Publication

Abstract : Blood cancer, arising from DNA mutations, presents a significant challenge in terms of early detection. To extract crucial physiological and morphological information from cancer cells, a combination method including machine learning algorithms and human image processing is required due to the disease’s complexity. In this research, we propose a system that leverages sophisticated Convolutional Neural Network (CNN) algorithms, along with transfer learning and fine-tuning techniques, to achieve accurate prediction of cancer subtypes, enabling early detection and facilitating timely treatment interventions. It also incorporates a user-friendly web-based chatbot interface,which is integrated to a sophisticated drug discovery module, enabling the identification of potential targets for novel drugs and potential drug combinations that are specifically effective against corresponding subtypes of blood cancer considering the patient’s unique circumstances along with the accuracy of the prediction by selecting the model. This paper aims to improve the precision of blood cancer detection and therapy by merging cutting-edge machine learning algorithms, user-friendly user interfaces, and cutting-edge drug discovery techniques. The potential impact of this research is immense, as improved detection and treatment can ultimately lead to saving lives and improving the prognosis of individuals affected by blood cancer.