Chronic diseases such as asthma, arthritis, diabetes, and heart disease affect about half of all adults in the US. Worldwide, they account for more than 90 percent of the morbidity and mortality rates among First-World nations. Here are even more sobering figures to think about:

  • Chronic diseases are a leading cause of death and disability in the US per the CDC. 
  • US healthcare spends approximately $1.65 trillion in the treatment of patients with one or more chronic diseases a year. 
  • 12 percent of those patients who have five or more chronic conditions account for 41 percent of total healthcare spending. 
  • The US Agency for Healthcare Research and Quality estimates 4.4 million hospital admissions cost healthcare $30.8 billion, half of which goes to treat heart disease and diabetes complications. 
  • A study by the American Public Health Association shows that preventing the most common chronic diseases could save healthcare $218 billion per year.

The public and private sectors have huge incentives to reduce chronic disease. Artificial intelligence (AI), which has proven beneficial for such industrial uses as smart factories and food waste management, is being looked upon by healthcare for Chronic Care Management (CCM). Specifically, four areas are being considered: the prevention of CCM, its detection, diagnosis, and treatment.

Prevention, or stop the disease before it happens or gets worse. 

One form of CCM is to prevent the diseases from happening in the first place. AI-algorithms can shift through the vast amounts of patient data stored in medical computer EMRs to identify those at risk for diseases like heart disease, hypertension, and pre-diabetes. Healthcare staff would then offer preventive treatments to those patients like vaccinations, blood pressure medications, and dietary advice.

Results from some of these AI-programs have proven to be surprising. In a 2018 study, researchers at Google used one to identify with 70 percent accuracy which patients would suffer a heart attack or other cardiac problem within five years. It did so based on the patients’ retinal scans.

Detection, or near-continuous monitoring for changes. 

Much of CCM is handled by the patient, if physically and mentally capable, family members, or any caretaker if not. In the past, medical professionals would check up on the patient’s condition at scheduled appointments or during their rounds if they were hospitalized. This proved to be inadequate as their health could change day to day or even hour by hour.

Telehealth for chronic care management has dramatically expanded healthcare’s scope to watch patients and their conditions. Vitals like weight, blood pressure, temperature, oxygen levels, and more can be monitored and transmitted continuously via wearables and remote patient monitoring systems. 

However, the amount of information coming from such biometric devices is enormous. This is where AIs can help. They can go through large amounts of data at lightning speed, looking for specific pieces, patterns, and more depending on their algorithms. They can alert the clinician if the patient is having a major asthma attack, or if their blood sugar levels have remained steady over several weeks under the new diabetic medications. 

Certain forms of chronic pain can be hard to diagnose, especially from patients who can’t communicate or easily express themselves during doctor appointments (example: after a major stroke). Or the pain may only flare up sporadically or under certain conditions outside the medical office. A telemedicine device with an AI paired with facial recognition software can monitor the face muscles of such patients. After they experience a bout of pain, the program can provide a score on the pain chart. The clinician can then adjust treatment plans accordingly. 

AI-detection can also be used to help minimize the readmission of patients back into a hospital 30 days after discharge. Such returning patients cost Medicare between $15 to $20 billion each year. The hospitals, too, suffer as they’re penalized financially for such 30-day readmissions. 

Machine Learning (ML), which is a specific subset of AI, can comb through the patient’s health records and other data like social determinants of health (unhealthy living conditions at home, lack of readily accessible transportation to a nearby medical clinic) to predict if there’s a high chance of readmission. Healthcare staff can then act to reduce its likelihood by providing frequent visits from social workers or even free transport for checkups.

Diagnosis, or finding the best treatments right now. 

Artificial intelligence can also be used to suggest pathologies for a patient’s chronic condition. This is accomplished through ML described above, and Deep-Learning (DL) algorithms. This would dramatically speed up clinicians’ work since only the most appropriate information would be displayed on their medical tablet. Right now, they have to scroll through tons of data to figure out that patient’s current condition and best course of action. 

One such example is the analysis of medical images. Researchers from the UK-based NHS Foundation compared the accuracy of finding diseases through DL-programs versus examination by clinicians like radiologists. In certain studies, DL was found to be just as accurate as the medical specialists. 

Treatment that’s the right one for the patient. 

For the longest time, clinicians used a one-box-fits-all approach to treat patients. This includes those under CCM. Now they can work with the patient to manage their health. ML, DL, and other forms of AI are able to sort through the massive volumes of patient data quickly enough so medical staff can provide suggestions tailored to the unique wants and needs of each patient. This results in more personalized care, less waste in ineffective treatments, and positive outcomes for everyone involved.

Virtual nurses are one example of such custom care. Based on AI, they can help patients stay on track by recording vital data, then issue reminders when necessary when they stray from the clinician-provided treatment plans. Patients using the AI-powered avatars from the medical start-up Sensely start their day by recording vitals like weight and blood pressure. The program then calculates a risk assessment, sending notifications to clinicians if their intervention is necessary.

Closing Thoughts

CCM is a big part of healthcare from the sheer number of patients to be treated to the prohibitive treatment costs. Tools like AI can aid medical clinics and hospitals in chronic disease prevention, detection, diagnosis, and treatment. 

Contact an expert at Cybernet if you’re interested in learning more about the benefits of artificial intelligence in the management of chronic disease for your medical group. 

Join the conversation and connect with us on this and other relevant topics – Follow us Facebook, Twitter, and Linkedin