4TH International Congress on Technology - Engineering & SCIENCE - Kuala Lumpur - Malaysia (2017-08-05)

Deep Convolutional-based Medical Retrieval System Using Radon Transformation

Image classification and retrieval systems have gained more attention because of accessing to high-tech medical imaging, however lack of availability of large-scaled balanced labelled data in medicine is still a challenge. Simplicity, practicality, efficiency, and effectiveness are the targets should be followed in medical domain. To achieve these gains, Radon transformation, which is well-known in medical society, is utilized along with a deep network to propose a retrieval system in a highly imbalanced medical benchmark. The main contribution of this study is to propose a deep structural model which is training on the Radon-based transformed input data. The experimental results show that applying this transformation, utilized many years ago in medicine, as input, feeding to a convolutional neural network, significantly increases the performance, compared with other retrieval systems, using Radon and traditional systems. Considering this contribution, properly increases the performance, compared with almost all models which use Radon transformation to retrieve medical images.
Amin Khatami, Morteza Babaie, Hamid Reza Tizhoosh, Abbas Khosravi, Saeid Nahavandi