An approach to automated detection of tumors in mammograms
Citations
526 citations
Cites background or methods from "An approach to automated detection ..."
...Global thresholding Global thresholding has been widely used for segmentation [39,42,43,46]....
[...]
...The classification hierarchy in [42] used the deterministic or Bayesian classifiers with four features to perform classification....
[...]
...A hierarchic classification example [42]....
[...]
...Binary decision tree [42,44,45,53,54,90] A binary decision tree recursively using a threshold to separate mammogram data into two classes each time Intensity features, shape features, texture features Low complexity Accuracy depends fully on the design of the decision tree and the features...
[...]
...[153,154] proposed a new cascade classifier ART2LDA combining an unsupervised classifier ART2 and a supervised classifier based on LDA to improve classification performance Hierarchical architecture Individual classifiers are combined into a tree structure, and each node is associated with a classifier In [42], the first level associated with deterministic classification and the area feature, and the rest three levels used Bayesian classifier and other types of features, such as shape descriptor, edge distance variation descriptor and edge intensity variation...
[...]
414 citations
388 citations
Cites background or methods from "An approach to automated detection ..."
...They demonstrated that this algorithm was directly correlated with the isodata clustering algorithm (Brzakovic et al., 1990)....
[...]
...Breast cancer remains, in the United States as well in the European Union, the leading cause of death for women after their 40 s (Eurostat, 2002; Buseman et al., 2003). However, although breast cancer incidence has increased over the past decade, breast cancer mortality has declined among women of all ages (Sickles, 1997). This favourable trend in mortality reduction may relate to the widespread adoption of mammography screening (Sickles, 1997; De Koning et al., 1995), and improvements made in breast cancer treatment (Buseman et al., 2003). Mammography remains the key screening tool for the detection of breast abnormalities. Vacek et al. (2002) show that the proportion of breast tumours that were detected in Vermont by mammographic screening increased from 2% during 1974–1984 to 36% during 1995–1999....
[...]
...One of the earliest approaches to mass detection was the work of Brzakovic et al. (1990), which was based on a multi-resolution fuzzy pyramid linking approach, a data structure in which the input image formed the basis of the pyramid and each subsequent level (of lower resolution) was sequentially…...
[...]
387 citations
382 citations
Cites methods from "An approach to automated detection ..."
...al [14] developed an automated system for the detection and classi cation of particular types of tumors in digitized mammograms....
[...]
References
37,017 citations
"An approach to automated detection ..." refers methods in this paper
...Next, the output is thresholded using algorithm described in [ 22 ], and possible tumors (that always have higher intensity) are assigned intensity 255, while the background is assigned intensity 0. The next processing stage is to determine if the detected regions are tumors, and, furthermore, to identify the benign tumors....
[...]
6,893 citations
"An approach to automated detection ..." refers methods in this paper
...Consequently, the regions around the boundary of the extracted region are subjected to local edge detection, using the Marr-Hildreth operator [ 20 ],...
[...]
3,237 citations
965 citations
452 citations