Optimized neuro fuzzy convolutional networks in gene expression analysis in microscopic image feature extraction and classification by deep learning architectures

With the use of gene expression microarrays, also referred to as "gene chips," scientists can monitor the pace at which hundreds of genes in each cell or tissue are expressed and converted into proteins. These gene expression pictures of biological activity can help with disease diagnosis, prognosis, and treatment planning. They can also reveal new targets for drug discovery and infer regulatory circuits in cells. The popularity of deep learning methodologies has grown recently because of their wide range of applications in several sectors for inference and prediction. This research proposes novel techniques in gene expression analysis in microscopic image data feature extraction with classification. Input is collected as microscopic image which is processed for noise removal and image smoothening. Then, these image features were extracted using neuro fuzzy convolutional networks. The extracted features have been classified using a self-learning-based genetic algorithm to obtain input images' gene expression. Using these organized gene expression immunity system and abnormality of the body function have been analyzed. It is proposed that existing approaches be improved. I developed a multivariate and hybrid feature selection strategy to achieve good classification performance for high-dimension classification issues. Experimental analysis is carried out for various microscopic image datasets in terms of accuracy, recall, precision, F-Score, RMSE, and MAP.