Providers, pharmaceutical and medical device companies, payors, and consumers are all preparing for deeper integration of artificial intelligence (AI) and machine learning into their processes.
Artificial Intelligence is a buzzword across the healthcare industry, and for a good reason. With an explosion of data in our society, 2.5 quintillion bytes of data generated each day, there is an increasing need to move toward parsing through what's actually important and utilizing the data effectively. In healthcare, data analysis can help decrease costs, improve care, streamline processes, improve patient outcomes, and predict population health trends.
How are healthcare organizations using AI?
From hospital care to clinical research, drug development and insurance, AI applications are revolutionizing how the health industry works to reduce spending and improve patient outcomes. Here are a few ways healthcare organizations are using AI:
Detecting and diagnosing - AI systems can analyze data faster than humans, which may make them more adept at processing multiple patient symptoms and determining medical diagnoses. An individual patient has hundreds of thousands of healthcare data points, if not millions in their electronic healthcare record (EHR). AI can sift through all of these disparate data points, identify patterns, and alert healthcare providers of any potential patient concerns or medication side effects or interactions.
Precision surgeries - AI-enabled robots have begun assisting with surgical procedures. In a surgical setting, a robotic system can be controlled by a surgeon, whose hand movements are converted into smaller, more precise movements that are performed by a set of “robot hands.” These hands stabilize potential tremors in the surgeon’s movements and the AI robot is also able to "learn" from each surgery, providing actionable feedback to surgeons on how to improve.
Population health - Using AI, healthcare professionals can get a complete picture of a patients’ health history, quickly compare it against that of patients with similar symptoms, and make more effective care decisions.
Wearables - Consumers are increasingly adopting wearable technologies like FitBits, Apple Watches, heart monitoring armbands, etc. With a growing percentage of the population utilizing these wearable technologies, clinicians are starting to use this data to track medical and health-related information from their patients, otherwise known as remote patient monitoring. This helps physicians monitor and treat chronic illnesses such as diabetes, cardiovascular disease, and asthma. AI is being used to help analyze a massive amount of patient data in real time for a more holistic approach to patient health.
3 emerging AI uses in healthcare
1. Drug research and development
Biotechnology and pharmaceutical companies can benefit from AI technologies in a number of ways:
Drug development and discovery - The pharmaceutical research and design process generates more data than ever. It can be incredibly difficult for drug researchers to wrangle large data sets, work across data silos, and extract meaningful data points. Artificial intelligence is used in various ways by those who produce drugs to develop better diagnoses or biomarkers, to identify the objectives of the drug, and to design new medicines. For example, in the pre-clinical stage (drug discovery), AI can help researchers identify drug candidate molecules that are both safe and effective. Previously, researchers had to sift through hundreds or thousands of compounds that must be screened and optimized against different properties (i.e. potency versus toxicity). AI can help researchers predict the most optimal compounds by picking up on patterns during early-stage testing.
Clinical trials - For biotech and pharmaceutical companies, it's becoming more important to invest heavily in commercial or "all-payor" claims, data and advanced analytics to better predict which clinical trial sites work best with patients. Artificial intelligence technology can help researchers apply predictive analytics to identify candidates for clinical trials through innovative channels such as social media and doctor visits. AI functionality can also help researchers with real-time remote monitoring and analysis of clinical data, which frees up their time to focus on the clinical trial's design and recruitment.
Identify Key Opinion Leaders (KOLs) - Now, more than ever, pharmaceutical companies need to lean into the power of big data and AI to identify and target individuals who possess the greatest influence over their respective therapeutic area, Key Opinion Leaders (KOLs). These leaders are particularly important during the late stages of drug development as consultants for marketing messaging and go-to-market strategy. Because KOLs have demonstrated expertise in specific fields, their perspectives are trusted and sought out at all phases of new drug development and distribution.
2. Image Analysis
Radiology: The first imperative for any radiologist is to prioritize the patients most in need of treatment, which often means sifting through piles of charts and data. However, this can become time-consuming; a high volume of patients and scans may delay any necessary treatments. AI technologies are working toward bridging this gap.
Artificial intelligence in the medical imaging market is estimated to rise from $21.48 billion in 2018 to a projected value of $264.85 billion by 2026, according to Data Bridge Market Research’s April 2019 report. Ultimately, AI has the potential to revolutionize the medical imaging industry by sifting through mountains of scans quickly and offering providers and patients life-changing insights into a variety of diseases, injuries, and conditions that may be hard to detect without the supplemental technology.
So, what are some imaging-related AI technologies to keep your eye on? First and foremost, Google’s DeepMind AI technology can read 3D retinal OCT scans and diagnose ophthalmic conditions with 99 percent accuracy, alerting the technician to which patients require the most urgent care. ICAD’s “ProFound AI” solution is revolutionizing digital breast tomosynthesis (DBT) by allowing radiologists to view each breast tissue layer independently, which helps detect cancer earlier (by 8 percent) and reduces radiologists’ time spent reading each breast scan by more than 50 percent on average.
White eye detection: Another interesting AI-powered invention comes in the form of an iPhone app. CRADLE was created by researchers at Baylor University in Waco, Texas. It uses the camera flash to detect a retinal reflection or glow, otherwise known as "white eye," in the eyes of infants, which can be a sign of several serious eye diseases such as retinoblastoma, pediatric cataracts, and Coats' disease. From there, the app uses AI to scan through thousands of photos to detect white eye earlier than was previously ever possible.
Skin cancer screening: In the future, identifying skin cancer could be as simple as taking a photo with your iPhone. Some AI technologies have the ability to scan images of cutaneous lesions and identify which individuals are at highest risk of cancer. This type of screening could increase the number of dermatology referrals, streamline patient visits, and prioritize limited resources for patients with the highest risk of developing cancer.
Facial recognition: In February 2019, Boston-based FDNA Inc. developed Face2Gene: an AI technology that can help diagnose genetic disorders in children, using a simple photo. This facial recognition tool scans a baby's face, sifts through thousands of photographic comparisons to known genetic conditions, and then helps geneticists identify a possible diagnosis. This technology is only made possible through artificial intelligence and its ability to cover immense amounts of data and narrow down the options at lightening speeds.
With care models switching from fee-for-service to value-based, patient satisfaction is more important than ever. That's why many healthcare providers and organizations are implementing chatbots on their websites to offer personalized and helpful online experiences for patients. These chatbots use artificial intelligence to process a patient's appointment inquiry, and then artificially respond like a "real" person by replicating patterns of human interactions.
AI-powered mobile applications are also becoming more mainstream. Virtual assistants like WoeBot (a mental health AI tool), Forsky (a calorie-tracking and healthy eating AI tool), and HealthTap (online doctors and consultations) all help patients manage their healthcare from the privacy of their own phones.
Visit us at HLTH 2019!
We will be exhibiting in our brand-new booth (#555)! Interested in learning more about Definitive Healthcare or signing up for a product demo? Contact us today to schedule a meeting:
ABOUT THE AUTHOR
Tory Waldron is a communications professional with a lifelong passion for writing and editing. Before joining Definitive Healthcare, she spent three years at a PR agency working with various B2B ...