The current state of healthcare has changed rapidly with the ever-increasing uses of newer and more efficient technologies. While healthcare’s primary premise is to treat the populous of their illnesses and other conditions, it is the practices that have drastically changed over time. These changes have come from a more meaningful understanding of the human body, genetics, infections, and sanitary practices. The advancements in medicine and improvements in patient outcomes can be attributed, not just to the increase in biological understanding, but also to technological advancements that have changed the face of the industry. For example, appendectomies consisted of large incisions and long recovery periods. However, with the innovation of laparoscopic surgeries, large-scale surgeries and recovery time, have both been greatly reduced.
Healthcare is chiefly conducted in hospitals and physician clinics, but with advancements in healthcare technologies through smartphones and the internet, there is a revolution of personalized, at-home solutions, helping patients treat all modes of ailments including emotional, physical, and mental. In this article, I will identify what I perceive to be the trends and technologies of tomorrow in healthcare.
1.0 — Artificial Intelligence
Ideally, the goal of better health should come without dispute or constraints, but with restrictions such as cost and regulations, significant limitations are placed on viable technology, its implementation, and its success in the marketplace. To offset these costs and regulatory restrictions and to enhance current healthcare practices, more intelligent technologies are crucial. Thus, the future of healthcare cannot be discussed without talking about Artificial Intelligence (AI).
Before I expand on the impact of AI in different aspects of healthcare over the next five years, I want to clarify the scope of this discussion. A misconception held by a majority of people is between true AI, and machine learning (Figure 1). The difference between the two can be, greatly, simplified to the following definition: AI relies on the principle that the technology/program is capable of independent thought — artificial, independent thinking. The purpose of AI-based programs is to create machines that can perform as humans, or better yet, outperform human capabilities. Machine learning, on the other hand, deals with the principles that a program can accurately predict trends and make decisions based on learnings from training data. Machine learning can be defined as a major subgroup of AI — it is a substantial part of the foundation of AI. Deep learning is a subset of machine learning in which more complex data requires a more sophisticated approach to building algorithms and models.
1.1 — AI in Pharma
The massive data collection and data utilization currently conducted in the pharmaceutical field (pharma) is rife for AI applications. Alongside the data, pharma possesses the talent, the resources and hunger for innovations, whereby to fund research and advancements into AI. One such aspect for pharma is the process of drug discovery, a complex and labour-intensive task requiring immense amounts of attention. Currently, high-throughput screening has proven to be successful for pharma, but the industry requires its next advancement.
With the help of big pharma companies, who have the means to enter new avenues of research and technologies by acquiring or partnering with independent, third-party organizations and keeping in mind the current state of AI, I believe there are a few companies which can produce and bring to market viable products within the next five years so [2].
In my opinion, one of the most important aspects of AI deals with making sense of incomplete, “dirty”, and unstructured data. For pharma, the process of drug discovery is complex and requires the most attention. AI-enabled technologies and software products can be used to parse literature, clinical trials, publicly accessible data, and other sources to narrow down a relevant data universe based on text queries. For example, the AI tool would make it easier for individuals to find a plethora of data and evidence related to a specific topic in a much more efficient way. This has the potential to significantly reduce the time spent gathering evidence and gaining an understanding of potential drug targets, therapies, and more. Iris is an AI assistant which uses natural language processing (NLP, a subset of deep learning), to consolidate research documents in a list format. This technology claims to reduce the time spent in creating reading lists by about 90%. While this is still a small start-up, I believe that their software will prove to be critical in the pharmaceutical industry since it has the capabilities to efficiently accelerate research and development [2]. Since it is a relatively young company, there is a possibility that pharma will either partner with, or, acquire Iris. According to Deloitte, the early adoption of AI technology in the pre-clinical research and development phase can effectively improve the process by a factor of 15 [3]. Furthermore, NLP tools can be extremely beneficial in the medical affairs and sales aspects of pharma where social media mining can be used to identify key opinion leaders in different healthcare sectors who have a substantial digital presence [4]. Linguamatics is a subsidiary of IQVIA, which has been leading the NLP space with its I2E and iScite products for the past few years; and because it is highly reputed and well-trusted, there is high scope for growth since the cost to bring a drug to market can be reduced through more accurate efficacy and safety testing [3].
