Deep Learning Models for Early Detection of Blood Cancer Disease

Cancer remains a significant global health challenge, emphasizing the crucial importance of early detection to improve patient outcomes. This research aims to improve the diagnosis of blood cancer through the application of deep learning techniques. Classification, a fundamental method in deep learning, finds widespread application across various domains. It involves assigning items in a dataset to specific categories or classes, enabling accurate prediction of the target class for each instance. The study employs advanced deep learning models, namely ResNetRS50 and RegNetX016, for cancer detection, marking a notable advancement in leveraging technology for early cancer diagnosis. Notably, the RegNetX016 model yielded superior results, achieving an accuracy of 97%. This is attributed to its more efficient architecture, which facilitates high prediction accuracies and faster, superior outcomes.