5 ways machine learning can improve patients' experience and outcomes

Artificial intelligence and machine learning improve processes though data-driven insights and automation. These tools are becoming increasingly integrated into healthcare, but their adoption on the clinical front line remains slow.

During an Oct. 15 webinar hosted by Becker's Hospital Review and sponsored by Health at Scale, two industry leaders discussed how machine learning can improve clinical operations and care delivery. The webinar was moderated by Elisabeth Berger, Health at Scale's director of growth.

Ms. Berger spoke with:

  • Mohammed Saeed, MD, PhD, chief medical officer of Health at Scale

  • John Guttag, PhD, chief technology officer of Health at Scale

The speakers discussed five ways machine learning can improve healthcare for patients and providers.

  1. Machine learning can eliminate the one-size fits-all approach. Dr. Saeed said "we are just at the dawn of using machine learning in healthcare," and noted that the technology has potential to greatly personalize and improve the patient experience by taking into consideration data about a patient's chronic conditions, genetics, location, age, activity level and other determinants of health.

  2. Machine learning can predict which patients are most at-risk. By using clinical, demographic and social data to determine which patients have the highest probability of experiencing adverse events, machine learning models can identify which patients need to be contacted and brought in for preventative or intervention-based care.

  3. Machine learning can match patients to the right facilities and specialists. Machine learning models can help patients find the healthcare providers that best fits their needs, Dr. Saeed said. These models can predict at which facilities patients will have the best outcomes and direct those who may be at risk for complications to facilities that are the most prepared to care for them.

  4. Machine learning can bring about faster interventions through anomaly detection. Dr. Guttag said machine learning tools, such as remote monitoring devices, can track patients' behavior and flag any activity that is unusual for them. For example, if a remote monitoring device picks up that a patient hasn't moved for nine hours, it can identify the anomaly so that the patient can be checked on.

  5. Machine learning can be applied to improve healthcare beyond clinical settings. Dr. Guttag pointed out hospitals use machine learning to do coding that maximizes reimbursement, and payers use it to look for fraud, waste and abuse. He also said the pharmaceutical industry uses machine learning to help design drugs, and researchers use the technology frequently for population studies.

To learn more about Health at Scale, click here. To watch the full webinar, click here.

© Copyright ASC COMMUNICATIONS 2021. Interested in LINKING to or REPRINTING this content? View our policies by clicking here.