Chronic liver disease detection using ranking and projection-based feature optimization with deep learning
The liver is a vital gland responsible for various essential functions such as digestion, metabolism, detoxification, and immunity. Liver diseases caused by infections, injuries, or genetic factors are dangerous and require prompt diagnosis and treatment to improve survival rates. Early detection of...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
AIMS Press
2025-02-01
|
| Series: | AIMS Bioengineering |
| Subjects: | |
| Online Access: | https://www.aimspress.com/article/doi/10.3934/bioeng.2025003 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The liver is a vital gland responsible for various essential functions such as digestion, metabolism, detoxification, and immunity. Liver diseases caused by infections, injuries, or genetic factors are dangerous and require prompt diagnosis and treatment to improve survival rates. Early detection of liver conditions is crucial, and recent advancements in machine learning (ML) have proven highly effective in predicting diseases like chronic obstructive pulmonary disease (COPD), hypertension, and diabetes. Additionally, the rise of deep learning has begun transforming liver research, offering powerful tools to aid doctors in diagnosis and treatment. This study presents a novel and efficient learning method to identify liver patients accurately. The approach integrates multiple ranking and projection techniques for features, utilizing deep learning to detect early signs of liver disease. Additionally, Shapley Additive exPlanations (SHAP) are applied to perform global interpretation analysis, helping to select optimal features by assessing their contributions to the overall model. Our experimental results demonstrate that this proposed model outperforms traditional machine learning algorithms, achieving superior accuracy. Cross-validation and various testing methods confirm that the deep neural network (DNN) we developed surpasses other classifiers, reaching an accuracy rate of 90.12%. This paper explores how machine learning can be integrated into healthcare, particularly for predicting liver disease. Our findings show that the proposed model can potentially improve diagnostic accuracy and support timely medical intervention, ultimately enhancing patient outcomes. |
|---|---|
| ISSN: | 2375-1495 |