1
|
A survey on multi-objective hyperparameter optimization algorithms for machine learning. Artif Intell Rev 2022. [DOI: 10.1007/s10462-022-10359-2] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/26/2022]
Abstract
AbstractHyperparameter optimization (HPO) is a necessary step to ensure the best possible performance of Machine Learning (ML) algorithms. Several methods have been developed to perform HPO; most of these are focused on optimizing one performance measure (usually an error-based measure), and the literature on such single-objective HPO problems is vast. Recently, though, algorithms have appeared that focus on optimizing multiple conflicting objectives simultaneously. This article presents a systematic survey of the literature published between 2014 and 2020 on multi-objective HPO algorithms, distinguishing between metaheuristic-based algorithms, metamodel-based algorithms and approaches using a mixture of both. We also discuss the quality metrics used to compare multi-objective HPO procedures and present future research directions.
Collapse
|
2
|
Improved deep convolutional neural network-based COOT optimization for multimodal disease risk prediction. Neural Comput Appl 2022. [DOI: 10.1007/s00521-022-07767-4] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/10/2022]
|
3
|
Abdulghafor R, Turaev S, Ali MAH. Body Language Analysis in Healthcare: An Overview. Healthcare (Basel) 2022; 10:healthcare10071251. [PMID: 35885777 PMCID: PMC9325107 DOI: 10.3390/healthcare10071251] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/25/2022] [Revised: 06/22/2022] [Accepted: 06/23/2022] [Indexed: 11/16/2022] Open
Abstract
Given the current COVID-19 pandemic, medical research today focuses on epidemic diseases. Innovative technology is incorporated in most medical applications, emphasizing the automatic recognition of physical and emotional states. Most research is concerned with the automatic identification of symptoms displayed by patients through analyzing their body language. The development of technologies for recognizing and interpreting arm and leg gestures, facial features, and body postures is still in its early stage. More extensive research is needed using artificial intelligence (AI) techniques in disease detection. This paper presents a comprehensive survey of the research performed on body language processing. Upon defining and explaining the different types of body language, we justify the use of automatic recognition and its application in healthcare. We briefly describe the automatic recognition framework using AI to recognize various body language elements and discuss automatic gesture recognition approaches that help better identify the external symptoms of epidemic and pandemic diseases. From this study, we found that since there are studies that have proven that the body has a language called body language, it has proven that language can be analyzed and understood by machine learning (ML). Since diseases also show clear and different symptoms in the body, the body language here will be affected and have special features related to a particular disease. From this examination, we discovered that it is possible to specialize the features and language changes of each disease in the body. Hence, ML can understand and detect diseases such as pandemic and epidemic diseases and others.
Collapse
Affiliation(s)
- Rawad Abdulghafor
- Department of Computer Science, Faculty of Information and Communication Technology, International Islamic University Malaysia, Kuala Lumpur 53100, Malaysia
- Correspondence: (R.A.); (S.T.); (M.A.H.A.)
| | - Sherzod Turaev
- Department of Computer Science and Software Engineering, College of Information Technology, United Arab Emirates University, Al-Ain, Abu Dhabi P.O. Box 15556, United Arab Emirates
- Correspondence: (R.A.); (S.T.); (M.A.H.A.)
| | - Mohammed A. H. Ali
- Department of Mechanical Engineering, Faculty of Engineering, University of Malaya, Kuala Lumpur 50603, Malaysia
- Correspondence: (R.A.); (S.T.); (M.A.H.A.)
