Pareto-Optimal Multi-Objective Dimensionality Reduction Deep Auto-Encoder for Mammography Classification

Although mammography is an effective modality for early breast cancer detection and diagnosis, on mammographic examinations, 10-30% of cancerous/noncancerous lesions may be misinterpreted [1]. To overcome this, computer aided diagnosis (CADx) systems have been developed. The accuracy of CADx for x-ray breast mammography still requires improvements to be useable as a flawless guide (an alert system flagging potential misclassification to the human operator) for radiologists or an independent clinical interpreter [2,3].

Source: Computer Methods and Programs in Biomedicine - April 13, 2017 Category: Bioinformatics Authors: Saeid Asgari Taghanki, Jeremy Kawahara, Brandon Miles, Ghassan Hamarneh Source Type: research