Stewart Wang (University of Michigan)
15 Apr. 2019. A new company provides a service that determines a person’s condition for disease treatments based on indicators derived from an analysis of huge medical-image databases. Applied Morphomics Inc. in Ann Arbor is a 2 year-old spin-off enterprise founded by surgery professor Stewart Wang at University of Michigan medical school, also in Ann Arbor.
Wang, who specializes in treatments for burns and other injuries, also studies morphomics, the analysis of biomarkers or molecular indicators of a person’s physical condition. Rather than identifying those biomarkers from specimen samples such as blood or urine, Wang and colleagues in the Morphomics Analysis Group find digital biomarkers in medical images, such as computed tomography or CT scans, compared to large-scale databases of similar images collected over the years. Specimen samples generally have a short shelf life and are discarded after tests are performed, while medical images can be stored almost indefinitely, and added to the database for later analysis.
“This is the ultimate selfie,” says Wang in a university statement. “A patient’s body is their biological medical record and contains a tremendous amount of information that clinicians to date have not been able to comprehend.”
Wang and colleagues write algorithms that analyze CT images looking for visual evidence of digital biomarkers. Their analysis so far covers CT scans from more than 100,000 individuals over a 20-year period. From that collection, the researchers calculate baselines for different ages, genders, and fitness levels. And from those baselines, the techniques make it possible to identify digital biomarkers related to specific diseases or conditions. For example, muscle mass identified in CT scans is a good indicator of an individual’s overall health, and can help predict recovery from major surgery or trauma, compared to similar age and gender groups at large.
In the past year, Wang’s lab published a study showing factors identified through CT scans, such as bone density and body fat, can predict successful lung transplant procedures, as well as highlight factors influencing the length of hospital stays and survival times. Another study published last year uses morphomics for computing an index to screen for malnutrition in adults, comparing people independently diagnosed with malnutrition to healthy kidney donors, and indicating nutrition levels among the kidney donors. The lab makes available a reference set of CT scans from 6,000 trauma patients at University of Michigan called Reference Analytic Morphomic Population, or RAMP, as well as a library of morphomics measures for body mass, spine, and other bones.
Wang’s company Applied Morphomics refines the algorithms to provide guidance to individual patients based on digital biomarkers in the person’s CT scans compared to baselines for the population at large. Applied Morphomics has an exclusive license from the university to commercialize the lab’s technology, which it offers as a tool for precision medicine decision-making.
“The deep analysis of the variety and often less apparent nuances of our physical structure that have been developed by the morphomics project,” notes Wang, “is a phenomenally deep and sophisticated benchmarking reference. The variety of applications for this collection of personalized structural human roadmaps is staggering in its potential applications.”
More from Science & Enterprise:
- Genetic Risk Analysis Shows Ethnic Bias
- Genomics, A.I. Calculate Immune Health
- Population Data Show Genetics-Health Links
- High-Speed Screens, A.I., Boost Nanoscale Medicines
- Large Molecule Library Enables On-Demand Drug Discovery
* * *
RSS - Posts
You must be logged in to post a comment.