Journal of Integrated Science and Technology
Keywords: item categories
Blood cancer is a global health issue that arises when abnormal cells grow uncontrollably in the blood, turning into cancer and having a high mortality rate around the world. It’s important to stop it before it’s too late, especially with things like leukemia. Over the past decade, there have been more cases than ever, making it more vital for better diagnostic methods. This research uses machine learning and the Internet of Things to diagnose blood cancer. SIFT (Scale-Invariant Feature Transform), DenseNet-169, and ResNet50 are methods they used to develop a classification system. Together, these steps help us get closer to accurate and effective diagnosis. Compared to other models, the combined model was extremely effective at diagnosing blood cancer. It had an accuracy of 99.75%, precision of 100%, recall of 99.48%, and an F1 score of 99.74%. The promising model uses advanced machine learning techniques to significantly improve diagnosis accuracy and efficiency when compared to models like logistic vector trees classifier (LVTrees), support vector classifier (SVC), and logistic regression (LR).
Discover more from RSYN FYI
Subscribe to get the latest posts sent to your email.