Medical imaging plays a role as a crucial source of data for disease detection and diagnosis. Recent advancements in machine learning and deep learning have become an efficient tool for medical image analysis. Medical image research laboratories are rapidly creating machine learning systems to achieve the professional performance of humans. However, both machine learning and deep learning methods are complex and require a lot of expertise, resources, knowledge, and time to train. Those create a significant barrier for researchers. In this study, we propose a convolutional neural network architecture to detect abnormalities in bone images. The proposed method provides insight into medical images and explains in detail how the model supports the diagnosis.