Smart explainable artificial intelligence for sustainable secure healthcare application based on quantum optical neural network

• Published in: Optical and quantum electronics Vol. 55; no. 10
• Main Authors: Suhasini, S., Tatini, Narendra Babu, Arslan, Farrukh, Bansal, Sushil Kumar, Babu, Suresh, Umaralievich, Mekhmonov Sultonali
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.10.2023; Springer Nature B.V
• Subjects: Artificial intelligence; Characterization and Evaluation of Materials; Communication networks; Computer Communication Networks; Decision making; Deep learning; Electrical Engineering; Energy consumption; Explainable artificial intelligence; Graph neural networks; Health care; Lasers; Machine learning; Multimedia; Neural networks; Optical Devices; Optics; Photonics; Physics; Physics and Astronomy; Quantum computing; Recall; Urbanization; Deep learning; Explaining artificial intelligence; Data analysis; Sustainable quantum optical application; Multimedia healthcare
• ISSN: 0306-8919; 1572-817X
• DOI: 10.1007/s11082-023-05155-3

Managing expanding urbanisation, energy use, environmental preservation, citizen economic and living standards, and people's ability to effectively use and adopt modern information and communication technology (ICT) are all objectives of smart cities. A branch of machine intelligence engineering known as explaining artificial intelligence (XAI) makes complex techniques approachable and adaptable for efficient decision-making in the sciences and technologies. The quantum uncertainty problem may be applied to the network state, which consists of several states and dimensions and requires real-time information. Specifically pertinent are the linkages between the emerging paradigms of machine learning (ML), quantum computing (QC), and quantum machine learning (QML), and traditional communication networks. This study provides a new method in explainable deep learning for analysing healthcare data in multimedia for long-term quantum photonic applications. Input has been collected from multimedia healthcare data and processed for noise removal and normalization. Processed data features have been extracted using a gradient quantum neural network, and classification is carried out using an attention-based graph neural network (GNN). The experimental analysis is carried out in terms of accuracy, precision, recall, F-1 score, NSE (normalized square error), MAP (mean average precision),and Jaccard index.The proposed technique attained an accuracy of 93%, precision of 81%, recall of 79%, F-1 score of 71%, NSE of 65%, MAP of 58%, and Jaccard index of 53%.