Data analytics for Drug Discovery. This sounds may sound obvious; you may even say it is too simple to be placed under the AI umbrella since this is something we have been doing for decades. However, in pharma, this is an important aspect to discuss because the better the data, the more diverse the data, and the more comprehensive the data, the easier it is to identify problems, predict solutions, and make changes to processes. Currently, pharma has access to limited amounts of data — patient data is difficult to acquire in an anonymized manner. Sensyne Health has partnered with the National Health Service (NHS) in the United Kingdom to acquire anonymized patient data from which larger datasets are created using AI algorithms. However, Sensyne Health not only curates these datasets, but uses sophisticated machine and deep learning algorithms to generate novel insights that can be used to enhance medical research, drug development, and improve patient care. This is an important company to watch out for because it has found and maintained footing in the UK market and is now looking to expand by forming working partnerships with Bayer and, most recently, Roche [5]. Furthermore, Intellegens is a UK based company that has been recognized by KPMG and Boeing for their AI algorithms that generate meaningful analytic insights from sparse and extremely unstructured datasets. Since their algorithm is diverse, in pharma, it would have the ability to help in drug discovery by offering insights from considerably incomplete data.
The use of AI in drug discovery is almost ubiquitous as of now, specifically, in identifying targets and molecules. One of the aspects of biology that has always fascinated me deals with protein folding and the numerous diseases that are based on improper protein folding. Scientists have been plagued with accurately predicting protein folding — but AI can change that. Google’s DeepMind has successfully made progress with accurately predicting protein folding structure [6]. Many startups focus on predicting protein structure — however, DeepMind is expected to outperform them all with Google’s resources. While there are other aspects of drug development that AI can affect, I think improving drug screening and clinical trial efficiency through better access to the data, improving comprehensive information access (for medical affairs and sales) is going to be a key trend for pharma because they are relatively easy services to implement but come with high reward. Additionally, protein folding diseases are prevalent in almost all areas of biology, especially in neurobiology, and any sort of progress in accurate protein modelling can take therapies to the next level [7].
1.2 — AI in Insurance Companies
AI in healthcare is not limited to the pharma industry, it is finding substantial traction in telemedicine, women’s health, personalized genetics, disease management and more. Sensentia is a company that uses AI and NLP to aid health insurance call-centre representatives to answer customer questions. Why do I think this is important? Getting answers to my insurance questions has always been a magnanimous task — having a software that can make it easier to get answers will not only improve the customer experience, but it can also relieve the insurance companies of the amount of time it takes to get through customer calls. Efficiency for both parties involved! Similarly, Eigen Technologies is a machine learning-based startup that uses its algorithm to gather insights from complex financial and contractual documents [8]. While this is the primary focus, it is expected that the company expands into healthcare and other industries that have burdensome documentation. Eigen and Sensentia seem to have similar value propositions and services, and while Sensentia has a head start in the healthcare industry, Eigen has significantly more financial backing.
AI domination in the future of healthcare is inevitable, it is already here.
2.0 — The 5G Revolution
We are currently undergoing a major change– The Fourth Industrial Revolution (4thR). The 4thR has been occurring for a while now and is a result of the increasing digital technologies that allow for the fusion of physical, biological, and digital aspects of life [9]. This revolution deals with the billions of possibilities that arise from the interconnectedness of devices and tools we use today, such as smartphones [9]. The 4thR will come into full force once 5G officially gets put into action — because not only will it increase internet speeds, it will decrease latency periods from 20 milliseconds to 2 milliseconds (10x). Imagine your internet flows through a pipe; until now the flow of the internet was being increased (speed), but 5G makes the pipe wider (latency) along with increasing the flow. The importance of 5G becomes apparent when its use-cases are discussed in various forms of technologies that already exist but will be augmented (significantly) with 5G. It is a matter of timing for the following technologies — they arrived too early, and the infrastructure was not in place to support their true potential.
