Healthcare in America today is nearly 18% of GDP. The average American spends over $10,000 per year on their healthcare. In other developed countries, with national healthcare systems, it is between 10 to 14% of GDP, spending far less (see 40 surprising healthcare facts). Politically there is a push for universal healthcare – this will be difficult without major changes in healthcare services. The bottom line is healthcare services needs major cost reductions in order to achieve these political and human objectives. Innovation will be the key – Healthcare artificial intelligence (AI). Healthcare AI is coming, soon it will be a luxury to see a doctor. Is this a bad thing?
The quest to create a medical tricorder (a portable, even personal, a device capable of diagnosing and treatment for healthcare issues in humans), has to date been only part of imaginary Sci-Fi movies. Today people scoff at the idea, but its day is coming. Some concerns can include; can AI really be better than a real doctor, can machines really replace humans in human care, and will cost really be lower. Let’s explore some of these concerns of healthcare AI.
Doctors at Moorfields Eye Hospital in London have taken a step into healthcare AI. They trained an algorithm that could recommend the correct treatment approach for more than 50 eye diseases with 94 percent accuracy. They compared that to eye specialists, and the machine didn’t miss one referral, but the eye doctors did. In another example, doctors in China used AI to diagnose polyps on the colon during a colonoscopy. In one arm of the randomized trial, the diagnosis was made by AI plus the gastroenterologist. In another arm, just the specialist made the diagnosis. The AI system significantly increased polyp detection (29% compared to 20%). In other words, AI is potentially better than doctors.
When it comes to our health, especially in matters of life and death, the promise of AI to improve outcomes is very intriguing. While there is still much to overcome to achieve healthcare AI, most notably data privacy concerns and fears of mismanaged care due to machine error and lack of human oversight, there is significant potential. Here is a list by News Forecasters of potential AI advances in healthcare that appear to have the most potential:
- Virtual doctor and nursing assistants – initial diagnosis,
- AI-assisted robotic surgery,
- Image and blood analysis,
- Aid medical patient oversight – treatment protocol,
- Aid in pharmaceutical development,
- Aid in clinical trial judgment and diagnosis,
- Improve gene editing,
- Aid in spotting medical patient trends, and
- Workflow and administrative tasks.
This is not a complete and comprehensive list but it’s a good start, for future healthcare activities performed by AI and non-humans. This list could grow 10 fold over the next 10 years.
Healthcare is emerging as a prominent area for AI investments. And nearly every area across the industry will be impacted by the technology’s rise. Image recognition, for example, is revolutionizing diagnostics. Recently, Google DeepMind’s neural networks matched the accuracy of medical experts in diagnosing 50 sight-threatening eye diseases. Even pharma companies are experimenting with deep learning to design new drugs. For example, Merck partnered with startup Atomwise and GlaxoSmithKline is partnering with Insilico Medicine. In the private market, healthcare AI startups have raised $4.3B across 576 deals since 2013, topping all other industries in AI deal activity. In one of the largest M&A deals in artificial intelligence, Roche Holding acquired Flatiron Health for $1.9B in February 2018. Flatiron uses machine learning to mine patient data. Clearly, healthcare AI is just at the beginning and offers exciting tremendous new opportunities – definitely a growth area for new investments. The key business drivers will have cost reductions and accuracy of healthcare service delivery.
As stated before the only practical way to get universal healthcare in America is via innovation. News Forecasters believes that the first area of innovation, healthcare AI introduced, will be in primary healthcare services. The idea of going and seeing a doctor on an initial visit will become antiquated. Similar to going into a bank to get cash and seeing a bank teller.
Healthcare AI primary services might work something like this. The patient will logon to their healthcare provider’s app to request initial healthcare services. The healthcare AI will ask questions, consulting the patient’s healthcare history database, and suggest other potential diagnostic tests, with tools that came with there healthcare AI toolkit. The healthcare AI toolkit may include capturing the patient’s temperature, heart rate, blood pressure, photographic imagery, body scanning technology, and analytical blood work. From this the healthcare AI may; give medical advice, pharmaceutical prescriptions, request you to speak to an online healthcare professional via video link for further human diagnoses, and then where appropriate refer you to to an actual medical specialist while making an appointment for the patient real-time. Other areas will be in the area of preventative checkups, following similar approaches.
This initial healthcare service provision, will be at a lower cost, provide improved diagnostic and treatment accuracy, and speed of service delivery, which can be critical in healthcare services. No need to decide if you’re going to the doctors or not – just go online. Initially offered as an employee healthcare based service option and to Medicare and Medicaid patients, it becomes a de facto healthcare standard. Government’s may play a role in setting standards, oversight and regulating the legal liability to healthcare AI providers. After primary healthcare services are under our belt, second and third-tier healthcare AIs will then be considered and adopted on a more plug and play basis for specific medical areas.
News Forecasters sees this happening within the next 10 years. There will be an initial resistance to change, issues for sure, but over time it will be accepted and preferred. It won’t be just in America but other countries from the developed and emerging markets will also adopt these standards and participate in the industry. In short, this all could be a very positive development for humankind.
A video presentation of this subject: