《高力国际:2023Q2美国医疗服务报告-医疗中的人工智能(英文版)(12页).pdf》由会员分享,可在线阅读,更多相关《高力国际:2023Q2美国医疗服务报告-医疗中的人工智能(英文版)(12页).pdf(12页珍藏版)》请在三个皮匠报告上搜索。
1、Healthcare ServicesArtificial Intelligence in HealthcarePart I:Adoption to Date and Future ProspectsColliers U.S.Research Report:Q2 20232|Q2 2023 Healthcare Services ReportOverviewArtificial Intelligence(AI)has become a constant topic in todays headlines.While AI is not a new concept,the development
2、 of generative AI applications,such as ChatGPT,is driving major interest in AIs impact on business and personal life.In this two-part report,we explore the role and scope of AI in the healthcare industry:Part I provides an introduction to AI and its current usage in healthcare,plus a review of the l
3、atest developments in AI,and their potential impact on healthcare services.Part II will examine AI from a management perspective.It will address how healthcare systems can develop an effective AI strategy,including key issues to resolve and implementation challenges.Key Findings The use of AI applic
4、ations in business is widespread and longstanding.Recent developments in generative AI,most notably OpenAIs chatbot ChatGTP,have raised interest in AI to an all-time high.AI is already being used in healthcare to identify a wide range of clinical conditions and generate treatment plans.It is also be
5、ing applied to data driven processes.Healthcare system gains from utilizing AI include greater efficiency,better diagnostics,a higher level of customer service and optimal talent management.The application of generative AI in healthcare is in its infancy but is poised to grow rapidly.Several health
6、systems are already experimenting with generative AI.The potential for generative AI usage ranges across a wide spectrum of healthcare services and functions.Colliers U.S.Research|3Defining Artificial IntelligenceIn its broadest sense,AI is a term for a wide array of technology focusing on creating
7、and training machines to perform intelligent tasks.Key AI functions include problem-solving,knowledge representation,machine learning,language processing,perception and image recognition,along with the development of machines with both general and social intelligence capabilities.The concept of AI f
8、irst emerged in the 1950s and has long been used for analytics to identify trends and patterns and make predictions across huge sets of data.At a workshop in 1956,the Dartmouth Institute commented that“every aspect of learning or any other feature of intelligence can be so precisely described that a
9、 machine can be made to simulate it.”Apart from a period in the 1980s,AI research did not truly flourish until the late 1990s and mostly centered on its application for problem solving in collaboration with other fields,including statistics,economics and mathematics.This led to rapid development in
10、AI and its capabilities,fueled by faster computers and access to large amounts of data.Moving forward,Bloombergs Jack Clark commented that“2015 was a landmark year for AI,with the number of software projects that use AI within Google increasing from sporadic usage in 2012 to more than 2,700 projects
11、.”Clark identified advances in cloud computing as a key driver of this growth.The results of a 2017 survey reported that one in five companies had“incorporated AI in some offerings or processes.”Similarly,the amount of research into AI,as measured by the number of publications,is estimated to have g
12、rown by 50%from 2015 to 2019.Global spending on AI is forecast to reach$301 billion in 2026,more than three times the volume seen in 2021.Source:IDC05003003502022420252026Global Spending on AI($Billions)4|Q2 2023 Healthcare Services ReportIn recent years AI technology has fuele
13、d the creation of virtual assistants.Apples Siri is one of the best-known examples of the use of AI in everyday life.Other popular applications include web search engines,such as Google Search,and recommendation systems,including Amazon,Netflix and YouTube.How is AI being used in business?Research f
14、rom TechTarget,published in the report“A Guide to Artificial Intelligence in the Enterprise,”identified seven key benefits:Productivity and efficiency gains Improved customer service Improved monitoring Faster product development Better quality Better talent management Business model innovation and
15、expansionTechTargets study also shows the wide range of industries using AI.These include agriculture,education,financial services,healthcare,legal services,manufacturing and transportation.Colliers U.S.Research|5Generative AIThe field of generative AI is not new,but recent high-profile advancements
16、 in performance and funding are causing significant interest.The current focus is on a few large language models,most notably OpenAIs ChatGPT,a publicly available chatbot service that Microsoft launched in November 2022.Over one million people signed up to use ChatGPT in its first five days.McKinsey
