Bio-inspired Algorithms for Solving Nonlinear Optimization Problems in Medical Diagnostics
Year: 2025
Type: article
Abstract: The growing burden of diagnostic errors and inefficiencies in Ghana’s healthcare system necessitates innovative solutions, and bio-inspired algorithms present a transformative approach to solving nonlinear optimization problems in medical diagnostics. As healthcare demands more accurate, faster, and cost-efficient diagnostic tools, the application of Genetic Algorithms, Particle Swarm Optimization, and Ant Colony Optimization offers a critical pathway for improvement. This study aimed to evaluate how these bio-inspired algorithms could enhance diagnostic accuracy, computational efficiency, convergence rates, and robustness within Ghanaian healthcare settings. Using a quantitative research design based on 105 secondary datasets from 2020 to 2024, the study applied descriptive statistics, correlation analysis, and regression models. Results showed that Genetic Algorithms improved diagnostic accuracy by 20%, Particle Swarm Optimization reduced diagnostic times by 50%, and Ant Colony Optimization enhanced solution robustness with only a -3.7% drop under noise, achieving an overall correlation coefficient of r = 0.734 between algorithm application and diagnostic efficiency. The findings concluded that bio-inspired algorithms significantly optimized medical diagnostics despite hardware and data limitations, and the regression model revealed that 65.1% of the variance in diagnostic efficiency was explained by these techniques. These results imply that strategic deployment of bio-inspired algorithms can revolutionize healthcare delivery in resource-constrained environments. Consequently, it is recommended that Ghana's healthcare institutions invest in capacity-building, government policies incentivize technological integration, and future research explore broader scalability across Sub-Saharan Africa.
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W6911977182
FWCI:
Cites:
Cited by:
Related to: