Temporal Mass Detection
暂无分享,去创建一个
We present a method to prompt a clinician to “suspicious” dense regions in temporal mammogram sequences. The particular context that we envisage is mammogram screening, when the clinician compares the most recent mammogram to previous ones in order to detect significant changes. The method uses anisotropic filtering as a pre-processing step in order to significantly reduce the number of candidate masses, while preserving the important anatomical information about each mass. The method has already been tested on 15 temporal pairs, where pathology has been diagnosed in the most recent image.
[1] J. Wolfe. Breast patterns as an index of risk for developing breast cancer. , 1976, AJR. American journal of roentgenology.
[2] Joachim Weickert,et al. Anisotropic diffusion in image processing , 1996 .
[3] Michael Brady,et al. Filtering hint Images for the Detection of Microcalcifications , 2001, MICCAI.
[4] Michael Brady,et al. Multi-scale landmark selection for improved registration of temporal mammograms , 2000 .