17、 defines generative AI as follows:“Generative AI describes algorithms(such as ChatGPT)that can be used to create new content,including audio,code,images,text,simulations,and videos.Recent new breakthroughs in the field have the potential to drastically change the way we approach content creation.”Ge
18、nerative AI systems fall under the broad category of machine learning,which allows models to learn from data patterns without human direction.ChatGPT has the potential to benefit a wide range of businesses that need to produce clear written materials.This can also be applied to more technical needs,
19、such as coding and high-resolution images.With the time saved,organizations can focus more on business development.Outside of large language models,the latest generation of generative AI applications are mostly in the research and development stage.There is no widespread commercial usage beyond pilo
20、t projects.The response to ChatGPT,and machine learning in general,has been largely positive.A 2022 McKinsey survey shows that AI adoption more than doubled over the past five years,along with a rapid uptick in AI investment.According to CB Insights,AI startups received$2.6 billion in funding in 202
21、2,up from just$271 million in 2020.However,the growing use of generative AI also brings fears,particularly from people who earn their living through content creation ranging from copywriters to academics.Concerns have also been expressed about the inappropriate use of generative AI,output reliabilit
22、y,reduced human involvement in business processes,data security and the need for regulation.Artificial Intelligence in HealthcareAccording to Wikipedia,“AI in healthcare is an overarching term used to describe the use of machine-learning algorithms and software,to mimic human cognition in the analys
23、is,presentation,and comprehension of complex medical and healthcare data by providing new ways to diagnose,treat,or prevent disease.”6|Q2 2023 Healthcare Services Report“The primary aim of health-related AI applications is to analyze relationships between clinical and patient outcomes.AI programs ar
24、e applied to practices such as diagnostics,treatment protocol development,drug development,personalized medicine,patient monitoring and care.”Clinical applications of AI include cardiovascular,dermatology,gastroenterology,infectious diseases,oncology,ophthalmology,pathology,primary care,psychiatry a
25、nd radiology.AI is also used in system applications,including disease diagnosis,drug interactions,electronic health records and telemedicine.The value of the global AI healthcare market,which stood at$15.1 billion in 2022,is set for a period of phenomenal growth.One forecast calls for a more than te
26、n-fold increase to$188 billion by 2030.A 2023 report from the Center for Connected Medicine at Pittsburgh-based UPMC identified the top seven uses of AI by healthcare systems as:Health/disease management and prediction Operational optimization Imaging Population health management Clinical research,V
27、alue-based care Patient/member managementHow is AI of benefit to healthcare systems?In a 2022 paper,the Advisory Board commented,“AI can enhance a wide range of objectives in healthcare by improving the cost,speed,accuracy and consistency of various processes.”AI in Healthcare Projected Global Marke
28、t Size($Billions)Source:Statista02040608000232024202520262027202820292030Colliers U.S.Research|7The Advisory Board study identified three key areas where AI systems could provide advantages:Improving speed,capacity,consistency and accuracy:Computers retain what they have learne
29、d,can work constantly,and recognize patterns in data more quickly and comprehensively than most humans.Address the increased amount of data:The breadth and depth of healthcare data are growing exponentially.AI can process the rapidly expanding amount of data from sensors,medical devices,mobile apps
30、and electronic health records.Enhancing the abilities and efforts of human workers:AI can help free human experts from repetitive or tedious tasks,provide constant monitoring,and discover patterns or anomalies,augmenting human capabilities.8|Q2 2023 Healthcare Services ReportCase Study 1:CastellCare
31、 coordinators working in primary care at Castell,an Intermountain Health company in Utah,were manually reviewing charts for four hours a day to identify patients due for preventative care services,preventing them from accomplishing higher-value tasks.Castell used Intermountains existing partnership
32、with AI provider Notable to introduce“digital assistants”to perform medical record searching.Four beneficial outcomes from this approach were identified:Improved efficiency by reducing the time that care coordinators spent reviewing charts.Cost savings which enabled Castell to increase the time avai