2.1 — Augmented Reality and Virtual Reality
Augmented reality (AR) is one where the real world is virtually changed, while virtual reality (VR) is one where a simulated world is created. The AR market is expected to grow, and 5G connectivity will be a leading cause of this growth since the basis of the technology is fast connectedness. Google Glass is an AR headset which failed to reach consumer markets when it was first launched, but it announced this week that a second edition of the technology will be available for direct purchase. While it is still not meant for consumer use, the removal of this purchasing barrier for organizations that will benefit from the product, adds hope for the success of Google Glass [10]. For example, surgeons can perform surgery while keeping a literal “eye” on their vital signs [11].
Other than Google Glass, there are other applications of AR in healthcare that I see as up and coming in the next few years. EchoPixel is a company that is leading the mixed-reality landscape in healthcare by being the first company that can render 3D images from 2D scans [12]. The best part? All of this is patient specific. EchoPixel can enable surgeons and doctors to view the anatomy of patients and devise operational/treatment plans accordingly. Additionally, I predict that once enough data is acquired for different physical conditions, AI can be implemented to enhance the experience by automatically designing operational plans and detecting problems in the anatomy. AR in healthcare is only just arriving — the onset of 5G and its speed and connectivity will play a significant role in its success.
2.2 — Remote Robotic Surgery
Another aspect of the 5G revolution that could not have gone undiscussed today is remote robotic surgery. Now, this is not an unheard-of concept, we have robots to do surgeries, and we have physicians that can use robotic surgical tools successfully. However, with the significantly reduced latency period, performing these robotic surgeries away from the actual patient is now a possibility [13]. The robotic surgery market is expected to grow to by $4 billion in the next 5 years — which provides enough incentive for the leading company, Intuitive Surgery, to invest in the remote surgery space [14]. That being said, there is competition in this aspect of the industry with giants; for example, Johnson and Johnson because of their acquisition of Auris Health [14]. Auris Health’s surgical tool is specifically to be focused on the bronchoscopy procedures, while Intuitive Surgery’s robot has wider applications. For this reason, I think Intuitive surgery will lead the remote robotic surgery space in the next 5 years [15].
2.3 — The Internet of Medical Things (IoMT)
When I was researching for this piece, I came across the IoT and IoMT acronym a countless number of times, and it became apparent that this will be a major trend for the next five years. The concept of IoMT is relatively well-known now, but it is about to become a lot more useful because of the breakthrough in connectivity that will arise from 5G implementation [16]. What is IoMT anyway? Deloitte describes it as medical products such as, pregnancy testing kits, imaging technologies, surgical tools, and more, that can be integrated into a network that collects, transmits, and analyzes data. Hence, in layman’s terms, IoMT is the web that connects different medical technologies that can be used to improve almost every aspect of healthcare [17].
A connected medical device is anything that uses the internet to function, transmit data, or access data. Wearables, connected inhaler devices, and insulin and glucose monitors are just some examples of such devices. Now picture this, John is a 30-year-old marketing lead at a company, he has diabetes and his work schedule makes it hard for him to get his meals on time. John also has an Apple watch and an insulin pump that will automatically administer insulin before blood glucose levels become dangerous. John uses his Apple Watch, which is connected to his insulin pump to assess how his diet impacts blood glucose levels and logs his daily meals on a food tracking app. John’s profile is a part of the IoMT in multiple ways due to his connectedness. If John’s data is anonymously stored, along with others like him, there is significant scope for understanding patient behaviour and its impact on health.