| |
Collapse
|
4
|
Rashid J, Batool S, Kim J, Wasif Nisar M, Hussain A, Juneja S, Kushwaha R. An Augmented Artificial Intelligence Approach for Chronic Diseases Prediction. Front Public Health 2022; 10:860396. [PMID: 35433587 PMCID: PMC9008324 DOI: 10.3389/fpubh.2022.860396] [Citation(s) in RCA: 12] [Impact Index Per Article: 6.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/22/2022] [Accepted: 02/22/2022] [Indexed: 12/23/2022] Open
Abstract
Chronic diseases are increasing in prevalence and mortality worldwide. Early diagnosis has therefore become an important research area to enhance patient survival rates. Several research studies have reported classification approaches for specific disease prediction. In this paper, we propose a novel augmented artificial intelligence approach using an artificial neural network (ANN) with particle swarm optimization (PSO) to predict five prevalent chronic diseases including breast cancer, diabetes, heart attack, hepatitis, and kidney disease. Seven classification algorithms are compared to evaluate the proposed model's prediction performance. The ANN prediction model constructed with a PSO based feature extraction approach outperforms other state-of-the-art classification approaches when evaluated with accuracy. Our proposed approach gave the highest accuracy of 99.67%, with the PSO. However, the classification model's performance is found to depend on the attributes of data used for classification. Our results are compared with various chronic disease datasets and shown to outperform other benchmark approaches. In addition, our optimized ANN processing is shown to require less time compared to random forest (RF), deep learning and support vector machine (SVM) based methods. Our study could play a role for early diagnosis of chronic diseases in hospitals, including through development of online diagnosis systems.
Collapse
Affiliation(s)
- Junaid Rashid
- Department of Computer Science and Engineering, Kongju National University, Cheonan, South Korea
| | - Saba Batool
- Department of Computer Science, COMSATS University Islamabad, Islamabad, Pakistan
| | - Jungeun Kim
- Department of Computer Science and Engineering, Kongju National University, Cheonan, South Korea
- *Correspondence: Jungeun Kim
| | - Muhammad Wasif Nisar
- Department of Computer Science, COMSATS University Islamabad, Islamabad, Pakistan
| | - Amir Hussain
- Data Science and Cyber Analytics Research Group, Edinburgh Napier University, Edinburgh, United Kingdom
| | - Sapna Juneja
- Department of Computer Science, KIET Group of Institutions, Ghaziabad, India
| | - Riti Kushwaha
- Department of Computer Science, Bennett University, Greater Noida, India
| |
Collapse
|
5
|
Wu C, Zhou T, Tian Y, Wu J, Li J, Liu Z. A method for the early prediction of chronic diseases based on short sequential medical data. Artif Intell Med 2022; 127:102262. [DOI: 10.1016/j.artmed.2022.102262] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/09/2020] [Revised: 02/18/2022] [Accepted: 02/23/2022] [Indexed: 11/30/2022]
|
6
|
Identification and Prediction of Chronic Diseases Using Machine Learning Approach. JOURNAL OF HEALTHCARE ENGINEERING 2022; 2022:2826127. [PMID: 35251563 PMCID: PMC8896926 DOI: 10.1155/2022/2826127] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 01/13/2022] [Revised: 02/01/2022] [Accepted: 02/07/2022] [Indexed: 01/01/2023]
Abstract
Nowadays, humans face various diseases due to the current environmental condition and their living habits. The identification and prediction of such diseases at their earlier stages are much important, so as to prevent the extremity of it. It is difficult for doctors to manually identify the diseases accurately most of the time. The goal of this paper is to identify and predict the patients with more common chronic illnesses. This could be achieved by using a cutting-edge machine learning technique to ensure that this categorization reliably identifies persons with chronic diseases. The prediction of diseases is also a challenging task. Hence, data mining plays a critical role in disease prediction. The proposed system offers a broad disease prognosis based on patient's symptoms by using the machine learning algorithms such as convolutional neural network (CNN) for automatic feature extraction and disease prediction and K-nearest neighbor (KNN) for distance calculation to find the exact match in the data set and the final disease prediction outcome. A collection of disease symptoms has been performed for the preparation of the data set along with the person's living habits, and details related to doctor consultations are taken into account in this general disease prediction. Finally, a comparative study of the proposed system with various algorithms such as Naïve Bayes, decision tree, and logistic regression has been demonstrated in this paper.
