A BiLSTM model enhanced with multi-objective arithmetic optimization for COVID-19 diagnosis from CT images

In response to the relentless mutation of the coronavirus disease, current artificial intelligence algorithms for the automated diagnosis of COVID-19 via CT imaging exhibit suboptimal accuracy and efficiency. This manuscript proposes a multi-objective optimization algorithm (MOAOA) to enhance the Bi...

Full description

Saved in:
Bibliographic Details
Published in:Scientific reports Vol. 15; no. 1; pp. 10841 - 19
Main Authors: Chen, Liang, Lin, Xin, Ma, Liangliang, Wang, Chao
Format: Journal Article
Language:English
Published: London Nature Publishing Group UK 29.03.2025
Nature Publishing Group
Nature Portfolio
Subjects:
631/114/1305
631/114/1564
631/114/2164
Accuracy
Algorithms
Artificial Intelligence
Automatic diagnosis
Automation
BiLSTM
Classification
Comparative analysis
Computed tomography
Coronaviruses
COVID-19
COVID-19 - diagnosis
COVID-19 - diagnostic imaging
Datasets
Deep learning
Diagnosis
Disease
Efficiency
Human error
Humanities and Social Sciences
Humans
Long short-term memory
Medical imaging
Medical research
MOAOA
Multi-objective optimization
multidisciplinary
Optimization algorithms
Optimization techniques
Patients
R&D
Research & development
SARS-CoV-2
Science
Science (multidisciplinary)
Tomography
Tomography, X-Ray Computed - methods
COVID-19
MOAOA
Multi-objective optimization
Automatic diagnosis
BiLSTM
ISSN:2045-2322, 2045-2322
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In response to the relentless mutation of the coronavirus disease, current artificial intelligence algorithms for the automated diagnosis of COVID-19 via CT imaging exhibit suboptimal accuracy and efficiency. This manuscript proposes a multi-objective optimization algorithm (MOAOA) to enhance the BiLSTM model for COVID-19 automated diagnosis. The proposed approach involves configuring several hyperparameters for the bidirectional long short-term memory (BiLSTM), optimized using the MOAOA intelligent optimization algorithm, and subsequently validated on publicly accessible medical datasets. Remarkably, our model achieves an impressive 95.32% accuracy and 95.09% specificity. Comparative analysis with state-of-the-art techniques demonstrates that the proposed model significantly enhances accuracy, efficiency, and other performance metrics, yielding superior results.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:2045-2322
2045-2322
DOI:10.1038/s41598-025-94654-2