人工智慧技術於智慧醫療之理論探討與實務應用

Artificial Intelligence in Smart Health: Investigation of Theory and Practice

林書弘(Shu-Hung LIN) ； 陳牧言(Mu-Yen CHEN)

《護理雜誌》 66卷2期 (2019/04) Pp. 7-13

https://doi.org/10.6224/JN.201904_66(2).02

摘要

世界衛生組織（World Health Organization）對於「智慧醫療」（smart healthcare）的定義為：「資通訊科技（information and communication technology）在醫療及健康領域的應用，包括醫療照護、疾病管理、公共衛生監測、教育和研究」。另外，許多學者也認為「智慧醫療」是指醫學資訊（medical informatics）、公共衛生、商業應用的整合，主要是透過網際網路及相關的人工智慧與資料探勘技術，提供更精準的個人化健康服務與健康資訊的應用。近幾年來，深度學習（deep learning）發展十分迅速且熱門，尤其是在2016年AlphaGo擊敗南韓棋王李世後，深度學習變得更廣為人知。而深度學習最常應用於圖像辨識與物體辨識，像棋譜、畫作或者是圖片中的人事物等之類的辨識，而另一個常用的範疇則是用來進行資料的特徵萃取。在過去的研究中，智慧醫療的研究與應用很少使用機器學習或是深度學習的方式進行分析，大多是使用傳統的統計、迴歸分析的運算進行預測。所以本文將從機器學習與深度學習方法來進行智慧醫療的整合運用。

關鍵字

智慧醫療 ； 人工智慧 ； 機器學習 ； 深度學習

並列摘要

The World Health Organization defines Smart Healthcare as ＂Information and Communication Technology applications in the medical and health fields, including medical care, disease management, public health monitoring, education, and research.＂ In addition, many scholars believe that ＂Smart Healthcare＂ refers also to the integration of medical informatics, public health, and business applications mainly through the Internet and related artificial intelligence and data mining technologies in order to provide more accurate personal healthcare services and health information. The concept of deep learning has gained ground rapidly in recent years. While deep learning is usually applied to the studies of image/object recognition such as board game notations, paintings, people/things/objects in pictures, and so on, it is also often applied to the extraction of features. However, researchers have rarely used deep learning methods to predict outcomes in the medical and healthcare fields, preferring instead to make these predictions using algorithms based in traditional statistical methods and regression analysis. This paper introduces and investigates deep learning methods in the context of predicting outcomes in the medical and healthcare fields.

並列關鍵字

smart healthcare ； artificial intelligence ； machine learning ； deep learning

參考文獻

Antipov, G., Baccouche, M., Berrani, S. A., & Dugelay, J. L. (2017). Effective training of convolutional neural networks for face-based gender and age prediction. Pattern Recognition, 72, 15–26. https://doi.org/10.1016/j.patcog.2017.06.031

Fukushima, K., & Miyake, S. (1982). Neocognitron: A self-organizing neural network model for a mechanism of visual pattern recognition. In S. Amari & M. A. Arbib (Eds.), Competition and cooperation in neural nets (pp. 267–285). Berlin, Germany: Springer. https://doi.org/10.1007/978-3-642-46466-9_18

Hinton, G. E., & Salakhutdinov, R. R. (2006). Reducing the dimensionality of data with neural networks. Science, 313(5786), 504–507. https://doi.org/10.1126/science.1127647

Hubel, D. H., & Wiesel, T. N. (1962). Receptive fields, binocular interaction and functional architecture in the cat's visual cortex. The Journal of Physiology, 160(1), 106–154. https://doi.org/10.1113/jphysiol.1962.sp006837

Indolia, S., Goswami, A. K., Mishra, S. P., & Asopa, P. (2018). Conceptual understanding of convolutional neural network-A deep learning approach. Procedia Computer Science, 132, 679–688. https://doi.org/10.1016/j.procs.2018.05.069