last updated on 27 April 2021
AI in the healthcare system: an overview by a Licensed Lebanese Dietitian
COVID-19 shook the healthcare system
In 2008, after the Great Financial Rescission, a boost for the financial technology sector took place. In 2019, a similar scenario occurred, in fact, COVID-19 shook the healthcare system. The fundamentals of this system are being reassessed and revaluated, indeed COVID-19 had shed the light on the strengths and weaknesses of the current healthcare system. COVID-19 has been pushing to a quicker digitalization of the world, especially the healthcare system. The emerging pandemic guided the enthusiasm and interest in healthcare technologies such as artificial intelligence (AI), digital health as well as telemedicine which will all lead to the revolutionization of the healthcare system in the upcoming years 1,2.
在金融海嘯2008年之後，金融技術發展提升。 而在2019年發生了類似的情況：COVID-19憾動了整個醫療系統。 確實COVID-19揭示了當前醫療體系的優點和缺點，使得整個醫療系統的基礎開始重新評估。 COVID-19加速了整個世界數位化，尤其在醫療系統。 這新興流行病導向人們對人工智能AI、數位醫療以及遠距醫療等醫療技術的熱情和興趣，而也預計在未來幾年內引發醫療系統的革命
AI has shown to play a crucial role in healthcare
AI has shown to play a crucial role in healthcare. AI has the ability to deliver high performing data-driven medicine, enhance patient care, improve administrative processes, recommend the right therapy for the right patient as well as support clinical decision making (Fig. 1). AI is defined as the ability of machines to perform tasks that are associated with human minds and intelligence- such as solving a problem. The different types of learning include deep learning, neural networks, and natural language processing. For example, AI could analyse large and complex datasets and discover patterns and trends in these data that could help and lead to predictive analysis. It has been shown that AI can perform as well as or even better than humans at some of the major healthcare duties such as diagnosis disease 3. For example, these models are outdoing radiologists at detecting malignant tumours.
AI提供更高效率的數據藥物、增強病人照護、改善管理流程、媒合病人、治療及支持由醫護人員的臨床決策（Fig. 1）。 AI被定義為「機器去執行與人類思維和智能相關的任務（例如解決問題）的能力」， 而這些分為不同類型：包括深度學習、神經網絡連結、及語言處理
One of the major features of care in digital health is patient engagement and adherence applications
One of the major features of care in digital health is patient engagement and adherence applications, which are known as the final obstacle between positive health results and ineffective outcomes. Examples of patient adherence to treatment include following a certain diet, improving lifestyle behaviour or taking prescribed medication which are all recommended by a healthcare professional4. As we all know, obesity rates and unhealthy behaviours are increasing globally (Fig. 2) 5, the effort to reach and sustain beneficial weight loss continues to be a big challenge for public health and nutrition specialists. It is well established that patients proactively engaging and contributing to their wellbeing end up with improved outcomes. Therefore, a lot of effort has been put on developing machine learning and business rules models to help in the continuum of care. These include sending or nudging patients with messaging alerts that causes a certain action to be taken using smartphones, smartwatches, and conversational interfaces 4. For example, these tools can be helpful in reminding people to go for a walk, drink water, or eat a healthy snack.
數位醫療中的主要特色之一：是病人參與度和遵從度的應用，這亦是正面健康結果與無效結果差別的最終障礙像是病人的遵從度至治療：遵循一定的飲食習慣、改善生活習慣或遵從服用處方用藥—來自醫療保健專業人員推薦的4。眾所周知，肥胖比例及不健康行為在全球劇增（Fig. 2）5，如何達到及保持減重的好處仍是公共衛生和營養專家面臨的重大挑戰。我們確定的是，積極參與並為他們的健康做出貢獻最終會改善。因此，許多努力在(1)開發機器學習Machine Learning、和(2)商業規則模版方面，去幫助實現持續照護。進一步來說，在使用智能手機，智能手錶和對話界面來通知或提醒病人做出行動，像是提醒人去散步、喝水或吃飯、甚至去選擇健康的零食5
An AI nutrition APP, which widely used in Japan
In addition, there is an AI nutrition mobile phone application widely used in Japan that helps people lose weight. This app is known for its main feature of having several food-logging interfaces, the most important one is the AI-powered photo analysis feature which has a huge database of more than 100,000 menus of food that can be homemade or from restaurants/markets. For example, if you post a picture of an ice cream (Fig. 3), the application will recognize the frame of each element, the menu and serving amount. After confirmation, it gives the user the ability to select the correct brand of the icecream from 82 types/brands in the database. And finally, the app calculates and analyse the energy and nutrient intakes the user is consuming through these food logs, helping him/her maintain a healthy and balanced food intake and behavioural changes.
