BIOMEDICAL IMAGES ENCODE TRACES OF AN UNDERLYING PHENOTYPE, WE ARE LEARNING TO TRANSLATE THEM INTO COMPUTATIONAL BIOMARKERS.
BIOMEDICAL IMAGES ENCODE TRACES OF AN UNDERLYING PHENOTYPE, WE ARE LEARNING TO TRANSLATE THEM INTO COMPUTATIONAL BIOMARKERS.
Our lab focuses on the theory, development, and application of multiscale imaging biomarkers. We study the imaging phenotype across length-scales (e.g., radiological {~10-3 m}, pathological {~10-6 m}, and molecular {~10-9 m} ) and time-scales (e.g., the natural history of disease, dynamic response to treatment, etc.). We develop mathematical methods, physics-based models, computational imaging techniques, and data fusion algorithms to quantify the appearance and behavior of disease across space and time.
Our lab focuses on the theory, development, and application of multiscale imaging biomarkers. We study the imaging phenotype across length-scales (e.g., radiological {~10-3 m}, pathological {~10-6 m}, and molecular {~10-9 m} ) and time-scales (e.g., the natural history of disease, dynamic response to treatment, etc.). We develop mathematical methods, physics-based models, computational imaging techniques, and data fusion algorithms to quantify the appearance and behavior of disease across space and time.
(a) Mathematical model of tumor shrinkage based on the applied analysis of stochastic differential equations. (b) Spatial-temporal clustering of the time dynamics reveal unique tumor habitats.
(a) Mathematical model of tumor shrinkage based on the applied analysis of stochastic differential equations. (b) Spatial-temporal clustering of the time dynamics reveal unique tumor habitats.
Interested in how we convert images into precision medicine?
Interested in how we convert images into precision medicine?
Funding
Funding