Purpose: This study introduces a pioneering hybrid model that combines genetic algorithms, neuro-fuzzy logic, and mobile agent technology to enhance predictive precision for early-stage prostate cancer diagnosis. Design/Methodology/Approach: One hundred and twenty records of prostate cancer patients were initially collected from the Delta State University Teaching Hospital, Oghara, Nigeria. Each patient’s record included relevant data on prostate disease, such as age, PSA levels, clinical history, symptom severity, biopsy results, and other demographic and clinical factors. This data was extracted and stored as rules in a MySQL database, with the MySQL Fuzzy Extension enabling fuzzy data storage and processing. Findings: Extensive simulations and clinical data analyses demonstrate the model’s superior sensitivity and specificity in detecting early-stage prostate cancer compared to traditional diagnostic methods. Medical expert evaluations validate the model’s effectiveness as a promising diagnostic alternative. Research Limitation: While results are promising, the study is limited to simulations and a controlled clinical dataset. Practical Implications: The system offers a practical, scalable early prostate cancer detection solution that could revolutionise current diagnostic practices. Social Implications: Potential social benefits include improved patient outcomes, reduced healthcare costs, and better quality of life. Originality/Value: This study presents an innovative integration of genetic algorithms, neuro-fuzzy systems, and mobile agent technology. This novel approach paves the way for advanced cancer diagnostics and precision medicine.
Read full abstract