uri icon
  • Contact Info
  • 202-994-5910
  • Websites

Murray H Loew Faculty Member

Positions

Key achievements in: 1. Research: Research underway:  Thermal breast imaging: with the goal of early detection of breast tumors, we are acquiring infrared images from normal and cancerous breasts to detect warm regions that may indicate the growth of a tumor. Aspects of the process include: (1) using both classical feature-based pattern-recognition methods and current deep-learning algorithms to identify tumors, and (2) validation of the methods using finite-element analysis to register the true location of a tumor as found at surgery, when the patient is supine, with the thermal image acquired when the patient is sitting.  Development of magnetic resonance (MR) image segmentation methods for detection and risk stratification of prostate cancer  Creation of pathology-preserving anonymization methods to prevent identification of faces in 3d renderings of computerized tomography images of the skull  Development of tools that quantitate the components of explainable artificial intelligence (especially with respect to image analysis), by characterizing separability, assessing data-augmentation techniques, and creating robust segmentation methods  Extension, improvement, and validation of methods of analysis of MR images for early detection of Alzheimer’s disease  Refining and benchmarking our reflectance transformation imaging dome, which provides high-resolution 3d images of surface characteristics of materials  Shape description for reliable discrimination between benign and malignant tumors of the breast and kidney  Combining genomic data with MR images for detection of subtle prostate tumors 2. Teaching: revised and taught a course in pattern recognition and machine learning (graduate, fall 2021), and digital image processing (graduate, spring 2021) to good reviews; on the basis of the courses, recruited several new graduate students to my lab. 3. Service: Member, Advisor Council on Research. Chair, BME search committee, BME personnel committee; ad hoc member of all BME standing committees

Research Areas

GW Expert Finder utilizes up to four primary sources that shares data to populate individual profiles. Reference these FAQs for information on how to update your GW expert profile.
GW Expert Finder utilizes up to four primary sources that shares data to populate individual profiles. Reference these FAQs for information on how to update your GW expert profile.