Artificial Intelligence in Healthcare: 2017 accomplishments and what’s next in 2018
2017 has seen the Big Giants enter the healthcare space and outstanding start-up innovations. Yet integration within clinical practice has been slow. Will 2018 deliver on 2017’s promise?
I read at the start of last year that 2017 would be the “Year of Artificial Intelligence”. In our day-to-day lives, this is true. AI is now all around us and is changing everything we know! Moreover, I believe it will soon redesign the healthcare industry…
When we think about AI, one may think initially of robots! Yet, the AI we are talking about today is the invisible “helping” hand behind improving our life experience. The Tweets and posts in our Facebook feeds, our search results, our Amazon Echo and Google home experience; it is all driven by AI learning machines. These AI technologies analyze collected data about us to predict our preferences and future needs for a better, personalized experience.
AI is just getting started. Soon we may be greeted by name next time we enter a store and robots will shop for our groceries. Programs such as Westworld, where AI-robots are used to create alternative realities (a theme park where humans play out their fantasies with robots) will no longer seem such far-fetched ideas.
The start of the AI journey is here now… and it has significant relevance for healthcare. Particularly, its ability to deliver the promise of personalized medicine.
Yet, in the tightly controlled healthcare sector, we are still not seeing AI make the life-saving difference. This article “HEALTH CARE IS HEMORRHAGING DATA. AI IS HERE TO HELP” in the Wired, does a very good job of summarizing some of our healthcare accomplishments of 2017.
The big giants got lost in translation
For me personally 2017 was the year the big giants, such as IBM, Google, Apple and Microsoft all staged AI as a “Game Changer” in health care. Yet have much to prove.
However, those companies who understood that AI is a “tool” to improve the health service experience did some interesting changes. Led by some really innovative start-ups, these may well pave the way for future healthcare transformations!
So what do I mean?
Healthcare is a specialized sector providing a vital multi-complex line of services. Different players such as physicians, practitioners, patients, payers, pharmaceutical companies and more are all involved in the provision of these services. With so many different stakeholders, healthcare requires a comprehensive design experience that meets each audience’s needs – one size does not fit all.
In this respect, being “Big” does not necessarily mean success when delivering the AI healthcare promise. I think IBM’s Watson performance is a good example of this approach.
Watson became an umbrella brand for AI promising to change the world. The idea was to build an intelligent machine that could be used to solve problems in a wide range of industries, including retail, finance, healthcare and more…
In healthcare, Watson set out to help practitioners better diagnose and treat cancer. For that, IBM signed a strategic collaboration with the world's leading cancer centre of excellence in Houston – MD Anderson. However, after $62 million dollars spent over 4 years, MD Anderson walked away from the Watson Big Data project. Though the vision is powerful, in my opinion, with no clear experience or service design aiming to answer doctors’ specific needs, the idea was not able to be leveraged into practical implementation.
I am now looking forward to seeing how the other AI-healthcare projects from the other big giants fair. There are many promising projects, let’s await the results.
In terms of small enterprises – there has been some great developments in 2017. I recommend you read the Wired article above.
Here in Israel, I have had a front-line view to some outstanding technologies and AI-driven service design approaches. AI-algorithms are assisting radiologists to identify visual abnormalities in medical scans, spotting symptoms of disease by analyzing your voice and predicting the probability of colon cancer development. Each of these start-ups are looking for new ways to approach and improve the diagnosis and treatment of current healthcare problems.
In my opinion, the AI solutions that will succeed in the early years will be those AI-prediction algorithms that take advantage of the overwhelming amount of existing clinical and genetic data now available on one hand, and have a deep understanding of the patient design experience on the other. In other words, AI solutions that are both practical and also bring previously unavailable health data into the clinical setting. Allowing doctors to quickly and easily make informed decisions to personalize treatment, while saving time and money in the process.
AI in 2018: Think big yet personalized
This is exactly the approach we are taking at Taliaz.
Our AI-prediction algorithm, Predictix, is an easy-to-use service for Family doctors/ GPs to help them better prescribe the right medication for their patients sooner. It brings personalized medicine into the doctor’s office: healthcare innovation that fits easily within GPs daily activities.
We are thinking big but being practical. And it is easy to lose sight of this fact when we see and read so much hype about AI and its potential in healthcare. Yet, we must separate the hype from what’s really happening.
For example, we have barriers for regulatory approval and user adoption.
Today, machine learning powers more and more medical device software. And because it is always learning and improving, it is constantly changing products on the fly. This makes regulating it extremely tricky.
I am interested to see what the FDA’s new unit dedicated strictly to digital health will do. The agency is piloting a new course—that certifies trusted companies with good track records, as opposed to individual software packages.
Rather than reviewing each line of code or medical device on its own merits for each of its intended uses, they are looking to flip that framework on its head. Instead, a model where trusted companies with demonstrated histories of excellence would be AI certified companies.
From the user adoption perspective, as within any change, there will be reluctance. The integration of AI in the clinical setting will face difficulties. Yet, I also believe that doctors and care staff understand the potential benefits from AI for both the prevention and treatment of disease.
AI won’t replace doctors anytime soon
In this context, I expect to see rapid integration of AI into the everyday fabric of healthcare in 2018. However, not as a disruptive force that totally changes healthcare as we know it, but a gradual, incremental change that supports our doctors. Where small, targeted AI improvements drive improved care, operational efficiencies and lower costs.
At the end of the day, care is about people and in my opinion, doctors. The role of listening and being there for a patient cannot be replaced by AI. However, we can empower our doctors, through AI, to do an even better job. To better diagnose, treat and empower patients to improve their care management, whether they are in hospital or at home.
We can make AI not a force to be scared of, but a helping hand that makes our lives easier and healthier.
I look forward to see what 2018 has in store!