Automatic malaria disease detection from blood cell images using the variational quantum circuit

Abstract

Variational quantum circuit (VQC) is a quantum-classical (QC) machine learning approach that accommodates quantum processes on a classical computer. Malaria is a worldwide deadly disease caused by Plasmodium parasites. This research designs an effective VQC-based approach to recognize the existence of malaria from the Red blood cell (RBC) image through the classification of the optimized feature set that has been extracted from a set of RBC images. Minimum redundancy maximum relevance (mRMR), and Principal component analysis (PCA), are used to optimize the feature set. Comparing to existing classical approaches we have found that mRMR with our input encoding and parameterized circuit of VQC shows satisfactory performance by using a lower number of features and a lower number of parameters. After ascertaining the presence of malaria from VQC we have also introduced a rule-based expert system to detect the types of malaria. The proposed mechanism is mainly designed to evaluate the potency of quantum machine learning (QML) in near-term quantum computers and using the ten-fold cross-validation the scheme gained an overall accuracy, precision, recall, and specificity of 99.06%, 99.08%, 99.05%, and 99.07% respectively for malaria disease diagnosis.