Disease prediction model: An efficient machine learning- based DNA classifier

When it comes to health care, everyone is always eager to identify diseases in their early stages, but doing so might be difficult because of the lack of knowledge on the patterns of specific diseases since DNA contains most of the genetic blueprints, DNA sequence classification can be used to predict the existence of certain conditions accurately. There are several machine-learning techniques available to classify DNA sequences. Traits from known diseases are extracted to train the model for new, unknown diseases. The expansion of patients' access to digital platforms for early disease diagnosis through knowledge transfer to artificial neural networks eliminates the need for clinical equipment. To analyze the model, DNA samples of four well-known viruses—human respiratory viruses, lung cancer viruses, and papilla-maviruses (HPV)—are gathered from Genbank (NCBI). These samples are then compared with five existing methods using seven different parameters—specificity, accuracy, Matthews correlation coefficient, recall, precision, F1-score, area under the receiver operating characteristic (ROC) curve (AUROC), and area under the Precision-Recall (PRC) curve (AUPRC)—to facilitate the analysis of the model. The outcome demonstrates that the proposed work provides significantly better precision and accuracy than the prior best results, where precision has increased by more than 5.124% and accuracy has increased by about 15.9%.