This was just one example of how the IoMT is enhanced by connected medical devices. 5G will impact IoMT since it will not only allow for more connectedness between devices, there is a possibility to create massive Machine-Type Communications (mMTC) through smaller sensors (used to monitor patients, compliance, adherence to treatment and more) at a more reasonable scale with fewer resources [18]. The key concept here is that while the “things” are important, the true value comes from real-time data acquisition. With the possibilities laid out, a company that I believe will be at the forefront of using real-time data from IoMT to its benefit is Medtronic [19]. Medtronic is a global medical device company that is heavily focused on using data to provide value-based healthcare — and a recent report identified ways that healthcare can be improved using data in a partnership with Harvard Analytics [20]. Medtronic has already identified that connected devices are an important part of the equation; and with their connected products and data analysis abilities, Medtronic has the most to gain from the IoMT.
The 4thR, in conjunction with the power of 5G, offers multiple ways that existing healthcare technologies can be optimally used to increase patient care, gather valuable data, and allow for the growth of the industry.
3.0 — 3D Printing
I am certain this is not a new concept to most people — we have all heard about amazing ways that 3D printing has been used across multiple industries, including healthcare. However, I wanted to highlight a use-case of 3D printing which I expect to grow in the next 5 years by about 29% (compounded-annual-growth-rate) [21]. Organovo is a pioneer in the 3D printing space, specifically, 3D bioprinting. They have been bioprinting tissues for use in research for a decade, however, the future lies in being able to 3D print entire organs that can be transplanted in patients [22]. While this may sound like science fiction at first, a small start-up company, Prellis Biologics, was recently featured in TechCrunch for successfully bio-printing a miniature sized heart which, when implanted, was able to form connections with the animal’s inherent vasculature [22]. Printing a fully functional and implantable organ does not seem likely in the next five years, but the progress the company has made is important. Not only does it put us one step closer to artificial organs, it certainly adds pressure upon more established companies like Organovo. If there is one thing to say for certain, it is that 3D printing in healthcare is not going anywhere.
4.0 — Conclusion
I have mentioned many companies today that I believe will improve healthcare in the coming five years — but before I sign off, I want to put my money where my mouth is. Figure 2 and Figure 3 describe how much of my investment would go into specific trends and the companies which I believe to be the leaders of specific use-cases of these trends based on current funding availabilities, size of the company, and stock trends.
References
1. Garbade, Michael J. “Clearing the Confusion: AI vs Machine Learning vs Deep Learning Differences.” Medium, Towards Data Science, 14 Sept. 2018, towardsdatascience.com/clearing-the-confusion-ai-vs-machine-learning-vs-deep-learning-differences-fce69b21d5eb.
2. Rees, Victoria. “Optimising Artificial Intelligence in the Pharmaceutical Industry.” European Pharmaceutical Review, 11 Dec. 2019, www.europeanpharmaceuticalreview.com/article/107772/optimising-artificial-intelligence-in-the-pharmaceutical-industry/.
3. “Intelligent Drug Discovery, Powered By AI.” Deloitte Research, Deloitte, 2019, www2.deloitte.com/content/dam/insights/us/articles/32961_intelligent-drug-discovery/DI_Intelligent-Drug-Discovery.pdf.
4. Reed, Jane. “How Natural Language Processing Enhances Sales Productivity in Pharma.” Copyright Clearance Center, 27 Mar. 2019, www.copyright.com/blog/natural-language-processing-sales-productivity-pharma/.
5. Adams, Ben. “Sensyne Health Pairs with Roche for AI Trial Work.” FierceBiotech, 29 Dec. 2019, 11:33, www.fiercebiotech.com/cro/sensyne-health-pairs-roche-for-ai-trial-work.
6. Hutson, Matthew. “AI Protein-Folding Algorithms Solve Structures Faster than Ever.” Nature, 22 July 2019, doi:10.1038/d41586–019–01357–6.
7. Chaudhuri, Tapan K., and Subhankar Paul. “Protein-Misfolding Diseases and Chaperone-Based Therapeutic Approaches.” FEBS Journal, vol. 273, no. 7, 27 Apr. 2006, pp. 1331–1349., doi:10.1111/j.1742–4658.2006.05181.x.
