Remember those commercials for LifeAlert? “I’ve fallen, and I can’t get up!” Those cheesy, melodramatic ads became a sort of cultural punchline in the 1990s, but they nonetheless underscored a very important need. The commercials represented an earnest attempt at bridging the gap between in-person care and technology-assisted care. The idea was that when someone fell, they could press a button to get immediate care and attention.
Today, health-care providers have much more sophisticated options to help doctors and nurses help patients, or allow families to monitor their loved one’s health. One of those options is AI.
Sure, AI has lately attained a sort of buzzword status, a term that businesspeople can throw into a conversation to seem in the know. But AI can honestly do some pretty amazing things, mostly by enabling health-care practitioners to improve the quality of care that they provide and interact more meaningfully and effectively with their patients.
Patients want better ways to interact with their health-care providers, but so far health-care organizations have been falling short in providing those opportunities for interaction. AI can help health-care providers offer better care, both in-person and remote.
Currently, AI can perform a variety of menial but important tasks, which frees up health-care practitioners to focus on their jobs, allowing them to focus on the patient.
For instance, Duke University reports that 700,000 infections occur per year in health-care settings, simply because surgical devices aren’t properly cleaned. This is partly because health-care staff are already busy, and may rush the job. Another factor is that some new medical tools are so advanced that staff just don’t know how to clean them. An AI-assisted surgical instrument can learn to detect rust, blood, bone, and other biological tissue on surgical tools to ensure they are properly cleaned.
Other AI technology can take patient vital signs as they sleep, allowing them to get more rest. AI uses deep learning to identify when patients are in their deepest sleep cycle and takes their vital signs at this point. This means that patients aren’t disturbed and can continue resting. This also helps health-care workers get more rest themselves, since they don’t have to take the vitals. This is an increasingly important benefit as burnout and overwork plague health-care practitioners.
Another area in which AI can help is in reading doctors’ notoriously bad handwriting. It’s an embarrassing open secret, but bad doctor handwriting causes 1.5 million injuries every year. Busy doctors scribble illegible prescriptions and clinicians will make dangerously wrong interpretations through no fault of their own. Machine learning can learn to recognize and predict doctor handwriting over time, turning those handwritten notes into legible text.
A lot of health care happens outside the hospital. Especially for chronically ill patients, or older patients, it just doesn’t make sense to stay in the hospital indefinitely. They want to be home, and you want to have bed space to help more patients. Also, the longer patients stay in the hospital, the more likely they’ll be exposed to infections.
But of course, care doesn’t end when the patient leaves. Patients still need to be confident in their own safety, both in managing their conditions and in getting a fast response should their condition deteriorate rapidly.
That can be tricky to balance. Nobody likes to feel watched 24/7. One study looking at informal caregivers in the UK noted that “patients…are reluctant to have external care provision in the home… preferring instead to be self-sufficient.” Plus, that kind of care is expensive if the patient is hiring external help, or time-consuming if the patient is relying on an unpaid family member to assist them. That’s where AI can come in.
Wearable technology sends patient vitals to physicians in real-time. AI can help physicians deal with this influx of data, helping spot trends and make decisions, as well as alerting physicians to dangerous drops or spikes in vitals that indicate immediate care is needed – kind of like an improved, more high-tech version of the “I’ve fallen and I can’t get up” button.
AI can help beyond monitoring and treatment. Those same wearables can even help with the early detection and prevention of cardiac disease. When older patients coming in for their annual check-ups wear monitoring devices, AI can help detect abnormal data patterns before patients even come in with their concerns.
This kind of AI use can improve patient outcomes as well as reduce costs for patients and health-care providers. If you can outsource as much care to AI while also potentially improving the quality of care patients receive, it’s a win-win.
AI is a useful tool, not a panacea
The truth is that AI in healthcare is now only helpful in simple ways, but it’s still extremely valuable. While AI can’t (yet) work miracles, it can improve patient care, reduce the burden on overworked practitioners, and minimize administrative costs in hospitals.