Music photonic signal analysis based health monitoring system using classification by quantum machine learning techniques

• Published in: Optical and quantum electronics 2024-03, Vol.56 (3)
• Main Author: Hou Dingjie
• Format: Article
• Language: English
• Subjects: Algorithms; Classification; Lidar; Lung cancer; Lung diseases; Machine learning; Nodules; Noise reduction; Photonics; Recall; Signal analysis; Signal monitoring
• ISSN: 0306-8919; 1572-817X
• DOI: 10.1007/s11082-023-05960-w

The next generation of lidar systems will need to be more flexible and have a higher bandwidth in order to achieve higher resolution. Due to the inherent digital nature of the technology, the detector bandwidth and sampling frequency of traditional lidar systems, and picture lidar systems in particular, are limited. One of the most frequent diseases that, if caught and treated early on, greatly increases a patient’s likelihood of survival is lung disease. The most challenging part of a radiologist’s job is making the diagnosis of cancer. Radiologists would benefit greatly from using a state-of-the-art computerised system. In order to diagnose lung cancer, ML algorithms have been used in several research. To predict lung cancer, a multistage classification is most usually utilised. The data categorization system for segmentation and enhancement is complete. Using quantum machine learning in tandem with radiation based on musical photonic signals, the authors of this work propose a novel approach of identifying lung cancer. Here, trumpet players’ data has been collected and reviewed for noise reduction, normalisation, and smoothing. The features of the processed data are extracted and classified with the aid of support kernel vector Gaussian learning and spatial convolutional perceptron learning. Accuracy, precision, recall, AUC, true positive rate (TPR), and false positive rate (FPR) are experimentally studied across a variety of lung cancer datasets. Experimental results show that the proposed technique is effective in detecting and classifying lung cancer nodules. Accuracy, precision, recall, AUC, and test and false positive rates all improved by 5–45% when using the suggested method.