Headache is a common disease that occurs in almost all kinds of people due to tension, changing lifestyle, and symptoms of other serious issues in the human body. The correct diagnosis of this lacks due to complex test procedures, irregular monitoring, visiting the physicians as an outpatient, and inexperience of physicians. To assist the specialists in detecting the accurate types of headaches, the proposed system is developed to classify the types of headaches using Modified Neural Network. The Introduction of the Brilliant Binary Chromosome Algorithm (BBCA) will generate brilliant chromosomes with the filter-based method that will give the most relevant features. The optimization of resource handling and reduced computational time is achieved by using the Fitness Based Crossover Algorithm (FBCA). The public Migbase dataset is taken for the study. Migbase contains collective information on headache attacks from 849 patient data from three different hospitals in Turkey. This Customized approach of the Genetic Algorithm is proposed to extract the elite features with a significant effect. The well designed modified Artificial Neural network model is implemented as the classifying model. The entire implementation of the proposed work is performed in Python 3.7 environment from which it is confirmed that the suggested novel combined methodology (CGA & MANN) will give optimal results compared with other current research methodologies concerning the proposed method. The proposed customized feature selection method improved the performance of the migraine classifier, producing a robust system that achieved over 98% accuracy. The results suggest that the proposed methods can be used to support specialists in the classification of migraines in patients undergoing treatment.