New computer aided diagnostic system using deep neural network and SVM to detect breast cancer in thermography

Medical diagnostic support system based on breast thermography using Raspberry Pi and cloud computing

“…Several segmentation techniques have been proposed in the literature to solve this problem, but they obtained moderate success rates [6], [15]. In an attempt to properly segment the breast regions, we used a deep learning U-net algorithm [16] which has showed great performance in similar pattern recognition and computer vision tasks [11], [17]- [19]. Figure 3 illustrates the result of this segmentation operation on breast thermograms.…”

“…In Figure 3(a) before processing and Fogure 3(b) after processing. More details on segmentation, model architectures and training configuration can be found in our previous work [16].…”

“…In order to obtain high accuracy in the classification step, the most relevant characteristics of interest in the image are extracted, in particular statistical and textural features [6], [16], [20]. Several supervised learning techniques were then tested as classifiers for the proposed computer aided detection (CAD) system using the extracted features.…”

“…SVM builds a model capable of distinguishing the belonging class (normal or abnormal breast) of any future data based on the support vectors obtained by the training dataset [20], [21]. More details on feature extraction and classification steps can be found in our previous work [16].…”

“…In the proposed method, we used the Lineage OS as the operating system in order to access the cloud computing facilities through an Android application. To correctly achieve this task, the algorithm [16] was written on the Pycharm integrated development environment using the python language.…”