Performance Analysis of Fitness Function in Genetic Algorithm for Fetal Brain MRI Classification

Medical image classification plays a vital role in aiding diagnosis and treatment planning. Fetal magnetic resonance imaging has appeared as a crucial diagnostic tool for prenatal evaluation, providing detailed insights into fetal development and potential abnormalities. This paper shows a comparative evaluation of various fitness functions in genetic algorithm for fetal MRI image classification. The fitness function is a pivotal component in genetic algorithms, playing a crucial role in guiding the evolutionary process towards optimal solutions. The proposed study utilizes accuracy, precision, recall, f1- score, specificity and cross- entropy loss as a fitness function in genetic algorithms for fetal MR image classification. The classification results are evaluated using various performance metrics. The simulation outcomes show the effectiveness of the genetic algorithm framework in accurately classifying fetal MR images. Furthermore, the analysis reveals insights into the impact of different fitness functions on the classification performance, providing valuable guidance for optimizing the algorithm for fetal MR image classification tasks. This research contributes to the advancement of medical image analysis techniques and has implications for improving prenatal diagnosis and patient care.