Enabling better healthcare with artificial intelligence

August 28, 2017



AI could fundamentally change the nature of patient care delivery. Are consumers and clinicians prepared for this much-anticipated innovation?

Today’s unprecedented increase in the volume of patient healthcare data has left the industry struggling to put that data to practical use. Artificial intelligence (AI), with its capability to draw “intelligent” inferences based on vast amounts of raw data, may hold the solution. Smart devices and robotics are already making tentative inroads to the healthcare marketplace. Follow the money, and you’ll see big bets on healthcare AI across the globe: 63% of healthcare executives worldwide already actively invest in AI technologies, and 74% say they are planning to do so.

The data that fuels AI is at the heart of today’s healthcare delivery, and as more information goes online, it continues to proliferate. Between 2010 and 2015, the amount of stored patient data increased 700%, 91% of which is unstructured. Unstructured data is the information that resides outside of organized databases such as electronic health records and lab reports. If we become capable of tapping the potential of that data, we could make patient care more efficient and cost-effective than ever before.

artificial intelligence in healthcare

But unstructured data—such as photos, videos, recorded dialogue, physician notes, sensor data, and genomic information—is difficult to organize using traditional computational algorithms. Today, most of that data languishes on servers and clouds, with payers, providers, and patients unable to realize its potential. But AI is changing that—quickly.

As processors get faster and faster, we are getting better and better at automatically parsing through, evaluating, and categorizing the enormous amount of patient data available to us. This is enabling us to begin to diagnose, predict, and even—with the help of robotics—treat human ailments. As we continue to enhance our ability to turn patient data into intelligence about that patient, we will increasingly challenge our assumptions about the limitations of healthcare delivery.

PwC’s Global Artificial Intelligence Study, which analyzed AI’s potential impact on each industry, found that healthcare (along with retail and financial services) is poised to reap some of the biggest gains from AI in the form of improved productivity, enhanced product quality, and increased consumption. AI stands to benefit healthcare diagnostics most by helping detect small variations within patients’ health data and comparing variations among similar patients; identifying potential pandemics early and tracking the incidence of diseases to help prevent and contain their spread; and enhancing imaging diagnostics in radiology and pathology.

Intelligent healthcare is already making inroads

Although not yet widespread, some AI-powered, consumer-operated diagnostic and monitoring devices are already on the market, including a smartphone device that captures digital images of the ear canal and sends them to a physician for diagnosis of possible infection. Smartphones from leading manufacturers are also incorporating built-in heart rate monitors. And some pioneering companies are offering online services they say can evaluate digital photos of rashes, moles, and skin conditions and then connect consumers with physicians for consultation.

Some harbingers of AI’s future applications dramatically expand these capabilities. For example, one prestigious industry competition offered millions of dollars to any team that develops a Star Trek-esque, consumer-friendly “tricorder” that is able to diagnose 13 health conditions and measure five vital signs without the intervention of healthcare professionals. Two teams were recently awarded prizes for their contest entries. Their devices drew on data created by patient histories, sensors, image processing, and biomarker testing to gauge patient vital signs–such as oxygen saturation, respiratory rate, and blood pressure–as well as health conditions, including sleep apnea, COPD, and pneumonia.

The implications of a device like the tricorder should not be underestimated. Although far from the independent thinking that remains the elusive goal of “true AI,” these technologies are able to independently perform tasks that have formerly been confined to the realm of the healthcare professional.

Self-serve, consumer-friendly diagnostic tools could replace the routine work of primary care physicians, freeing them up for more complex patient care. Care providers would just have to confirm the results indicated by a patient’s device and follow up with treatment plans, negating the need for an office visit. According to HRI’s 2015 clinician survey, 42% of US doctors say they are willing to prescribe medications based on the results of consumer-operated diagnostic technologies. Widespread use of these technologies could mean big cost savings. Researchers developing a strep throat test for consumer use estimate that the device has the potential to eliminate 780,000 doctor visits a year, resulting in $94 million in savings annually.

Consumers are ready

But the question of whether a majority of patients would accept the replacement a physician’s personal care with devices they operate themselves remains speculative. There are some early positive signs. Nearly 59% of US consumers in our HRI survey said they would be very or somewhat likely to consider using an at-home test for strep throat. Nearly 55% say the same about minor skin conditions, and 47% would consider using a phone attachment to diagnose an ear infection.

Other parts of the globe are even more enthusiastic. We found in a recent survey of 12,000 people across 12 countries that most consumers are willing to substitute the care of human clinicians with the use of AI and robotics technologies. From fitness advice to minor surgery, consumers across demographic groups on a global scale are willing to consider non-traditional options for managing and treating their health.

AI in healthcare

For example, 94% of survey respondents in Nigeria, 85% in Turkey, 41% in Germany, and 39% in the UK are willing to talk to and interact with a device, platform, or AI-guided robot that can answer health questions, perform tests, make diagnoses based on those tests, and recommend and administer treatment. Specifically, consumers are most willing to receive four types of services from AI-guided technology: heart monitoring, heartbeat rhythm check, fitness and health advice, and blood sampling.

artificial intelligence in healthcare

Our global health survey also found that many respondents would consent to minor surgical procedures performed by a robot if studies showed that the robot could out-perform a human surgeon. Respondents in Nigeria (73%), Turkey (66%), and South Africa (62%) were the most willing; the least willing was the UK, which still garnered a healthy 36% agreement rate.

Indicators are that the AI healthcare market is poised for dramatic growth. Frost & Sullivan predicts that, worldwide, the AI market for healthcare will increase by 40% between 2014 and 2021, growing from $633.8 million to $6.66 billion. AI is also heating up the global deals market. In 2011, there were fewer than ten deals related to healthcare AI. That number climbed to almost 70 deals in 2016.

A potentially huge payoff awaits

The current advent of smartphones, wearables, and EHRs has resulted in a tsunami of data, the extent to which was not fully anticipated. If properly organized, manipulated, and executed, this raw data has the potential to transform the healthcare industry. But if left in disarray, such data can do little more than occupy server space.

Established and startup companies are tackling the challenge of putting an enormous amount of patient data to work for them. This includes experimentation with new technologies and market models, the formation of new business model solutions, and a significant number of deals as AI companies consolidate their resources. This evolution of AI in healthcare can be thought of in four stages.

But while the potential is there, there are still many unknowns at this early stage. The integration of AI-enabled technologies in the healthcare marketplace depends largely on the widespread implementation of those technologies. Just because they exist does not mean they will be offered to every consumer. Clinicians and payers will have to make hefty investments in AI-enabled devices, which will require their conviction that the new tech will in fact improve care and reduce costs long-term. From EHRs to streamlined coding, providers and payers are currently under an enormous amount of pressure to balance multiple IT investments. Convincing them that costly AI-enabled tech will be worth their investment will require compelling proof.

The road toward the vision of a self-serve, consumer-directed, personalized healthcare delivery model powered by artificial intelligence will likely be repeatedly disrupted by unexpected stops and starts. But, to more and more clinicians and technologists on the leading edge of healthcare IT development, it’s just a matter of time.


Chris Curran

Principal and Chief Technologist, PwC US Tel: +1 (214) 754 5055 Email

Anand Rao

Global Artificial Intelligence Lead, PwC US Tel: +1 (617) 530 4691 Email