Abstract: The lack of labelled data poses a main challenge in applying deep learning to medical imaging. Despite of the availability of large amounts of clinical data, it is difficult to acquire labelled image data, in particular for bone Scintigraphy (i.e. 2D bone imaging). This paper presents a neural network model that can classify bone cancer metastases in the chest area in a semi supervised manner. This deep learning model, classifies each instance independently, utilizes global information through an additional connection from the core network thereby achieving higher accuracy.

Keywords: Bone Scintigraphy, Neural Network, Semi-supervised learning, Clinical Imaging

| DOI: 10.17148/IJIREEICE.2021.9633