Abstract
Cancer-survivability prediction is one of the popular research topics, that attracted great attention from both the health service providers and academia. However, one remaining question comes from how to make full use of a large number of available factors (or features). This paper, accordingly, presents a novel autoencoder algorithm based on the concept of sparse coding to address this problem. The main contribution is twofold: the utilization of sparsity coding for input feature selection and a subsequent classification using latent information. Precisely, a typical autoencoder architecture is employed for reconstructing the original input. Then the sparse coding technique is applied to optimize the network structure, with the aim of selecting optimal features and enhancing the generalization capability. In addition, the refined latent information is further cast as alternative features for training a sparse classifier. To evaluate the performance of the proposed autoencoder architecture, we present a comprehensive analysis using a publicly available data repository (i.e., Surveillance, Epidemiology, and End Results, SEER). Experimental study shows that the proposed approach has the ability of extracting important features from high-dimensional inputs and achieves competitive performance than other state-of-the-art classification techniques.
Highlights
Cancer is the second major cause of death globally, according to World Cancer Report
This paper presents the following contributions: (i) e sparse coding technique is introduced for autoencoders, with the aim of performing feature selection and data reconstruction simultaneously; in particular, the feature selection is done by determining features that contribute most to the subsequent classification (ii) Instead of using the raw inputs, the latent information is manipulated as the training inputs for classification (iii) e training process of the proposed autoencoder algorithm is formulated and represented using one unique objective function, which is later solved using an iterative computational strategy e remainder of this paper is organized as follows
Experiments have been performed using a dataset of 500 patient records and 13 features, and the results indicate that the C5.0 algorithm achieves the highest classification accuracy (86.4%)
Summary
Cancer is the second major cause of death globally, according to World Cancer Report (http://publications. iarc.fr/Non-Series-Publications/World-Cancer-Reports/ World-CancerReport-2014). Cancer is the second major cause of death globally, according to World Cancer Report Due to its huge economical and health influence, cancer survivability prediction has received a lot of attention in the last decades. Is research question has been of great interest to either healthcare providers and individual patient as the prediction results provide an effective suggestion and/or measurement to evaluate the prognosis and reduce the significant suffer. Two main aspects studied in this paper are discussed, including existing applications for predicting cancer survivability and the conventional autoencoder algorithm. Cancer is reported as the second major cause of human death globally. Due to its large economic and social impact, healthcare providers and authorities have been spending a lot of effort on cancer-related research, including effective treatment, accurate diagnose with early time and less cost, medical imaging, etc
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
