International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 10 Issue: 07 | July 2023
p-ISSN: 2395-0072
www.irjet.net
Predicting disease from several symptoms using machine learning approach. MD. Atikur Rahman1, Tania Ahmed Nipa2, Md. Assaduzzaman3 1Student, Department of Computer Science and Engineering, Daffodil International University, DSC, Ashulia, Savar,
Dhaka, Bangladesh
2Student, Department of Computer Science and Engineering, Daffodil International University, DSC, Ashulia, Savar,
Dhaka, Bangladesh
3Lecturer, Department of Computer Science and Engineering, Daffodil International University, DSC, Ashulia,
Savar, Dhaka, Bangladesh ---------------------------------------------------------------------***--------------------------------------------------------------------Cause a major effect on someone’s health and sometimes might also come to death if they ignore[1]. People usually want to know if their symptoms are indicative of any serious diseases. However, they cannot access any tools that would provide them with precise information. Mainly machine Learning technology gives us a superior platform in the medical field so that healthcare issues will be solved effectively[10]. From the Centers for Medicare and Medicaid Services, 50% of Americans have multiple chronic diseases with a total US healthcare expenditure in 2016 being about $3.3 trillion, which amounts to $10,348 per person in the US[11]. This project intends to give them the tools they need to tell end users about disease prediction. If review mining can be used to create a prediction system for physicians and medicine, a lot of time will be saved. Understanding complex medical terms, such as scientific names, might be challenging while using this type of system's user interface. The user is confused by the wealth of medical knowledge on numerous symptom categories that are provided. This system's objective is to change to meet the particular user interaction requirements of the health area. A crucial part of treatment is using symptoms to anticipate sickness. In our experiment, we give an effort to accurately predict a disease by examining the patient's symptoms.
Abstract - Humans are the most intelligent species on the
earth and are very health conscious. The evolution of recent technologies like data science and machine learning has opened the trail for healthcare communities and medical establishments, to observe the diseases earliest as potential and it helps to supply higher patient care. For many medical organizations, disease prediction is very important for making the best possible healthcare decisions. Machine Learning is a field where we can develop a model to learn machines to make decisions on their own from real-time data and from past experience. We proposed a model to predict disease from some symptoms. So, in this experiment, we propose a new knowledge-based system for disease prediction using KNN, SVM, NB, DT, RF, and LR for data modeling and we got maximum 98.36 percent accuracy from the KNN algorithm. This paper is planned to develop multi-disease prediction using the machine learning concept. Our main contribution is to implement feature engineering and standard scal-ing to optimize our algorithm and better performance. The decision pro-posed support system for disease diagnosis might be implemented using the suggested methodology, aiding doctors in their work and enhancing patient outcomes. So, future studies will concentrate on increasing the dataset to cover a wider range of patient demographics and improving the machine learning algorithms to better prediction accuracy.
In the field of health informatics, machine learning becomes more popular to diagnosis, prognosis, personalized medicine and identify any disease by using some machine leaning technique. So, if we can predict any disease at it’s early stage, it will be more easy to give treatment to the patient by any medical servant. Overall, machine learning has the potential to change the medical industry by giving medical experts strong tools to more precisely and effectively predict, diagnose, and treat diseases.
Key Words: Machine Learning, Disease, Syndromes
1.INTRODUCTION Machine learning is the programming field of computers where computer systems learn from data and experience. Nowadays, demands for health-related information are transforming information-seeking behavior, as demonstrated globally. Finding accurate health information online on symptoms, diagnoses, and treatments can just be timeconsuming and expensive for many people. Today, billions of searches are performed daily, and sometimes the results are significant, and sometimes they are not. Such search terms generate thousands of results linked to medical advice. Diseases and health-related problems like Pneumonia, dengue, AIDS, Diabetes, Hepatitis, Jaundice, Arthritis, etc.
© 2023, IRJET
|
Impact Factor value: 8.226
2.LITERATURE REVIEW Grampurohit, S. and Sagarnal, C., [1] Via assisting doctors in early clinical diagnosis and prediction, a classification method that was constructed employing machine learning techniques was intended to greatly aid inside this settlement for mental well-being challenges. One representative collection
|
ISO 9001:2008 Certified Journal
|
Page 836