By Julie Campbell, Vice President
We know you’ve heard the buzz around artificial intelligence (AI) in healthcare. After all, it has been utilized in healthcare since the 1970’s. AI is easily recognized in the context of Siri, Google Home, or self-driving cars, but the evolution of AI in healthcare is also rapidly expanding. There are significant high value use cases, and most of the applications seen today involve pattern recognition, prediction, and natural language understanding. In a CB Insights article on the 100 top AI startups of 2022, healthcare was the most represented category. The low hanging fruit has been in administrative activities, such as scheduling, prior authorization, claims management, supply chain management, and medical billing and coding.
However, we have barely tapped the surface of its potential- and it remains largely untapped when it comes to clinical optimization. From the use of AI-assisted robotic surgery, precision medicine, to clinical diagnosis, AI has made its presence known in healthcare space. Adoption in these use cases can go a long way toward addressing the significant barriers to value-based care and clinician shortages. By 2025, some predict that the U.S. will be facing shortages of over 600,000 healthcare workers, resulting in a need to help reduce these losses. AI, while not a replacement for healthcare workers, enables professionals to be more efficient, operate to the top of their license, and make more informed decisions. This leads to a decrease in burnout, increase in employee retention, and higher quality care.
AI in clinical applications is still growing and in its early phases. While the use cases for AI are practically endless – and include diagnosis and treatment recommendations, patient engagement and adherence, administrative activities, and a lot more – the reality is that healthcare has been much slower to adopt clinical applications of AI. In the near future, we expect to see an acceleration of AI utilized in clinical operations. We will explore some of the numerous use cases, and many are in early stages of development and utilization today.
Optimize how effectively care is delivered
AI enables healthcare professionals to diagnose diseases faster, earlier, and more accurately. This is done using big data and insights at the right place and time, as well as through AI-enhanced medical devices. For example, next generation blood tests – including Thrive Earlier Detection and Freenome – are changing the way blood is tested through the use of AI. These platforms can help healthcare professionals detect cancer earlier, giving patients a higher chance of survival. And Beth Israel Deaconess developed AI-enhanced microscopes for blood tests, resulting in a faster diagnosis. AI use in healthcare is also improving the accuracy of diagnostics through machine learning to assist pathologists in making more accurate diagnosis- PathAI. And other companies – like NanoxAI (which acquired Zebra Medical) – are using AI to power radiology in order to increase accuracy.
AI can go beyond diagnosis – it can enable active treatment. Companies are introducing AI-assisted robotic surgery to improve care delivery. The Da Vinci platform by Intuitive has been enabling precise surgical care for many years, but it continues to get smarter; it now makes recommendations for future surgeries, while other organizations like Carnegie Mellon are implementing a miniature mobile robot that facilitates therapy on the heart.
Improve care access and engagement for patients and consumers
During the pandemic, there has been a proliferation of organizations implementing intelligent symptom trackers that use natural language processing to understand a patient’s symptom description. Then the platform guides the patients through symptoms to choose from and produces care recommendations after information is collected. Some advanced trackers will also immediately recommend emergency care or direct to 911 if conditions seem severe such as conditions of a stroke or heart attack. Healthcare organizations can improve optimal level of care delivery through tools like intelligent chatbots – including Babylon Health, Buoy Health, and Gyant. In addition to the intelligent symptom tools above, AI is becoming even more prevalent in tools that consumers are using outside of the hospital setting. AI enables the optimal patient-physician match through companies like Health in Her Hue and Spring Health.
AI is also present in care even when a physician isn’t present, which supports patient engagement and outpatient accountability. Woebot Health and Real are personal mental health allies that help patients manage their mental health in real time with intelligent chat bots, virtual games, and smart communities. Digital therapeutic musculoskeletal companies – Kaia Health, Sword Health and Hinge Health – are using computer vision to generated physical therapist-grade automated feedback coaches. AI can also be utilized to help reverse chronic conditions like diabetes. Solutions, like Virta Health and Twin Health, can provide personalized and precise nutrition, sleep, activity and breathing guidance to each member along with medication management guidance.
Utilize clinical resources effectively
AI can be used to be able to prioritize use of limited healthcare resources. It can predict and triage resources needs by combining big data and AI to predict clinical, financial, and operational risk. These platforms can predict things from who might get sick to what is driving up a hospital’s healthcare costs. For example, TQIntelligence is using voice biomarkers to identify patients at highest risk for mental health crises, which enables physicians to identify how to prioritize treatment plans given the physician shortages.
AI can even predict when to intervene during the patient’s journey (i.e., sepsis, readmission). CloudMedX does this by managing patient data, clinical history, and payment information to intervene at critical junctures. AI can impact patient flow optimization in the health system setting. Platforms like Qventus prioritize patient illness/injury, tracks hospital waiting times and can even chart the fastest ambulance routes. By implementing AI solutions, organizations can predict staffing needs based on historical data, call offs and trends in demand. And AI solutions make case management more efficient through optimal discharge planning, transportation needs, and easing post-acute transfer.
Improve provider performance
AI can be utilized as a training tool for physicians in order to maximize their skills. Using AR/VR and 3D printed models, physicians can practice life-saving skills in a controlled and safe environment, allowing them to learn and improve before needing to perform these skills on a patient. Whether you are training for a generic procedure (OssoVR, Level Ex) or training for a specific complex patient surgery (D.A.S.H. Orlando), these solutions can be utilized for surgeons to practice procedures pre-op. Clinicians can also be trained on bedside manner in a telehealth environment, with AI-enabled facial recognition to monitor performance.
Develop more effective medicines faster and at a lower cost
AI potential in drug discovery is significant. Healx is the next-generation AI platform built to find novel connections between drugs and diseases. By combining frontier AI technology with deep drug discovery and development expertise, Healx can accelerate the pace, increase the scale, and improve the chance of success of rare disease treatment. Recursion Pharmaceuticals has built a drug discovery platform which lets the company repurpose drugs and compounds that already exist. Being able to repurpose existing drugs will save on costs opposed to creating new drugs to affect new issues.
The future of AI
AI has established its place in the healthcare industry and will only continue to grow. The implementation of AI in healthcare settings can lead to reduced burnout, higher accuracy and efficiency, and the development of new technologies and medicine that will revolutionize healthcare. AI will never replace healthcare workers, nor do we want it to. AI should be utilized to lighten the load from healthcare workers and optimize their decision making so they can focus on what is most important- the patient’s care. AI has advanced from a simple question asked to your Amazon Alexa, to saving you an hour to schedule your appointment to get new life saving medicine produce. AI is the future of healthcare and embracing it will advance the wellbeing and health of healthcare staff and patients.