By Simon Gao, Advisor – Artificial Intelligence, February 20, 2022
Patient care journey optimization is realized in a number of ways, across a variety of health care settings, from family care and specialist providers, to acute, emergency, and long-term care providers. In this blog we’ll talk about some successful healthcare Artificial Intelligence (AI) use cases and show how they have improved the patient care journey.
Pre-care:
Healthcare AI for Early Disease Detection and Wellness Management
It’s hard for a healthcare provider to have an overview of every single patient round the clock. It’s arguably even harder for every patient – or potential patient – to determine whether their symptoms require a clinic visit. For these reasons, patients showing early symptoms of diseases – or very gradually worsening conditions – have tended to fly under the radar of healthcare providers until things get a lot more serious. Fortunately, AI in healthcare is about to change all of this.
By playing the role of a patient’s 24/7 personal health advisor, while being able to interpret sets of symptoms with accuracy and make smart, validated recommendations on the patient’s next best actions, AI applications at the pre-care phase of the patient journey are going to transform patient care for good.
Diagnosis:
Healthcare AI uses patient input data and medical history data for machine learning – making diagnostic assessment and recommendations to patient and doctors. Because a lot of the data involved is image or text data, where machine learning has made tremendous advancement in recent years, AI has the potential to revolutionize medical diagnostics, such as through clinical decision support systems or even a front door medical information portal – a use case that’s trending and led by startup healthtech companies lately. The system can help patient with what is the next best action – seeing primary care physician or specialists or in-patient hospital, and making appointments accordingly.
Using AI in the doctor’s office:
One of the best-known use cases revolves around predicting a major medical condition in patients. For example, atrial fibrillation, is a common problem in which people have an irregular heartbeat that’s associated with an increased risk of stroke and heart disease. Because it can happen in episodes, it can be hard to detect in the doctor’s office.
But AI can use big data with many records of health insurance claims and clinical data, along with years’ worth of electronic medical records, to predict who might get atrial fibrillation. The model was able to correctly detect 70% of the patients who were later diagnosed with atrial fibrillation, doctors could then use that data to keep a closer eye on those patients to keep them healthy.
Using AI in Hospital Care:
In the hospital setting, AI can make sense of the overwhelming amount of data created from genomics, biosensors, smartphone apps, the electronic health record (EHR), unstructured notes and data on social determinants of health, and create a broader context for clinicians to deliver high-quality, patient-centered care.
- ER triage optimization: AI models can quickly recommend what to do with the patients and the next best triage recommendation for the patients.
- Patient care level prediction: At the point of admission, AI models can predict the level of care the patient may require, either it is ICU or regular care, what specialist clinicians needs to be involved. There was a study using COVID-19 patient data and sought to predict level-of-care requirements based on clinical and laboratory data. The need for hospitalization, ICU care, and mechanical ventilation were predicted. This information was fed into a capacity management system to optimize ICU bed management.
- Length of stay (LOS) prediction: U.S. hospital stays cost the health system at least $377.5 billion per year and recent Medicare legislation standardizes payments for procedures performed, regardless of the number of days a patient spends in the hospital. This incentivizes hospitals to identify these high LOS risk patients early and have their treatment plan optimized to minimize LOS and lower the chance of getting a hospital-acquired condition such as staph infection. Another benefit is that prior knowledge of LOS can aid in logistics such as room and bed allocation planning.
- Patient discharge management: Streamlining and automating the patient discharge process is a major focus in gaining efficiency among hospitals. The system can use artificial intelligence to drive action across care teams in real time by providing situational awareness to future capacity constraints, identifying and orchestrating discharge activities, and communicating actionable insights from discharge barriers to bottlenecks. By predicting and prioritizing discharges, the care teams quickly could align on the high-priority patients and would be provided with the insights needed to manage throughput and remedy capacity challenges before they arise.
- Patient care monitor and community care: Patient outreach are important for patient follow ups after discharge. Healthcare organizations conduct patient outreach based on health data and key risk factors. Patients who fit certain risk factors may warrant outreach about chronic disease management plans or referrals to social determinants of health services. AI can be used to identify high-risk patients, predict their healthcare needs, and push out personalized communications.
- the number of days a patient spends in the hospital. This incentivizes hospitals to identify these high LOS risk patients early and have their treatment plan optimized to minimize LOS and lower the chance of getting a hospital-acquired condition such as staph infection. Another benefit is that prior knowledge of LOS can aid in logistics such as room and bed allocation planning.
- Patient discharge management: Streamlining and automating the patient discharge process is a major focus in gaining efficiency among hospitals. The system can use artificial intelligence to drive action across care teams in real time by providing situational awareness to future capacity constraints, identifying and orchestrating discharge activities, and communicating actionable insights from discharge barriers to bottlenecks. By predicting and prioritizing discharges, the care teams quickly could align on the high-priority patients and would be provided with the insights needed to manage throughput and remedy capacity challenges before they arise.
- Patient care monitor and community care: Patient outreach are important for patient follow ups after discharge. Healthcare organizations conduct patient outreach based on health data and key risk factors. Patients who fit certain risk factors may warrant outreach about chronic disease management plans or referrals to social determinants of health services. AI can be used to identify high-risk patients, predict their healthcare needs, and push out personalized communications.
