|Microcalcification cluster diagnosis in digitized mammograms|
|Written by Ramón Gallardo Caballero|
Page 1 of 6
This line is an application for Computer Aided Diagnosing (CADx) based on pattern classification and image analysis. Our methodology is centred on the use of Independent Component Analysis (ICA) as a feature extractor engine, Genetic Algorithms as feature selector and Neural Networks as classification system. Source data is taken from "Digital Database for Screening Mammography" (DDSM) provided by Island Vision Group at USF.
The incidence of breast cancer in western women varies from 40 to 75 per 100000, being the most frequent tumour among the Spanish feminine population. Some estimations states that around 6000 women die as a consequence of this disease, converting it in women’s first cancer death cause, with a mortality rate of 28.2 per 100000. The current probability for a Spanish woman to develop a breast cancer before reach 75 years old is almost 5%. This implies that one among 20 women will develop a breast cancer before that age.
This pessimistic statistics illustrates the problem magnitude. Although some risk factors have been identified, effective prevention measures or specific and effective treatments are unknown.
The absence of a clear risk factor, different of age, with high significance in disease appearance makes difficult to establish any effective measure in breast cancer prevention. Nowadays, early detection of breast cancer constitutes the most effective step in this battle.
The commonest detection method in different health systems is mammography analysis, an X based technique which provides high spatial resolution images, better than ultrasound imaging. Most early detected lesions in mammography are done by microcalcification cluster localisation.
A microcalcification is a very small structure (typically lower than 1 millimetre), when they appear grouped in some characteristic shapes (microcalcification cluster, MCC) usually indicates the presence of a growing abnormally.
The detection of such structures sometimes presents some difficulty. Microcalcifications are relatively small, sometimes they appear in low contrast areas and must be detected by a human expert, who can be fatigued or can reduce his attention level. This later reason makes very interesting the possibility to use a Computer Aided Diagnostic system (CAD) as a way to reduce the possibilities of misdetection of a developing breast cancer.
|Last Updated ( jueves, 25 noviembre 2010 )|