Artificial intelligence (AI) propelled by increasing availability of data and analytics is creating a revolution in the way technology works in solving complex problems. The fact that it utilizes, both structured and unstructured data to deliver powerful, conclusive result makes it highly sought after in areas of healthcare, entertainment, finance, transportation and more.
Thanks to AI, the voluminous data which was previously untapped has now been unplugged. Coupled with predictive analysis, through AI massive amounts of data have been scrubbed to produce results that have made a paradigm shift in the way healthcare operates for all – providers, patients and professionals.
What is AI actually and how does it work in healthcare?
Defined as “the stethoscope of the 21st century”, AI simply refers to the capability of a program or a machine to simulate intelligent human behavior like learning, decision making, and problem-solving. In other words – there are algorithms running on machines or computers which are programmed to follow a certain path, pattern or equation for a given input, which then produces the desired output.
With this technology at hand, AI is fast gaining its ground in healthcare. Whether it is diagnosis, labs, radiology, appointments, or procedures, the healthcare industry is rife with data. AI helps interpret these humongous data quickly and accurately, revealing patterns and predictions that aid physicians in many ways – detecting diseases early, design treatment plans accordingly and enhance overall patient health and safety. It is forecasted that these algorithms have the ability to synthesize complex data to design therapies based on an individual’s unique genetic makeup.
“AI not only compiles information but also translates it in usable intelligence. For example, it might crunch millions of EMR’s, journal articles and cancer registry entries to predict the optimal treatment for a patient with a rare form of breast cancer. Recommendations would be personalized based on the person’s genome, health history and response to past treatments. (Such programs are currently in the works at the cancer centers of Cedars-Sinai Medical Center and Memorial Sloan Kettering.)” – According to Ai and Healthcare article by Anne Bruce and Dolly Hinshaw.
How AI complements Healthcare?
AI combined with predictive analysis has helped change the landscape of disease prevention and treatment – providing the right information at the right time to the right person. With improved image analytics, concrete clinical and diagnostic decision-making, AI has been highly beneficial for the treatment of chronic diseases like cancer, neurology, and cardiology. AI-based systems are also demonstrating the ability to detect blood infections and colon cancer.
A striking example of AI in healthcare – Diabetic retinopathy is the most common cause of vision loss among more than 30 million Americans living with diabetes and the leading cause of vision impairment and blindness among working-age adults. In early April, FDA approved the first medical device that combined a special camera and artificial intelligence to detect a mild level of diabetic retinopathy in adults who have diabetes. Once AI detects mild retinopathy, primary care physicians will refer patients to a specialist. This early detection will prevent serious vision problems in the future.
AI’s inclusion in healthcare is vastly being discussed and one of the thought that plagues everyone’s mind is – Physicians versus Bots – Will robots replace physicians in the near future? Still a debatable issue with millions of questions – will I want to be operated by a robot vs a physician? In the event of an incorrect incision, what will the robot do? Is Gen X stopping the progress by introducing emotionless bots?
To answer this, we need to throw light on a physicians’ role – one which is a combination of providing care (mental, emotional and physical) along with cure, and this powerful role can never be replaced by machines, but yes, AI can certainly assist physicians in making error-free predictions, taking concrete diagnostic decisions and carrying out procedures with precision and accuracy.
What does a successful AI strategy in healthcare look like?
AI and big data companionship
The amount of data, to name a few – vitals, lab results, electrocardiograms, medical images, and biopsies —is overwhelming by itself. To add to this there are medical claims, clinical trials, prescriptions, billing and more. Luckily, with the help of AI, companies can assimilate and make sense of all this structured and unstructured data like never before.
Some companies are evaluating DNA in order to diagnose illnesses like a concussion, lung function of those suffering from chronic respiratory diseases, blood pressure, hemoglobin levels and even evaluate coughs, and other diseases– all by integrating smartphone apps with AI.
A few benefits of AI in Healthcare-
- Data analytics along with AI are being used by some hospitals to spot typical patterns in lab data which can improve patient care.
