Mammography is a non-invasive diagnostic technique largely used for early cancer detection in women's breasts. One of the main indicants of such disease is the presence of microcalcifications, appearing as small bright spots in the mammographic image. An automatic detection and recognition of malignant clusters of microcalcifications, although very useful for a mass screening of the female population at risk, is very difficult to accomplish because of the small size of the microcalcifications and of the poor quality of the mammographic images. In this paper we propose a novel approach, based on the adoption of a multiple expert system (MES). Such a system aggregates several experts, some of which are devoted to classifying the single microcalcifications while others are aimed at recognizing the malignancy of the cluster considered as a whole. The final classification decision of the system results from the combination of the outputs of the single experts. The approach has been successfully tested on a standard database of 40 mammographic images, publicly available
Automatic classification of clustered microcalcifications by a multiple expert system
DE VITO, SAVERIO;TORTORELLA, Francesco;
1999-01-01
Abstract
Mammography is a non-invasive diagnostic technique largely used for early cancer detection in women's breasts. One of the main indicants of such disease is the presence of microcalcifications, appearing as small bright spots in the mammographic image. An automatic detection and recognition of malignant clusters of microcalcifications, although very useful for a mass screening of the female population at risk, is very difficult to accomplish because of the small size of the microcalcifications and of the poor quality of the mammographic images. In this paper we propose a novel approach, based on the adoption of a multiple expert system (MES). Such a system aggregates several experts, some of which are devoted to classifying the single microcalcifications while others are aimed at recognizing the malignancy of the cluster considered as a whole. The final classification decision of the system results from the combination of the outputs of the single experts. The approach has been successfully tested on a standard database of 40 mammographic images, publicly availableI documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