Collapse
|
7
|
N.V. RK, M. A, E. B, J. SJP, A. K, S. P. Detection and monitoring of the asymptotic COVID-19 patients using IoT devices and sensors. INTERNATIONAL JOURNAL OF PERVASIVE COMPUTING AND COMMUNICATIONS 2020. [DOI: 10.1108/ijpcc-08-2020-0107] [Citation(s) in RCA: 11] [Impact Index Per Article: 2.8] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 11/17/2022]
Abstract
Purpose
Many investigations are going on in monitoring, contact tracing, predicting and diagnosing the COVID-19 disease and many virologists are urgently seeking to create a vaccine as early as possible. Even though there is no specific treatment for the pandemic disease, the world is now struggling to control the spread by implementing the lockdown worldwide and giving awareness to the people to wear masks and use sanitizers. The new technologies, including the Internet of things (IoT), are gaining global attention towards the increasing technical support in health-care systems, particularly in predicting, detecting, preventing and monitoring of most of the infectious diseases. Similarly, it also helps in fighting against COVID-19 by monitoring, contract tracing and detecting the COVID-19 pandemic by connection with the IoT-based smart solutions. IoT is the interconnected Web of smart devices, sensors, actuators and data, which are collected in the raw form and transmitted through the internet. The purpose of this paper is to propose the concept to detect and monitor the asymptotic patients using IoT-based sensors.
Design/methodology/approach
In recent days, the surge of the COVID-19 contagion has infected all over the world and it has ruined our day-to-day life. The extraordinary eruption of this pandemic virus placed the World Health Organization (WHO) in a hazardous position. The impact of this contagious virus and scarcity among the people has forced the world to get into complete lockdown, as the number of laboratory-confirmed cases is increasing in millions all over the world as per the records of the government.
Findings
COVID-19 patients are either symptomatic or asymptotic. Symptomatic patients have symptoms such as fever, cough and difficulty in breathing. But patients are also asymptotic, which is very difficult to detect and monitor by isolating them.
Originality/value
Asymptotic patients are very hazardous because without knowing that they are infected, they might spread the infection to others, also asymptotic patients might be having very serious lung damage. So, earlier prediction and monitoring of asymptotic patients are mandatory to save their life and prevent them from spreading.
Collapse
|
8
|
Detecting coronavirus contact using internet of things. INTERNATIONAL JOURNAL OF PERVASIVE COMPUTING AND COMMUNICATIONS 2020. [DOI: 10.1108/ijpcc-07-2020-0074] [Citation(s) in RCA: 8] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 11/17/2022]
Abstract
Purpose
The purpose of this paper is to identify coronavirus contact using internet of things. The disease is said to be highly contagious with the contact of infected persons. Feared to be air-borne, droplets of body fluids can transmit the disease in a matter of hours. The predominant symptoms of the COVID-19 are high fever, cough, breathing problem, etc. Recent studies have demonstrated the evolution of the disease to hide its symptoms. As it is highly transmissible, this disease might spread at an exponential rate costing the lives of thousands of people. The chain of transmission has to be detected with utmost priority through early detection and isolation of infected people. Automated internet of things (IoT) devices can be used in design and implementation of a prediction scheme for reporting the health-care risks of the patients with various parameters such as temperature, humidity and blood pressure.
Design/methodology/approach
IoT is a configuration of multiple autonomous and embedded wireless devices for serving a purpose. Every object possesses an individual identity and will serve to register critical events as entries for future learning and decisions. IoT plays an inevitable role in medical industries, detection of vital signs of diseases and monitoring. Among other life-threatening diseases, a new pandemic is on rise among world nations. COVID-19, a novel severe acute respiratory syndrome virus originated from animals in December 2019 and is becoming a serious menace to Governments, despite serious measures of lockdowns.
Findings
In this paper, the authors defined an architecture of an IoT system to predict the Covid-19 disease by getting the data from the human through sensors and send the data to the doctor using mobile, computer, etc. The main goal is early health surveillance by predicting COVID-19. Accordingly, the authors are able to identify both symptomatic and asymptomatic patients, which will help in the early prediction of disease.
Originality/value
Using the proposed method, the authors can save the time of both patient and doctor by ensuring timely medical treatment and contribute toward breaking the transmission chain. In so doing, the method also contributes toward avoiding unnecessary expenses and saving human lives.
Collapse
|