- With AI, closer monitoring of intensive care unit (ICU) telemetry data can give doctors and nurses a heads-up on their patient’s condition.
- Research on brain tumors and other cancers is progressing due to AI and big data approaches.
- With AI, Sepsis detection rates have improved drastically, leading to fewer deaths and lesser financial burden on the hospital.
- It is predicted that in some cases, Artificial intelligence will enable the next generation of radiology tools to replace the need for tissue samples.
AI and wearable technology
Wearables started with a bang – it was a new found way of tracking activity to gauge one’s fitness and recently it has taken an even bigger turn – health apps – to track all the vital health metrics for an individual.
Use of health apps, on smartphones, to track health metrics such as steps and heart rate, more than doubled in two years, from 16 percent in 2014 to 33 percent in 2016, according to the Accenture survey.
The wearables powered by AI are providing a two-fold benefit – not only are they becoming increasingly compact, they are also utilizing the enormous data points in a human body to configure a comprehensive health status-enhancing overall health management by an individual and decreasing cost for the healthcare organization.
One of the greatest impacts of AI coupled with wearables is in the field of clinical trials. By capturing data using wearables, smartphone apps and sensors transmit it to a secured cloud-based structure and then utilize AI to analyze data and quantify aspects like medication efficacy or dose response, these AI capabilities address critical aspects of a clinical trial.
If done correctly, clinical trials decrease operational costs and hold up to the promise of delivering a new treatment or a new drug development.
AI and EMR
In the world of EHR, unstructured data (medical images, lab reports, email communications, clinical notes, patient billing) remain vastly untapped. Some of this data cannot be easily assimilated by physicians to make informed decisions and provide quality care. AI can hence play a pivotal role in harnessing these unused data to redesign physician workflow to better coordinate patient care, treatment and follow up.
Ways in which AI could enhance EHR
- Users spend the majority of their time on clinical documentation, and order entry, EHR developers are now using AI to create more intuitive interfaces and automate some of the routine processes that are time-consuming.
- Clinical decision support (CDS) tools that provide functionalities like population health management, value-based care, and personalized medicine are in high demand. Driven by growth in AI and natural language processing (NLP) techniques, more and more healthcare providers are anxious to turn their documentation-based EHR systems into clinical support and analytics platforms that enhance and supplement the clinician’s decision-making skills.
- NLP, a subset of AI is also helping EHR to improve clinical documentation process through voice recognition, dictation, and video analytics.
- Artificial intelligence may also help to process routine requests from the inbox, like medication refills and result in notifications. It may also help to prioritize tasks that require the clinician’s attention.
- Electronic health record data can help to identify infection patterns and highlight patients at risk before they begin to show symptoms.
The revenue cycle is another area ripe for machine learning, according to Stuart Hanson, senior vice president of Change Healthcare.
Hanson cited two examples: the ability to predict what is relevant for a particular patient and deliver smart messaging, such as wellness and prevention tips and price transparency, as well as the opportunity to drive down costs associated with useless billing by better understanding how patients interact with various types of payment statements.
“There’s clear ROI in the revenue cycle for physicians and hospitals,” Hanson said.
One hospital in Silicon Valley has used AI to capture data from health records, assessing them for risk factors and then combining this data with that provided from real-time tracking. The result could predict which patients were likely to suffer falling over and, in turn, meant they reduced falls for patients by 39%.
EHR’s are a goldmine of data. Leveraging machine learning and AI tools to drive these data and analytics can bring about a powerful change in early detection and diagnosis of a disease, create faster, more accurate alerts for healthcare providers and result in preventive care – saving complexity of a disease in future for a patient and reducing overall healthcare costs.
In a recent move, Fitbit has partnered with Google to be more deeply involved in the healthcare sector. The fitness tracker maker announced that it would use Google’s recently announced health data standards for apps, known as the Google Healthcare API, to connect its wearable devices to the electronic medical records systems used by doctors and hospitals. The aim eventually is to allow doctors to get health data straight from Fitbit on their patients’ wrists.