BCCT.core Model Enhancement With Interpretability and Lateral Information

• Type: Article in International Conference Proceedings Book
• Date: 2010

Authors

Hélder P. Oliveira (Author) FEUP; Jaime S. Cardoso (Author) FEUP

Conference proceedings National

Title: Actas do RecPad 2010

Pages: 1-2

Portuguese Conference on Pattern Recognition

Vila Real, 29 de Outubro de 2010

Scientific classification

FOS: Engineering and technology > Electrical engineering, Electronic engineering, Information engineering

Other information

Abstract (EN): Breast Cancer Conservative Treatment (BCCT) is considered the gold standard of breast cancer treatment. The heterogeneity of the aesthetic result and the limited reproducibility of the subjective evaluation motivated the research towards objective methods, such as, the recent computer system named BCCT.core, based on machine learning techniques, namely support vector machines (SVMs). In the current work we investigate the accuracy of different interpretable methods against the model currently deployed in the BCCT.core software and the improvement of the model by introducing lateral information extracted from patients images. Experimental results show only a marginal improvement in the performance of the new models, suggesting that is essential to use more robust models, such as 3D approaches.

Language: English

Type (Professor's evaluation): Scientific

No. of pages: 2