A Novel End-to-End Classifier Using Domain Transferred Deep Convolutional Neural Networks for Biomedical Images

In the past, biomedical image analysis for diagnosis of diseases for patients was performed using traditional radiological screening techniques and manual identification, which is rather time-consuming and suffers from several critical limitations including the subjectivity of the diagnosed result and the variability between laboratories. Furthermore, with the recent drastic advances in the biomedical image technology [1], many types of biomedical images have been generated in a variety of modalities using advanced medical imaging devices, including CT (Computed Tomography), X-Ray, MS (Microscopic), US (Ultrasound) and MRI (Magnetic Resonance Imaging).

Source: Computer Methods and Programs in Biomedicine - January 5, 2017 Category: Bioinformatics Authors: Shuchao Pang, Zhezhou Yu, Mehmet A. Orgun Source Type: research