Camati et al.
Solving the virtual machine placement problem as a multiple multidimensional knapsack problem
IARIA, IEEE.
(2014)
B. Suseela et al.
A multi-objective hybrid aco-pso optimization algorithm for virtual machine placement in cloud computing
IJRET
(2014)
J. Zhao et al.
A heuristic placement selection of live virtual machine migration for energy-saving in cloud computing environment
PloS ONE
(2014)
A. Abdelaziz, M.Elhoseny, A.S. Salama, A.M. Riad and A. Hassanien, Intelligent Algorithms for Optimal Selection of...
S. Sharma
Cervical cancer stage prediction using decision tree approach of machine learning
IJARCCE
(2016)
C.B. Kumar et al.
Data analysis and prediction of hepatitis using Support Vector Machine (SVM)
IJCSIT
(2014)

T. Prerana et al.

Prediction of heart disease using machine learning algorithms – Naïve Bayes, Introduction to PAC Algorithm, comparison of algorithms and HDPS

IJSE

(2015)

P. Tintu et al.

detect breast cancer using fuzzy c means techniques in Wisconsin Prognostic Breast Cancer (WPBC) Data Sets

IJCAT

(2013)

A.M. Hamad, Lung cancer diagnosis by using fuzzy logic, ijcsmc 5(3) (2016)...

C. Arjun et al.

Diagnosis of Diabetes Using Support Vector Machine and Ensemble Learning Approach

IJEAS

(2015)

K. Parikh et al.

Virtual machine allocation policy in cloud computing using CloudSim in Java

IJGDC

(2015)

N.R. Darwish et al.

Applying swarm optimization techniques to calculate execution time for software modules

IJARAI

(2016)

N.R. Darwish et al.

A hybrid machine learning model for selecting suitable requirements elicitation techniques

IJCSIS

(2016)

X. Fang et al.

An Energy efficient ant colony system for virtual machine placement in cloud computing

Comput Intell Soc, IEEE

(2016)

N. Almezeini et al.

Task scheduling in cloud computing using lion optimization algorithm

IJACSA

(2017)

Cited by (184)

