Early Detection of Breast Cancer on Mammograms Using: Perceptual Feedback, Computer Processed Images and Ultrasound.
Abstract
Three approaches are being explored for improving the detection of small tumors in the breast: (1) Perceptual feedback to decrease errors in missing tumors that are actually visible on the initial screening mammograms. The observers head-eye-position is recorded while viewing mammograms and the gaze where the eye-position dwell time is concentrated, will be feedback to the observer. The reduction in the number of incorrect decisions with bio-feedback is being studied. (2) Computer processing of screening mammograms is being developed. Computer-aided-detection of clusters of microcalcifications and the relationship between breast carcinoma and the parenchymal pattern associated with the nodular and homogeneous tissues on the mammogram are being studied. The mammogram is finely digitized (42m) with 16-bit depth. This wide dynamic range may facilitate identification of microcalcifications in high and low optical density regions of the mammogram. (3) Ultrasound echoscanning mammography is under development. Wavefront distortion of acoustic waves in breast tissue presently limits the resolution of ultrasound mammography. Existing compensation algorithms correct for the phase distortion of the receiving waveform, however from our studies of tri-vivo ultrasound breast data amplitude distortion of the wavefront is a more serious limitation. New algorithms are being developed to correct for these amplitude distortions in ultrasound mammography.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 1995
- Accession Number
- ADA291599
Entities
People
- Peter Bloch
Organizations
- University of Pennsylvania