Early Identification of Blood Cancer Through Automated Anomaly Detection With a Convolutional Neural Network

Diagnosing and assessing blood cancer requires significant time and effort due to its complexity. Improving the accuracy of information acquired through manual analysis techniques largely depends on automation tools and models. Research is advancing with early detection techniques for examining the features of human blood cells. At this point, researchers have developed deeper convolutional neural network (CNN)-based learning models, primarily utilizing hybrid ensemble DCNN approaches. With accuracy rates exceeding 99 %, this deep learning technology significantly enhances the ability to monitor the progression of blood cancer with precision. Therefore, using progressive technology to offer remedies in clinical diagnosis has become easier. The information obtained from this research, when compared with previous studies using LDSVM, demonstrates the potential to provide a better solution.