人工智能可以帮助我们以更有效的方式抵抗传染病

“如果未来几十年内某个东西会杀死超过1000万人的话，那么很大的可能它是一种具高度传染性的病毒，而不是战争。它会是微生物而不是导弹。 造成这种情况的一部分原因是因为我们在核威慑上投入了大量资金，而实际在遏制流行病系统的投入资金很少。” ——比尔·盖茨2015年3月的TED演讲片段

> Bill Gates speaking at a TED Talk in March 2015

Throughout history humans have struggled against infectious diseases. Although many effective medicines have been developed, new viruses continue to challenge us. The COVID-19 pandemic has created a sense of urgency to improve current approaches to prevent and treat infectious diseases. Fortunately, currently available AI and data science approaches can help us fight infectious diseases in a more effective way人类在整个历史里都一直在与传染病作斗争。尽管到目前为止已经开发了许多有效的药物，但新的病毒却不停地在挑战我们。COVID-19大流行使得我们有了一种紧迫感，我们要改善目前传染病的预防和治疗方法。幸运的是，时下的人工智能（AI）和数据科学方法可以帮助我们以更有效的方式抵抗传染病。

Three major breakthroughs in the 20th century - sanitization,  immunization, and antibiotics, improved human health and increased average life expectancy by at least 20 years. In the last 100 years, humans have challenged viruses more than they have in thousands of years. Although we’ve made progress, new infectious diseases have continued to emerge and some have lead to serious outbreaks and pandemics.

20世纪里有三大突破：消毒、免疫和抗生素，三大突破改善了人类健康，人类的平均预期寿命至少延长了20年。 在过去的100年里，人类对病毒发起的挑战比过去几千年的挑战还多。尽管我们已经取得了很多进步，但新的传染病仍在不断出现，其中的一些病导致了大爆发甚至流行病。

**Exponential Growth of Biotech Science**

生物科技呈指数增长

In 2015, a biotech conference entitled [_Opportunities and Risks in Exponential Growth of Biotech Science_](https://www.cser.ac.uk/events/the- future-of-biotech-enterprise/ "https://www.cser.ac.uk/events/the-future-of- biotech-enterprise/") __ was held at the University of Cambridge. At the conference, subject matter experts examined how bioscience technologies have the power to build or destroy our world and discussed how to leverage entrepreneurial opportunities while avoiding catastrophic risk. The speakers at the conference, Professor of Chemical Engineering and Biotechnology Cris Law, Professor of Infectious Disease Informatics Derek Smith, and founder of Deep Knowledge Ventures Dmitry Kaminskiy, reached two major conclusions. They concluded that basic forms of AI and data science should be used to optimize the management of preventive medicine, and strong AI should be applied for the development of new vaccines. They cautioned that significant attention should be focused on preventing mistakes where viruses could spread from laboratories, and preventing advanced biohazardous technologies from reaching the hands of terrorists.

2015年，一个名为“生物技术科学迅猛增长的机会和风险”的生物科技会议在剑桥大学举行。一些生物科技领域的专家在会上就生物科技对于建设或摧毁我们世界的威力进行了探讨，他们还讨论了如何利用创业机会而同时避免灾难性风险等问题。化学工程与生物科技学院的Cris Law教授、传染病信息学教授Derek Smith以及Deep Knowledge Ventures 创始人Dmitry Kaminskiy三位是会议的演讲者。他们得出了两个主要结论。他们的结论是，基础AI和数据科学应该用于优化预防医学的管理，而强势AI则应该用于新疫苗的开发。他们还提出警告，一定要特别注意防止病毒可能从实验室传播出来的失误以及要防止先进的生物危害技术落入恐怖分子之手。

[](https://www.cser.ac.uk/events/the- future-of-biotech-enterprise/)

Centre for the Study of Existential Risk University of Cambridge

生存风险研究中心（剑桥大学）

In a pandemic, as the number of people infected with the virus increases, the rate of growth also increases. Technology and pandemics are both evolving exponentially, but sources of mega-pandemics from viruses and microbes are evolving at a double exponential rate. This includes natural viruses but could potentially include weaponized viruses. An example of a weaponized virus would be influenza + HIV embedded in mosquitoes, used in biological warfare. AI could help neutralize this disproportion and allow biopharma to catch up to the rate of pandemic evolution. Current technology can be applied to decrease the threat of pandemics. Even without next-generation AI systems progress can be made.

传染病大流行时，感染病毒的人数增加，因而增长速度也随之增加。科技和传染病大流行都呈指数发展，而病毒和微生物的流行源头却是以双倍指数发展。这些包括了自然病毒，但也可能包括用于武器化病毒。武器化病毒的一个例子是将流感+HIV病毒内置于蚊子，用于生物战。人工智能可以帮助消除这一不平衡关系，使生物制药商可以赶上大流行的发展速度。目前的技术已经可以减少大流行的威胁，即便不利用新一代AI系统都能取得好的进步。

> “The Canadian military is preparing for multiple waves of COVID19 over the next 12 months”   >“加拿大军方正在为未来12个月内可能多次出现的COVID19高峰做准备。”——加拿大武装部队乔纳森·万斯将军

> General Jonathan Vance, Canadian Armed Forces

**Personalized Targeted Immunization**

个性化靶向免疫

There are several approaches that can be taken with currently available technology. One approach is to apply data science techniques to personalized vaccination on a massive scale. Another is to optimize immunization management using AI. By using AI, big data, and small data techniques together, vaccines could be distributed on a massive scale in a more sophisticated and precise way. This could be accomplished by using AI to analyze personal data, such as genetics, to predict an individual’s risk and potential sensitivity to specific vaccines. Then, AI could be applied to design personalized sets of vaccines optimized for each individual. **** AI based individual profiling for optimization of immunization management could be applied now. The required technology is available. AI could be used to analyze personal data such as ethnicity, age, gender, blood type, weight, BMI, frailty, and pre-existing conditions that may put an individual at increased risk to certain viruses. Once profiled, people could receive recommendations for vaccines against specific viruses based on their vulnerability.

目前可用的技术可以采用几种方法。一种方法是将数据科学技术应用于规模性的个性化疫苗接种。另一个方法是利用AI优化免疫的管理。将AI、大数据和小数据技术结合在一起，疫苗的分发就能够规模性地以更精致准确的方式进行。实现上可以是通过AI分析一些诸如遗传学之类的个人数据，预测每个个体对特定疫苗的风险和潜在敏感性。然后再利用AI设计出针对每个个体的经优化过的个性化疫苗。到这里就可以利用基于AI的个人概况优化免疫管理。所需的技术已经存在。AI可以用于分析个人数据，例如种族、年龄、性别、血型、体重、BMI、虚弱程度以及一些可能令个体面临某些病毒风险增加的历史状况。在获取了个体概况后，一些人由于自身存在某个脆弱性就可以收到针对特定病毒的疫苗建议。

[](https://www.dkv.global/covid)

（编辑：应用网_阳江站长网）

【声明】本站内容均来自网络，其相关言论仅代表作者个人观点，不代表本站立场。若无意侵犯到您的权利，请及时与联系站长删除相关内容!