33、lable for patient outreach.Greater care coordinator engagement by allowing them to work at the top of their skill set.Improved quality of care by freeing up time for doctors to interact with patients.Case Study 2:Flagler HealthcareHospital-acquired conditions(HACs)such as pneumonia are difficult to
34、detect and costly to treat.The most common approach to this situation is identifying patients at risk of deterioration,then implementing a standard treatment plan.Such an approach starves systems of valuable resources and targets patients when they have become ill,rather than preventing deterioratio
35、n in the first place.Flagler Hospital,based in St.Augustine,Florida,partnered with machine intelligence software company Symphony AyasdiAI to proactively identify care paths that prevent HACs.Within the first few months of implementation,length of stay,mortality and costs all decreased.Flagler first
36、 applied this approach to pneumonia and has since moved on to cover 11 other conditions.Additionally,a 2023 study by Futurescan defined three top opportunities for using AI in healthcare:clinical decision tools,diagnostics and imaging and patient safety.Two case studies are presented below to illust
37、rate real world examples of the usage of AI in healthcare.Colliers U.S.Research|9The Potential for Generative AI in HealthcareThe use of generative AI in healthcare has been limited to date.However,recent high-profile advancements and increased funding suggest it will likely impact healthcare soon.T
38、he Advisory Board has identified several potential areas in which generative AI could be used in healthcare:Expanding the scope of telehealth and triage.Using synthetic data in drug discovery and clinical trials.Integrating disparate data sets to identify rising-risk populations earlier.Summarizing
39、medical conversations in plain language for patients.Possible benefits resulting from the use of Generative AI include:Faster and cheaper drug discovery.Improved clinician workflows.Better patient experience through support tools.Expanded access to care via digital therapeutics.Reflecting on the buz
40、z around the use of generative AI,John Brownstein,Chief Innovation Officer at Boston Childrens Hospital,commented,“Right out of the box,I dont think Ive seen anything as transformative since the iPhone or Google.”Generative AI is already being tested and investigated by both providers and healthcare
41、 systems.Microsoft is working with healthcare software firm,Epic Systems,to apply OpenAIs GPT-4 model to electronic health records.Epic has already found three partners for its GPT-4 integration;Wisconsin-based UW Health and UC San Diego Health have signed on,while Stanford Health Care in California
42、 looks set to adopt this functionality soon.Baptist Health and Cleveland Clinic are experimenting with generative AI by collaborating with Microsoft on using generative AI tools for clinical and administrative functions.This includes summarizing data for registry review and providing relevant diagno
43、stics information.Both systems hope to implement generative AI into clinical workflow later this year.Parting ThoughtsGenerative AI is set to be applied to the healthcare industry at a rapid pace.Early adoption has mostly focused on data analysis and administrative tasks,but the potential for AI usa
44、ge ranges across the entire spectrum of healthcare services and functions.The introduction of ground-breaking technology,such as generative AI,brings fear as well as optimism.Some of these concerns include accuracy,the need for human involvement,job losses,data security and regulation.Part II of our
45、 report will examine these issues in more detail and look at how to implement a successful,system-wide AI strategy.Key issues for healthcare systems and barriers to implementation will also be addressed.10|Q2 2023 Healthcare Services ReportColliers U.S.Research|11This document/email has been prepare
46、d by Colliers for advertising and general information only.Colliers makes no guarantees,representations or warranties of any kind,expressed or implied,regarding the information including,but not limited to,warranties of content,accuracy and reliability.Any interested party should undertake their own
47、 inquiries as to the accuracy of the information.Colliers excludes unequivocally all inferred or implied terms,conditions and warranties arising out of this document and excludes all liability for loss and damages arising there from.This publication is the copyrighted property of Colliers and/or its licensor(s).2023.All rights reserved.This communication is not intended to cause or induce breach of an existing listing agreement.Healthcare ServicesShawn JanusNational Director,Healthcare Services+1 312 612 ResearchStephen NewboldDirector,National Office Research|U.S.+1 202 534