An RBF-PSO Based Approach for Modeling Prostate Cancer

TL;DR

This paper introduces a novel RBF-PSO based method to model prostate cancer by estimating disease parameters, aiding early diagnosis and understanding tumor growth dynamics.

Contribution

It proposes a new hybrid approach combining Particle Swarm Optimization with meshfree interpolation for parameter estimation in prostate cancer modeling.

Findings

Effective parameter estimation for prostate cancer models

Improved understanding of tumor growth dynamics

Potential for early diagnosis enhancement

Abstract

Prostate cancer is one of the most common cancers in men. It is characterized by a slow growth and it can be diagnosed in an early stage by observing the Prostate Specific Antigen (PSA). However, a relapse after the primary therapy could arise and different growth characteristics of the new tumor are observed. In order to get a better understanding of the phenomenon, a mathematical model involving several parameters is considered. To estimate the values of the parameters identifying the disease risk level a novel approach, based on combining Particle Swarm Optimization (PSO) with a meshfree interpolation method, is proposed.