Mayo Clinic's Health Engineering and Analytics Laboratory (HEAL) develops computer engineering systems to enable precision medicine and improve well-being.
Multi-omics for precision medicine
The laboratory develops machine-learning workflows to integrate multiple modalities of longitudinal and cross-sectional omics data with clinical, social, nutritional and demographic factors. The goal of these workflows is to identify biomarkers that can:
- Predict drug response.
- Infer disease pathophysiology.
- Explain the impact of environmental exposures on human functioning and disease pathophysiology.
The multi-omic paradigm represents a significant step toward understanding the implications for precision medicine. Specifically, multi-omics are valuable because individuals' health, disease and drug-response dispositions have many factors and are often not explained by genomics alone. HEAL has demonstrated multi-omic integration capabilities to identify biomarkers of antidepressant response, suicide ideation and impact of environmental exposures in rare diseases.
Digital health for child and adolescent psychiatry
The adoption of wearables in children is increasing, and experts expect wearables to serve as a platform for parents to monitor their children's activities and well-being. The prevalence of child and adolescent mental health conditions also is increasing — at an alarming rate. However, access to related mental health specialists still remains limited.
To address this gap, HEAL develops new analytical platforms that use physiological biomarkers from wearables such as smartwatches to predict children's behavior states. Predicting disruptive behavior in children 4 to 10 years old — before it manifests with disruptive behavior disorder, autism or attention-deficit/hyperactivity disorder — allows parents time to seek help. It provides a window to learn about and start using therapeutic parenting practices, such as parent-child interaction therapy.
Digital health for worker well-being, burnout and mental health
When job demands exceed job resources, occupational burnout occurs. Unfortunately, workers often work without knowing they are in burnout, which is often associated with errors in the workplace. Burnout also is associated with increased risk of chronic health conditions such as depression and cardiovascular diseases.
Dr. Athreya's laboratory has developed groundbreaking new longitudinal, decentralized digital health studies using wearables such as smartwatches. These studies employ artificial intelligence approaches that use physiological, psychological and workplace data to predict impending risk of burnout.
As a case study, HEAL is investigating burnout in registered nurses across the Mayo Clinic enterprise. Data sources for this study include:
- Hospital systems, such as patient acuity and staffing.
- Psychological well-being burnout surveys.
- Patient satisfaction feedback.
The research team pools these factors to predict impending burnout at the individual level and unit level.