A Novel Approach to Machine Learning Application to Protection Privacy Data in Healthcare: Federated Learning

Abstract

Aim: Today, data banks contain unpredictable data. Together with the advances in data science, large data offer the potential to better understand the causes of diseases. This potential results from the processing, analysis or modeling of machine learning algorithms. Various data sets stored in different institutions are not always shared directly due to privacy and legal concerns. This problem limits the full use of large data in health research. Federated learning is aimed at developing artificial intelligence systems based on both high accuracy and data privacy. Materials and Methods: In this study, a federated learning approach was proposed in order to access any data and develop machine learning applications without sharing personal information within the scope of data privacy. Firstly, the structure of the Federated learner has been studied. It was then determined how federated learning should be used in machine learning models in different health applications. Results: In federated learning, the model is trained on local computers and its updates are transferred to a central server. The updated model is then transferred to local models. In this way, the central model is trained without seeing the data. Conclusion: It is necessary to make machine learning models in which confidentiality is applied with data obtained from health. For this, federated learning must be integrated into traditional machine learning applications. Thus, high performance is envisaged to be achieved with big data where data confidentiality is adopted.

Keywords

• federated learning,machine learning,healthcare,privacy,personal data

Sağlık Alanında Veri Mahremiyetinin Korunmasına Yönelik Makine Öğrenmesi Uygulamalarına Yeni Bir Yaklaşım: Federe Öğrenme

Öz

Amaç: Günümüzde veri bankalarını tahmin edilmeyecek büyüklükte veriler içermektedir. Veri bilimindeki gelişmelerle birlikte büyük veriler hastalıklarının oluşum sebeplerini daha iyi anlama potansiyeli sunmaktadır. Bu potansiyel verilerin işlenmesi, analiz edilmesi veya makine öğrenmesi algoritmaları ile modellenmesi sonucunda ortaya çıkmaktadır. Farklı kurumlarda depolanan çeşitli veri kümeleri gizlilik ve yasal kaygılar nedeniyle her zaman doğrudan paylaşılmamaktadır. Bu sorunda sağlık araştırmalarında büyük verilerin tam olarak kullanılmasını sınırlamaktadır. Federe öğrenme hem yüksek doğruluk hem de veri mahremiyetine göre yapay zekâ sistemlerinin geliştirilmesi amaçlanmaktadır. Materyal ve Metot: Bu çalışmada veri mahremiyeti kapsamında kişisel bilgiler paylaşılmadan, herhangibi bir veriye erişmek ve makine öğrenmesi uygulamaları geliştirebilmek için federe öğrenme yöntemi önerilmiştir. Öncelikle federe öğrenmeni yapısı incelenmiştir. Daha sonra federe öğrenmesin farklı sağlık uygulamalarındaki makine öğrenmesi modellerine nasıl kullanılması gerektiği belirlenmiştir. Bulgular: Federe öğrenmede model, yerel bilgisayarlarda eğitilerek merkezi bir sunucuya güncellemeleri aktarılmaktadır. Yerelden gelen güncellemeler merkezi modeli günceller. Daha sonra güncellenmiş model yerel modellere aktarılır. Bu sayede merkezi model veriyi görmeden eğitilmektedir. Sonuç: Sağlıktan elde edilen veriler ile gizliliğin uygulandığı makine öğrenme modellerinin geliştirilmesi gerekir. Bunun için geleneksel makine öğrenme uygulamalarına federe öğrenmenin entegre edilmesi gereklidir. Böylece veri gizliliğin benimsendiği büyük veriler ile yüksek performans elde edilmesi öngörülmektedir.