A new cloud-based method for composition of healthcare services using deep reinforcement learning and Kalman filtering
2024, Computers in Biology and Medicine
Healthcare has significantly contributed to the well-being of individuals around the globe; nevertheless, further benefits could be derived from a more streamlined healthcare system without incurring additional costs. Recently, the main attributes of cloud computing, such as on-demand service, high scalability, and virtualization, have brought many benefits across many areas, especially in medical services. It is considered an important element in healthcare services, enhancing the performance and efficacy of the services. The current state of the healthcare industry requires the supply of healthcare products and services, increasing its viability for everyone involved. Developing new approaches for discovering and selecting healthcare services in the cloud has become more critical due to the rising popularity of these kinds of services. As a result of the diverse array of healthcare services, service composition enables the execution of intricate operations by integrating multiple services' functionalities into a single procedure. However, many methods in this field encounter several issues, such as high energy consumption, cost, and response time. This article introduces a novel layered method for selecting and evaluating healthcare services to find optimal service selection and composition solutions based on Deep Reinforcement Learning (Deep RL), Kalman filtering, and repeated training, addressing the aforementioned issues. The results revealed that the proposed method has achieved acceptable results in terms of availability, reliability, energy consumption, and response time when compared to other methods.
Early prediction of pathological complete response to neoadjuvant chemotherapy in breast cancer MRI images using combined Pre-trained convolutional neural network and machine learning
2023, Measurement: Journal of the International Measurement Confederation
Timely diagnosis of breast cancer can ameliorate the treatment plan, thus reducing the mortality rate. We propose a model integrating pre-trained Convolutional neural network (CNN) with machine learning for prognosticating pathologic complete response(PCR) using breast cancer dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) prior to commencement of neoadjuvant chemotherapy(NACT).
For predicting pathologic complete response (PCR) using dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) for breast cancer prior to the start of neoadjuvant chemotherapy, we present a hybrid model integrating a pre-trained Convolutional neural network (CNN) with machine learning (NACT).
In this retrospective study, 64 patients receiving NACT for invasive breast cancer are examined. Deep learning-based pre-trained CNN models ResNet-50 and ResNet-18 were used to extract features from patient visit 1 MRI images (before the initiation of NACT). Mann-Whitney U tests is used to assess features and their relevance (significance level p less than 0.05 and confidence interval is 95 %). Furthermore, features extracted and features selected were independently given as an input to different machine learning classifiers for the prediction of response of NACT. Classification performance was assessed under different data division protocols using accuracy, specificity, sensitivity, and area under the receiver operating characteristic curve (AUROC).
The hybrid combination using ResNet-18 as feature extractor, fine K-nearest neighbor(KNN) as classifier and feature selected using Mann-Whitney U test outperformed the result. Accuracy of 99.8% and AUROC of 1 is obtained under hold-out validation protocol while accuracy of 99.3%, and AUROC of 0.99 is obtained under 10-fold cross-validation.
The proposed model employing DCE-MRI images acquired before starting chemotherapy has considerable accuracy in classifying PCR and non-PCR patients. The efficacy of the prediction model can improve considerably on the back of a larger dataset.
Review of fog and edge computing–based smart health care system using deep learning approaches
2023, Deep Learning in Personalized Healthcare and Decision Support
Fog/edge computing enables Internet of things (IoT) applications by employing computational node resources to evaluate the gathered data, meet latency requirements, and boost the scalability and energy efficiency of IoT systems. But there are a few challenging problems that need to be resolved before fog/edge computing for IoT can be fully utilized. Fog/edge computing, for instance, has a tendency to be unstable because the bulk of its compute nodes run on batteries. Further research is required on how to effectively provide computing resources to latency-sensitive IoT operations (like amplified actuality and essential realism, haptic technology, intelligent manufacturing, connected autonomous driving, and others), as well as how to seamlessly integrate fog/edge computing with cloud computing to deliver scalable services. The current research piece gives a thorough analysis of the device knowledge and subterranean erudition techniques utilized to address various tribulations in submissions connected to fog and edge computing.
Diabetes prediction using Shapley additive explanations and DSaaS over machine learning classifiers: a novel healthcare paradigm
2024, Multimedia Tools and Applications
Performance enhancement of cloud security with migration algorithm for choosing virtual machines in cloud computing
2024, Engineering Research Express
Machine learning methods for service placement: a systematic review
2024, Artificial Intelligence Review

View all citing articles on Scopus

View full text

A machine learning model for improving healthcare services on cloud computing environment

Abstract

Introduction

Section snippets

Parallel particle swarm optimization

Related work

The proposed could computing based optimization model for VMs

CKD diagnosis using LR analysis

Experimental results

Conclusion and future work

Int J Distrib Sensor Netw

Eng Sci Technol Int J

J Biomed Inform

Cluster based bee algorithm for virtual machine placement in cloud data centre

JATIT

Solving the virtual machine placement problem as a multiple multidimensional knapsack problem

IARIA, IEEE.

A multi-objective hybrid aco-pso optimization algorithm for virtual machine placement in cloud computing

IJRET

A heuristic placement selection of live virtual machine migration for energy-saving in cloud computing environment

PloS ONE

Cervical cancer stage prediction using decision tree approach of machine learning

IJARCCE

Data analysis and prediction of hepatitis using Support Vector Machine (SVM)

IJCSIT

Prediction of heart disease using machine learning algorithms – Naïve Bayes, Introduction to PAC Algorithm, comparison of algorithms and HDPS

IJSE

detect breast cancer using fuzzy c means techniques in Wisconsin Prognostic Breast Cancer (WPBC) Data Sets

IJCAT

Diagnosis of Diabetes Using Support Vector Machine and Ensemble Learning Approach

IJEAS

Virtual machine allocation policy in cloud computing using CloudSim in Java

IJGDC

Applying swarm optimization techniques to calculate execution time for software modules

IJARAI

A hybrid machine learning model for selecting suitable requirements elicitation techniques

IJCSIS

An Energy efficient ant colony system for virtual machine placement in cloud computing

Comput Intell Soc, IEEE

Task scheduling in cloud computing using lion optimization algorithm

IJACSA