A Deep Learning Based Multiple Chronic Disease Detection Model

• Author(s): Fati Oiza Ochepa ; John Patrick ; Malik Adeiza Rufai ; Adamu Isah
• Paper ID: 1703895
• Page: 107-116
• Published Date: 28-11-2022
• Published In: Iconic Research And Engineering Journals
• Publisher: IRE Journals
• e-ISSN: 2456-8880
• Volume/Issue: Volume 6 Issue 5 November-2022

Abstract

The rapid advancement of technology in the health sector has paved the way for automating health related processes. Diagnosing of diseases is one of the most important and sensitive tasks performed by health practitioners that if not done efficiently, can lead to dire consequences for the patients. This study developed and implemented a model for diagnosing four life threatening diseases; pneumonia, malaria, breast cancer and skin cancer using Deep learning. The datasets used for this study was acquired from Kaggle and features were selected using the hybrid technique. A Convolutional Neural Network (CNN) model was deployed by using 80% of the data for training while the remaining 20% served for validation of the model. Based on the model, a web based application diagnostic tool was deployed to enable patients gain easy access to efficient diagnosis. The model was assessed by using performance metrics such as precision, recall and F-measure. The overall accuracy of the model when tested on the four diseases dataset was 86.33%, 96.0%, 95.38% and 88.45% for pneumonia, malaria, breast cancer and skin cancer detection respectively.

Keywords

Chronic Disease, Convolutional Neural Network, Deep Learning, Detection, Diagnosis.

Citations

IRE Journals:
Fati Oiza Ochepa , John Patrick , Malik Adeiza Rufai , Adamu Isah "A Deep Learning Based Multiple Chronic Disease Detection Model" Iconic Research And Engineering Journals Volume 6 Issue 5 2022 Page 107-116

IEEE:
Fati Oiza Ochepa , John Patrick , Malik Adeiza Rufai , Adamu Isah "A Deep Learning Based Multiple Chronic Disease Detection Model" Iconic Research And Engineering Journals, 6(5)