Care2Vec: a hybrid autoencoder-based approach for the classification of self-care problems in physically disabled children

Accurate classification of self-care problems in children who suffer from physical and motor affliction is an important problem in the healthcare industry. This is a difficult and a time-consuming process, and it needs the expertise of occupational therapists. In recent years, healthcare professionals have opened up to the idea of using expert systems and artificial intelligence in the diagnosis and classification of self-care problems. In this study, we propose a new deep learning-based approach named Care2Vec for solving the self-care classification problem. We use a real-world self-care activities dataset that is based on a conceptual framework designed by the World Health Organization. The conceptual framework is known as the International Classification of Functioning, Disability and Health for Children and Youth (ICF-CY), which is a widely used standard framework for analyzing self-care activity records. Deep learning is a form of representation learning, and in recent years, it has been very successful in various fields such as computer vision, speech processing and more. We propose a hybrid autoencoder-based method Care2Vec, where we use autoencoders and deep neural networks as a two-step modeling process. We also propose a variation of the Care2Vec method. We then compare Care2Vec with traditional methods reported in the literature for solving the self-care classification problem (such as deep neural networks and decision trees) in both multi-class classification and binary classification settings. We use k -fold cross-validation while applying the different methodologies. The evaluation metrics used were the mean accuracy and the mean area under the curve (AUC). We found that the Care2Vec method has a better prediction accuracy than the prevalent methods and so a recommended approach for the self-care classification problem. The adoption of Care2Vec can help expert therapists in making better diagnostic decisions and will thus lead to better treatment.