Related Experiment Video
Updated: Apr 17, 2026
Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
Published on: August 30, 2013
[Use of information technologies to detect a breast cyst area on mammograms]
Objective:
To use the mammographic image processing algorithm developed by the authors in order to identify a cyst area.
Material And Methods:
The investigations used statistical methods for processing the mammographic films in women with different breast abnormalities, which were taken in the mammography room, Departmental Hospital at the Murom Station, OAO "RZhD", in the first half of 2012. Taking into account the incidence of breast cysts and difficulty in its detection, the authors proposed a specially oriented algorithm to identify a cyst area. It allowed one to assert with a higher degree of probability that the found tumors are cysts due to their typical inner structure and outer outlines. The application of the known algorithms for film processing failed to yield desirable results in the detection of round lumps as they were aimed at solving other tasks.
Results:
During the investigation, 27 mammograms were processed, which showed 12 cysts cytologically diagnosed; the detection rate of this type of a round breast lump was 80-90%.
Conclusion:
The use of this algorithm by a radiologist in his practice will be able to improve the quality of diagnosis particularly during screening tests for women and to reduce the number of pre-examinations using additional methods to identify a round breast lump.