在日本，廣泛運用AI營養APP來幫忙減重。這APP擁有多個食物記錄界面，而其中最重要的是由AI驅動的照片辯識功能，該功能具有龐大的數據庫，包含超過100,000種可自製或來自餐廳/市場的食物菜單。 例如，張貼一張冰淇淋圖片（Fig. 3），則應用程序將識別每個元素框架、菜單和食用量。 經確認後，它給予使用戶從數據庫中的82種/品牌中選擇正確的漢堡品牌。 最後，該應用程序通過這些食物記錄來計算，並分析用戶消耗量和攝食量，進而幫助保持健康且均衡的飲食攝取、和行為改變
Without any geographical restrictions, P’t are able to access RD services and have better quality care.
Another method of integrating nutrition and digital health is through remote healthcare and nutrition consults. This method is shown to be crucial during delicate times such as the COVID-19 pandemic. Following up with patients virtually through telehealth is more convenient, accessible, and safer for patients as well as the dietitians and nutritionists. In Lebanon, this method is widely used, in fact, the largest hospitals in Beirut, Lebanon implemented and started using telehealth in their practice (Fig. 4). This also comes in handy in the nutrition field, most dietitians shifted their practice to online consultations- Telenutrition. This virtual shift helped nutritionists and dietitians to open more doors for their careers on a national and international level, where patients all over the world, without any geographical restrictions, are able to access their services and have better quality care.
另一種方法，是通過遠程醫療和營養諮詢來整合營養和數位健康。 在COVID-19這種流行病傳染的脆弱時期，被證明這是至關重要的。通過遠程醫療實際上使患者以及營養師更加方便、可訪問且更安全。 在黎巴嫩，這種方法得到了廣泛使用。實際上，黎巴嫩的貝魯特(Beirut, Lebanon)最大的醫院正在實施使用遠距醫療（Fig. 4）。這也展現在營養領域：大多數營養師將他們的做法轉向了線上諮詢-Telenutrition。這種線上的轉變，幫助營養師在國家和國際層面上打開了更多的大門：不受任何地理限制之下，世界各地的病人能夠獲得營養諮詢及更好的營養照護
- Davenport T, Kalakota R. The potential for artificial intelligence in healthcare. Future Health J. 2019;6(2):94-98.
- Leung, A.W.Y.; Chan, R.S.M.; Sea, M.M.M.; Woo, J. An Overview of Factors Associated with Adherence to Lifestyle Modification Programs for Weight Management in Adults. Int. J. Environ. Res. Public Health 2017, 14, 922.
Lara Ghandour, MSc, RD
Language: Arabic, French, English
Lara Ghandour is a licensed dietitian in Lebanon. After graduating with a bachelor’s degree in Nutrition and Dietetics, she worked for a year as an in-patient dietitian at a hospital. She assessed the nutrition needs of more than 300 patients with different acuity and type of diseases. She also participated in several health awareness events in refugee camps all over Lebanon aiming to raise awareness on nutrition and its effect on chronic diseases. Then she decided to continue her education in London, she graduated from University College of London (UCL) with a masters’ degree in Clinical and Public Health Nutrition. She is passionate about research and is currently working in medical research. To keep up with new healthcare trends especially the ones that arose due to COVID-19, she decided to enhance her skills and took few online courses on Artificial Intelligence in Healthcare from Stanford University, which will be discussed in this article.
Contact: Lara.firstname.lastname@example.org // +9613146416