A Predictive Study of Machine Learning and Deep Learning Procedures Over Chronic Disease Datasets
-
Published:2024-02-01
Issue:42
Volume:
Page:34-47
-
ISSN:2799-1172
-
Container-title:Feb-Mar 2024
-
language:en
-
Short-container-title:JAIMLNN
Abstract
People's health and well-being are not given priority in the technological and Internet-savvy world we live in. People are becoming worse because they don't regularly attend the hospital for checkups due to job and unanticipated events. Most people nowadays suffer from one or more chronic illnesses, such as diabetes, hypothyroidism, heart disease, breast cancer, and dermatology. According to the World Health Organization (WHO), these chronic illnesses account for half of all fatalities in most nations and are the main cause of premature mortality. Patients who are identified early on potentially have their condition stop progressing. Many dispersed studies clearly demonstrated that conventional approaches to diagnosing chronic illnesses are prone to prejudice and heterogeneity among physicians, making it difficult to promptly and precisely diagnose problems. Still, Despite the availability of up-to-date information and a variety of machine learning-based methods, there have been enormous published efforts demonstrating that machine learning (ML)/deep learning (DL) based approach can considerably enhance the timely estimation of various health conditions. However, precise diagnosis of such diseases remains a difficulty. There are many machine learning-based techniques and current knowledge available, however despite this, a great deal of published research has shown that machine learning/deep learning based approach can considerably enhance the timely estimation of various health conditions. However, precise diagnosis of such diseases remains a difficulty. In order to tackle this problem, this work uses the UCI/KAGGLE ML/DL disease dataset to evaluate various ML/DL procedures and explores how different machine learning algorithms forecast chronic diseases. Accuracy and confusion matrix are used to verify the results. In order to help inexperienced researchers comprehend the disease prediction function of ML/DL-based techniques and determine the direction of Upcoming research, this study also discusses the advantages and disadvantages of accessible disease prediction schemes.
Reference46 articles.
1. TEstes, C.; Anstee, Q.M.; Arias-Loste, M.T.; Bantel, H.; Bellentani, S.; Caballeria, J.; Colombo, M.; Craxi, A.; Crespo, J.; Day, C.P.; et al. Modeling NAFLD disease burden in China, France, Germany, Italy, Japan, Spain, United Kingdom, and United States for the period 2016–2030. J. Hepatol. 2018, 69, 896–904. 2. Droz˙dz˙ , K.; Nabrdalik, K.; Kwiendacz, H.; Hendel, M.; Olejarz, A.; Tomasik, A.; Bartman,W.; Nalepa, J.; Gumprecht, J.; Lip, G.Y.H. Risk factors for cardiovascular disease in patients with metabolic-associated fatty liver disease: A machine learning approach. Cardiovasc. Diabetol. 2022, 21, 240. 3. S. Ghwanmeh, A. Mohammad, and A. Al-Ibrahim, “Innovative artificial neural networks-based decision support system for heart diseases diagnosis,” Journal of Intelligent Learning Systems and Applications, vol. 5, no. 3, 2013, pp. 176–183. 4. D. Al-Jumeily, S. Iram, F. B. Vialatte, P. Fergus, and A. Hussain, “A novel method of early diagnosis of Alzheimer’s disease based on EEG signals,” Scientific World Journal, vol. 2015, article 931387, 11 pages, 2015. 5. E. Keogh and A. Mueen, “Curse of dimensionality,” in Encyclopedia of Machine Learning and Data Mining, C. Sammut and G. I. Webb, Eds., Springer, Cham, Switzerland, 2017.
|
|