Anahtar Kelimeler

• federe öğrenme,makine öğrenmesi,sağlık kuruluşu,gizlilik,kişisel veri

References

1. 1 . Huh, S., Cho, S., & Kim, S. (2017). Managing IoT devices using blockchain platform. In 2017 19th international conference on advanced communication technology (ICACT) (pp. 464-467). IEEE.
2. 2 . Lee, I., & Lee, K. (2015). The Internet of Things (IoT): Applications, investments, and challenges for enterprises. Business Horizons, 58(4), 431-440.
3. 3 . Li, H., Ota, K., & Dong, M. (2018). Learning IoT in edge: Deep learning for the Internet of Things with edge computing. IEEE Network, 32(1), 96-101.
4. 4 . Diro, A. A., & Chilamkurti, N. (2018). Distributed attack detection scheme using deep learning approach for Internet of Things. Future Generation Computer Systems, 82, 761-768.
5. 5 . Shakeel, P. M., Baskar, S., Dhulipala, V. S., Mishra, S., & Jaber, M. M. (2018). Maintaining security and privacy in health care system using learning based deep-Q-networks. Journal of medical systems, 42(10), 186.
6. 6 . Demirhan A., Kılıç Y. A., Güler İ. Tıpta Yapay Zekâ Uygulamaları. Yoğun Bakım Dergisi 2010;9(1):31-41.
7. 7 . Lisboa P.J.G. A Review Of Evidence Of Health Benefit From Artificial Neural Networks İn Medical İntervention. Neural Networks 15, p 11-39, 2002.
8. 8 . Topol, E. J. (2019). High-performance medicine: the convergence of human and artificial intelligence. Nature Medicine, 25(1), 44–56. doi:10.1038/s41591-018-0300-7