8. Lunden, Ingrid. “Eigen Nabs $37M to Help Banks and Others Parse Huge Documents Using Natural Language and ‘Small Data’.” TechCrunch, TechCrunch, 14 Nov. 2019, techcrunch.com/2019/11/14/eigen-nabs-37m-to-help-banks-and-others-parse-huge-documents-using-natural-language-and-small-data/.
9. Schwab, Klaus. “The Fourth Industrial Revolution: What It Means and How to Respond.” World Economic Forum, 14 Jan. 2016, www.weforum.org/agenda/2016/01/the-fourth-industrial-revolution-what-it-means-and-how-to-respond/.
10. Statt, Nick. “Google Opens Its Latest Google Glass AR Headset for Direct Purchase.” The Verge, The Verge, 4 Feb. 2020, www.theverge.com/2020/2/4/21121472/google-glass-2-enterprise-edition-for-sale-directly-online.
11. “The Impact of Google Glass on Healthcare.” Meditek, 15 Mar. 2017, www.meditek.ca/google-glass-healthcare/.
12. “EchoPixel Introduces 3-D Holographic Intraoperative Software.” DAIC, 27 Sept. 2019, www.dicardiology.com/content/echopixel-introduces-3-d-holographic-intraoperative-software.
13. Frost, Caroline. “5G Is Being Used to Perform Remote Surgery from Thousands of Miles Away, and It Could Transform the Healthcare Industry.” Business Insider, Business Insider, 16 Aug. 2019, www.businessinsider.com/5g-surgery-could-transform-healthcare-industry-2019-8.
14. Landi, Heather. “Report: 5G Has the Potential to Revolutionize Robotic-Assisted Surgery, Improve Availability of Healthcare.” FierceHealthcare, 26 June 2019, www.fiercehealthcare.com/tech/report-5g-has-potential-to-revolutionize-robotic-assisted-surgery-and-improve-availability.
15. Speights, Keith. “Where Will Intuitive Surgical Be in 10 Years?” The Motley Fool, The Motley Fool, 14 Apr. 2019, www.fool.com/investing/2019/04/14/where-will-intuitive-surgical-be-in-10-years.aspx.
16. Research and Markets. “Global Internet of Medical Things Market — Forecasts from 2018 to 2023.” Research and Markets — Market Research Reports — Welcome, Sept. 2018, www.researchandmarkets.com/research/6cmfn3/global_internet?w=4.
17. Haughey, John, et al. “Medtech and the Internet of Medical Things.” Deloitte United States, 12 Sept. 2018, www2.deloitte.com/global/en/pages/life-sciences-and-healthcare/articles/medtech-internet-of-medical-things.html.
18. Looper, Christian de. “5G’s Arrival Is Transforming Tech. Here’s Everything You Need to Know to Keep Up.” Digital Trends, Digital Trends, 2 Feb. 2020, www.digitaltrends.com/mobile/what-is-5g/.
19. Biswas, Urmimala. “IoMT Becomes the Buzzword in New Age MedTech Investing.” Yahoo! Finance, Yahoo!, 13 Sept. 2019, finance.yahoo.com/news/iomt-becomes-buzzword-age-medtech-131301685.html.
20. “Three Keys to Unlocking Data-Driven Health Care.” Medtronic, Harvard Business Review, 2019, www.medtronic.com/content/dam/medtronic-com/global/Corporate/Initiatives/harvard-business-review/downloads/three-keys-data-driven-healthcare_infographic_hbr_av_mi_corpmark.pdf.
21. Modor Intelligence. “3D Printing Market: Growth, Trends, and Forecast (2020–2025).” Market Research — Consulting, Reports, Advisory, Sizing, 2020, www.mordorintelligence.com/industry-reports/3d-printing-market.
22. Shieber, Jonathan. “3D-Printing Organs Moves a Few More Steps Closer to Commercialization.” TechCrunch, TechCrunch, 11 Aug. 2019, techcrunch.com/2019/08/11/3d-printing-organs-moves-a-few-more-steps-closer-to-commercialization/.
