Optical bio sensor based cancer cell detection using optimized machine learning model with quantum computing
G. Balamurugan,C. Annadurai,I. Nelson,K. Nirmala Devi,A. Sheryl Oliver,S. Gomathi
DOI: https://doi.org/10.1007/s11082-023-05684-x
IF: 3
2023-12-09
Optical and Quantum Electronics
Abstract:Recent developments in information technology are due to the development of smart cities, which act as a crucial facilitator for the building of next-generation intelligent systems to improve security, reliability, and efficiency. Intelligent biosensors and biochips for molecular robotics are being developed with the help of integrated optoelectronic devices, which are essential. These tools combine optics and electronics to provide very sensitive and accurate detection, analysis, and manipulation of biological molecules and cells. This research proposed novel method in detection of lung cancer utilizing feature selection as well as classification utilizingdeep learning architectures with optical bio sensor. Here input MRI lung image has been collected based on the pre-historic medical data of the patients. This image is processed and segmented for noise removal as well as normalization. Features of this processed image is selected using Lasso regression and classify the selected features using Bayesian neural networks.The cells of this imagesisclassified for identification of nanoparticle Fe3O4 in the images in detecting tumor region. Patients diagnosed with COCO dataset, TEM dataset, S2NANOdataset were used to test developed feature selection, classification and prediction methods on 4various data sets for HDdetection. Testing results revealed that developed classification, as well as prediction methods, attained an accuracy of 98%, precision of 95%, specificity of 91%, F-1 score of 90%, Area under the ROC Curve (AUC) of 88% and mAP(Mean Average Precision)of 81% for CCF dataset; for HIC dataset proposed method obtained accuracy of 95%, precision of 90%, specificity of 88%, F-1 score of 86%, AUC of 88% and mAP of 71%; proposed method obtained accuracy of 96%, precision of 94%, specificity of 81%, F-1 score of 89%, AUC of 85% and mAP of 68% for LBMC dataset; for SUH dataset proposed method obtained accuracy of 97%, precision of 92%, specificity of 88%, F-1 score of 91%, AUC of 77% and mAP of 71%exceeding accuracies of previously published research.
engineering, electrical & electronic,optics,quantum science & technology