Privacy-Preserving Multiclass Lung Disorder Classification via CNN with Cosine Similarity in Big Data Framework

Jaya Sharma, D Franklin Vinod, Urvashi Chugh

Abstract


Annotating large-scale medical data manually takes a lot of time and human resources, and it requires specific medical knowledge and experience. Big data and Deep learning are two advancing technologies being widely used in the medical field for improved analysis. Because of the recent advancements in imaging technology, computer vision researchers still have unsolved problems related to automatically identifying medical images. Images, however, could include sensitive information about specific body parts and specifics of diseases. In actuality, sharing medical images that contain extremely sensitive information for each user may expose sensitive information to adversaries. One of the main issues between a user and a databank is privacy, we present in this study a Multi-layered convolutional neural network (MLCNN) integrated with PPCS (Privacy preserved cosine similarity) for feature extraction from large-scale medical image data. The framework that uses fully homomorphic encryption (FHE), CSSK(Cheon-Kim-Kim-Song) scheme to search for safe and to enable the categorization CNN is used for large-scale encrypted images. This study aims to diagnose various lung disorders such as COVID-19, lung cancer, pulmonary tuberculosis, pneumonia, and differentiate them from normal conditions by analyzing computed tomography (CT) images. The model's results included 98.54% F1 score, 97.11% Matthew's correlation coefficient (MCC), 98.89% accuracy (AC), 98.38% recall, and 98.81% precision (PC). We compare and contrast our privacy-preserving method with a CNN-based multiclass classification model that offers quick and effective classification.


Full Text:

PDF

References


Hellerstein, Joe. (2012). Parallel Programming in the Age of Big Data. Gigaom Blog. Archived from the original on 7 October 2012.

Segaran, T. & Hammerbacher, J. (2009). Beautiful Data: The Stories Behind Elegant Data Solutions. O'Reilly Media. p. 257. ISBN 978-0-596-15711-1.

Hilbert, M. & López, P. (2011). The world's technological capacity to store, communicate, and compute information (PDF). Science 332, 60(6025), 1095-9203.DOI : https:// 10.1126/science.1200970.

Reinsel, D. et al. (2017). Data Age 2025: The Evolution of Data to Life-Critical (PDF). seagate.com. Framingham, MA, US: International Data Corporation.

Luo, J. et al. (2016). Big data application in biomedical research and health care: a literature review. Biomedical Inf insights,8:BII–S31559

Viceconti, M., Hunter, P., Hose, R.(2015) Big data, big knowledge: big data for personalized healthcare. IEEE J biomedical health Inf, 19(4),1209–15.

Oussous, A., Benjelloun, F., Lahcen, A., Belfkih, S.(2018) Big data technologies: a survey. J King Saud University-Computer Inform Sci,30(4):431–48.

Belle, A., Thiagarajan, R., Soroushmehr, S., Navidi, F., Beard, D., Najarian, K.(2015). Big data analytics in healthcare. BioMed Res Int. https:// DOI : http: // 10. 1155/ 2015/ 370194.

Yang, A., Troup, M., Ho, J.W.(2017). Scalability and validation of big data bioinformatics software. Comput Struct Biotechnol J,15,379–86.

Kouanou, AT., Tchiotsop, D., Kengne, R., Zephirin, DT., Armele, NM., Tchinda R. (2018). An optimal big data workflow for biomedical imageanalysis. Inf Med Unlocked,11,68–74.

Andreu-Perez, J., Poon, C, Merrifield, R., Wong, S., Yang, G.(2015). Big data for health. IEEE J biomedical health Inf,19(4),1193–208.

Cirillo, D. & Valencia, A.(2019). Big data analytics for personalized medicine. Curr Opin Biotechnol,58,161–7.

Pääkkönen, P. & Pakkala, D. (2015). Reference architecture and classification of technologies, products and services for big data systems.Big data research,2(4),166–86.

Schmidhuber, J. (2015). Deep learning in neural networks: an overview, Neural Netw. 61,85–117.

Dieleman, S., Willett, Dambre, J.(2015). Rotation-invariant convolutional neural networks for galaxy morphology prediction, Mon. Notices R. Astron. Soc. 450,1441–1459.

Huval, B., Wang, T., Tandon, S., Kiske, J., Song, W., Pazhayampallil, J., Andriluka, M., Cheng-Yue, Mujica, F. (2015). An Empirical Evaluation of Deep Learning on Highway Driving, arXivPrepr. arXiv,1504,01716.

H. Li et al. (2015). A convolutional neural network cascade for face detection, In, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition,5325–5334.

Farfade, S. et al. (2015).Multi-view face detection using deep convolutional neural networks, In, Proceedings of the 5th ACM on International Conference on Multimedia Retrieval, ACM, 643–650.

Karpathy, A. et al. (2014). Fei-Fei, Large-scale video classification with convolutional neural networks, In, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 1725–1732.

Zbontar, J. & LeCun, Y. (2015). Computing the stereo matching cost with a convolutional neural network, In, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 1592–1599.

Ji S. et al. (2013). 3d convolutional neural networks for human action recognition, IEEE Trans. Pattern Analysis Mach. Intell. 35,221–231.

Sudholt, S. & Fink, G. (2016). Phocnet: A Deep Convolutional Neural Network for Word Spotting in Handwritten Documents, arXiv preprint arXiv,1604.00187.

Tang, et al. (2014). Feature Selection for Classification: A Review. CRC Press, 37–64.

Guyon, I. &Elisseeff, A. (2003). An introduction to variable and feature selection. J. Mach. Learn. Res. 3 (null), 1157–1182.

Vinod, D. & Vasudevan, Vinod. (2019). PPCS-MMDML: Integrated Privacy-Based Approach for Big Data Heterogeneous Image Set Classification: Proceedings of ICTIS, Volume 1. 10.1007/978-981-13-1742-2_43.

Qi, Z., MaungMaung, A., Kiya, H.(2023). Privacy-Preserving Image Classification Using ConvMixer with Adaptative Permutation Matrix and Block-Wise Scrambled Image Encryption. J. Imaging 2023, 9, 85.

DOI :https://doi.org/10.3390/jimaging9040085

Jia, H. et al. (2023) .Efficient and privacy-preserving image classification using homomorphic encryption and chunk-based convolutional neural network. J Cloud Comp 12, 175. DOI :https://doi.org/10.1186/s13677-023-00537-0

Alishahi, M. &Moghtadaiee V &Navidan, Hojjat. (2022). Add Noise to Remove Noise: Local Differential Privacy for Feature Selection. Computers & Security. 123. 10.1016/j.cose.2022.102934

Kassani, S. &Kassani, P.(2019). A comparative study of deep learning architectures on melanoma detection. Tissue Cell 2019, 58, 76–83.

Choe, J., Lee, SM. , Do K-H. , Lee G. ,Lee J-G. , Lee SM. et al.(2019). Deep learning–based image conversion of ctreconstructionkernels improves radiomics reproducibility for pulmonary nodules or masses. Radiology ,292(2),365–73, 2019.

Ioffe, S. & Szegedy, C. (2015). Batch normalization: Accelerating deep network training by reducing internal covariate shift,arXiv:1502.03167. http://arxiv.org/abs/1502.03167

Paillier, P. (1999). Public-Key Cryptosystems based on Composite Degree ResiduosityClasses. EUROCRYPT, 223–38.

Gulli, A. & Pal, S.(2017). Deep Learning with Keras. Packt Publishing Ltd




DOI: https://doi.org/10.31449/inf.v49i10.9531

Creative Commons License
This work is licensed under a Creative Commons Attribution 3.0 License.