Artificial Intelligence & Machine Learning
A Health Nucleus assessment is unique because we apply our proprietary Artificial Intelligence (AI) to all of the data collected about you. This enables us to provide you with personalized health reports that give you detailed information to make informed, data-driven decisions for a healthy future.
Artificial Intelligence (AI) uses technology to perform tasks usually reserved for the human brain, mimicking cognitive function.
Machine Learning (ML) looks for patterns to predict outcomes and help make decisions.
Deep learning is a type of machine learning. Inspired by the human brain, it uses multi-layered neural networks to find complex patterns and relationships in large data sets that traditional ML may miss.
Discovering What Matters
Your Health Nucleus assessment produces more than 150 gigabytes of data, compared to the average 80 megabytes a patient typically generates each year.* The ability for a human to integrate and access that vast amount of data is simply impossible. However, applying AI to all that data allows hidden correlations to be uncovered.
The AI developed at Health Nucleus pull from our proprietary database of imaging, metabolic and genomic data to automatically identify disease. Neural networks derive quantitative biomarkers to monitor your health status and risk, particularly for the top chronic age-related diseases: cancer, cardiovascular, neurologic and metabolic.
How Does AI Benefit Your Health Decisions?
The information derived from our application of AI is used to provide a greater understanding about your individual genetics and current phenotypic measurements. Examples of phenotypic measurements derived with deep learning include age-related brain atrophy and detailed body composition derived from your whole body MRI.
The personalized health reports integrate this information leveraging machine learning to inform changes in lifestyle that can reduce your future risk of disease. The value is in discovering what is most important about your health; today, for your future, and your family’s future.
Alzheimer’s and Dementia
To understand the value of the Health Nucleus approach versus the existing model, let’s apply it to accessing Alzheimer’s risk.
The current approach for predicting an individual’s risk of developing disease relies primarily on a single set of data or biomarkers. For example, your APOE genetic status is a well-documented genetic biomarker for risk of developing Alzheimer’s disease (need reference). Hippocampal occupancy score (HOC), looking at the volume of your hippocampus, measured by brain MRI, is another biomarker associated with the onset of dementia. (reference). The gold current standard for Alzheimer’s detection is amyloid PET screening, however, this is a costly and invasive procedure that requires injection of radioactive isotope to obtain the images.
Health Nucleus Approach
Health Nucleus has developed a proprietary MRI imaging protocol and genetics model to access Alzheimer’s risk equally as well as amyloid PET screening without the use of radioactive isotope, thus reducing the patient’s exposure to radiation. Our Alzheimer’s risk analysis model can predict your risk of Alzheimer’s disease 10 years prior to diagnosis, and is more accurate in predicting onset of dementia than stand-alone HOC testing.
Integrated Approach Leads to New Models
Our integrated approach is not just producing cutting-edge standards for disease risk analysis, we are redefining the measure of health. On-going research will continue to push the limits of understanding as we develop integrated disease risk models to be used in clinic practice. These actuarial-like risk models will provide personalized assessments of a person’s risk of developing dementia, coronary artery disease, and diabetes, providing high-risk individuals ways to evaluate their health status and potentially mitigate their risk of disease progression using modifiable risk factors.