9. 9 . Hashem, E. M., & Mabrouk, M. S. (2014). A study of support vector machine algorithm for liver disease diagnosis. American Journal of Intelligent Systems, 4(1), 9-14.
10. 10 . Ulagamuthalvi, V., & Sridharan, D. (2012, March). Automatic identification of ultrasound liver cancer tumor using support vector machine. In International Conference on Emerging Trends in Computer and Electronics Engineering (pp. 41-43).
11. 11 . Xian, G. M. (2010). An identification method of malignant and benign liver tumors from ultrasonography based on GLCM texture features and fuzzy SVM. Expert Systems with Applications, 37(10), 6737-6741.
12. 12 . Chu, F., Xie, W., & Wang, L. (2004, June). Gene selection and cancer classification using a fuzzy neural network. In IEEE Annual Meeting of the Fuzzy Information, 2004. Processing NAFIPS'04. (Vol. 2, pp. 555-559). IEEE.
13. 13 . Li, W., Jia, F., & Hu, Q. (2015). Automatic segmentation of liver tumor in CT images with deep convolutional neural networks. Journal of Computer and Communications, 3(11), 146.
14. 14 . Chaudhary, K., Poirion, O. B., Lu, L., & Garmire, L. X. (2018). Deep learning–based multi-omics integration robustly predicts survival in liver cancer. Clinical Cancer Research, 24(6), 1248-1259.
15. 15 . Ye, Q. H., Qin, L. X., Forgues, M., He, P., Kim, J. W., Peng, A. C., ... & Ma, Z. C. (2003). Predicting hepatitis B virus–positive metastatic hepatocellular carcinomas using gene expression profiling and supervised machine learning. Nature medicine, 9(4), 416.
16. 16 . Li, Y., Hara, S., & Shimura, K. (2006, August). A machine learning approach for locating boundaries of liver tumors in ct images. In 18th International Conference on Pattern Recognition (ICPR'06) (Vol. 1, pp. 400-403). IEEE.
17. 17 . Sağlıkla İlgili Uluslararası Belgeler, TTB Yayınları, 2. Baskı, 2009, s:177 18 . İzgi, M. C. (2014). Mahremiyet kavramı bağlamında kişisel sağlık verileri The concept of privacy in the context of personal health data. Türkiye Biyoetik Dergisi, (s 1), 1.
18. 19 . Dülger, M. V. (2015). Sağlık hukukunda kişisel verilerin korunması ve hasta mahremiyeti. İstanbul Medipol Üniversitesi Hukuk Fakültesi Dergisi, 1(2), 43-80.
19. 20 . Hartmann, F., Suh, S., Komarzewski, A., Smith, T. D., & Segall, I. (2019). Federated Learning for Ranking Browser History Suggestions. arXiv preprint arXiv:1911.11807.
20. 21 . Konečný, J., McMahan, H. B., Yu, F. X., Richtárik, P., Suresh, A. T., & Bacon, D. (2016). Federated learning: Strategies for improving communication efficiency. arXiv preprint arXiv:1610.05492.
21. 22 . Yang, Q., Liu, Y., Chen, T., & Tong, Y. (2019). Federated machine learning: Concept and applications. ACM Transactions on Intelligent Systems and Technology (TIST), 10(2), 12.
22. 23 . H. Brendan McMahan, Eider Moore, Daniel Ramage, and Blaise Agüera y Arcas. (2016). Federated learning of deep networks using model averaging. CoRR abs/1602.05629 (2016). arxiv:1602.05629 http://arxiv.org/abs/1602.05629.
23. 24 . Gang Liang and Sudarshan S. Chawathe. (2004). Privacy-preserving inter-database operations. In International Conference on Intelligence and Security Informatics. Springer, 66–82. 25 . Arivazhagan, M. G., Aggarwal, V., Singh, A. K., & Choudhary, S. (2019). Federated Learning with Personalization Layers. arXiv preprint arXiv:1912.00818.
24. 26 . Niknam, S., Dhillon, H. S., & Reed, J. H. (2019). Federated learning for wireless communications: Motivation, opportunities and challenges. arXiv preprint arXiv:1908.06847.
25. 27 . Yang, Q., Liu, Y., Chen, T., & Tong, Y. (2019). Federated machine learning: Concept and applications. ACM Transactions on Intelligent Systems and Technology (TIST), 10(2), 12.
26. 28 . Leroy, D., Coucke, A., Lavril, T., Gisselbrecht, T., & Dureau, J. (2019, May). Federated learning for keyword spotting. In ICASSP 2019-2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp. 6341-6345). IEEE.
27. 29 . Nilsson, A., Smith, S., Ulm, G., Gustavsson, E., & Jirstrand, M. (2018, December). A performance evaluation of federated learning algorithms. In Proceedings of the Second Workshop on Distributed Infrastructures for Deep Learning (pp. 1-8). ACM.
28. 30 . Qian, Y., Hu, L., Chen, J., Guan, X., Hassan, M. M., & Alelaiwi, A. (2019). Privacy-aware service placement for mobile edge computing via federated learning. Information Sciences, 505, 562-570.
29. 31 . Li, T., Sahu, A. K., Talwalkar, A., & Smith, V. (2019). Federated learning: Challenges, methods, and future directions. arXiv preprint arXiv:1908.07873.
30. 32 . Nishio, T. and R. Yonetani, "Client Selection for Federated Learning with Heterogeneous Resources in Mobile Edge," ICC 2019 - 2019 IEEE International Conference on Communications (ICC), Shanghai, China, 2019, pp. 1-7. doi: 10.1109/ICC.2019.8761315
31. 33 . Dwork, C. and A. Roth. The algorithmic foundations of differential privacy. Foundations and Trends in Theoretical Computer Science, 9:211–407, 2014.
32. 34 . Li, Q., Wen, Z., & He, B. (2019). Federated learning systems: Vision, hype and reality for data privacy and protection. arXiv preprint arXiv:1907.09693.
33. 35 . McMahan, H. B., Moore, E., Ramage, D., & y Arcas, B. A. (2016). Federated learning of deep networks using model averaging.
34. 36 . Truex, S., Baracaldo, N., Anwar, A., Steinke, T., Ludwig, H., Zhang, R., & Zhou, Y. (2019, November). A hybrid approach to privacy-preserving federated learning. In Proceedings of the 12th ACM Workshop on Artificial Intelligence and Security (pp. 1-11). ACM.
35. 37 . Xu, J., & Wang, F. (2019). Federated Learning for Healthcare Informatics. arXiv preprint arXiv:1911.06270.
36. 38 . Li Huang and Dianbo Liu. Patient clustering improves efficiency of federated machine learning to predict mortality and hospital stay time using distributed electronic medical records. arXiv preprint arXiv:1903.09296, 2019
37. 39 . Yejin Kim, Jimeng Sun, Hwanjo Yu, and Xiaoqian Jiang. Federated tensor factorization for computational phenotyping. In Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pages 887–895. ACM, 2017.
38. 40 . Theo Ryffel, Andrew Trask, Morten Dahl, Bobby Wagner, Jason Mancuso, Daniel Rueckert, and Jonathan Passerat-Palmbach, (2018). A generic framework for privacy preserving deep learning. arXiv preprint arXiv:1811.04017.
39. 41 . Google, (2019). Tensorflow federated. https://www.tensorflow.org/federated
40. 42 . Webank’s AI, (2019). Federated ai technology enabler. https://www.fedai.org/cn/
41. 43 . doc.ai. Declarative, on-device machine learning for ios, android, and react native. https://github.com/doc-ai/tensorio, 2019.
42. 44 . Gregor Ulm, Emil Gustavsson, and Mats Jirstrand. Functional federated learning in erlang (ffl-erl). In International Workshop on Functional and Constraint Logic Programming, pages 162–178. Springer, 2018.