Convolutional neural network for classifying the stages of the cell cycle

The cell cycle is a highly coordinated process that ensures the duplication and transmission of genetic information from one cell generation to the next. The detailed observation constitutes a fundamental starting point for diagnosing and preventing diseases such as cancer. Computational techniques, such as deep learning, offer a way to understand the cell cycle behavior. This study introduces a convolutional neural network model specifically designed to classify the stages of the cell cycle of a cancerous cell. Images from the Jurkat cell line were used to train the model. A rigorous evaluation was conducted to verify the effectiveness of the proposed model, comparing the results with other deep learning models that also use the same database. The results showed the high performance of the proposed model. The model demonstrated better alignment with the data characteristics, resulting in a more accurate classification of the cell cycle stages. This outcome underscores the distinctive ability of a convolutional neural network to identify patterns in the cell cycle more precisely than human perception, which, at times, may be susceptible to errors. Utilizing a convolutional neural network reduced complexity and heightened suitability, primarily owing to the available volume of data, which aligns seamlessly with the requisites of a moderately deep network. Furthermore, deploying a convolutional neural network is associated with reduced computational overhead and faster data processing.