Full Program »
Use of Deep learning methods in the domain of psychiatry
Over the past few years, the field of Deep Learning (DL) based algorithms showed a great promise in extracting features and learning patterns from complex data. The use of these techniques especially in the spatial-temporal domain is gaining a lot of attention. DL architectures and algorithms have made impressive advances in fields such as image recognition and speech processing, followed by significant contributions with state-of-the-art results for some common Natural Language Processing (NLP) tasks. These techniques, widely used in face recognition and other images - and speech-recognition applications, has shown promise in many areas from farming towards high energy physics, astronomy and medicine. Due to the promising results of DL in various fields as also in medical applications, DL is spreading into the industry. However, these techniques still gravitate in the area of Computer Vision and Natural and Language Processing, while time series analysis still mostly integrates classical probabilistic approaches. Similar classical statistical approaches and machine learning (ML) techniques for classification and nonlinear regression, like support vector machines (SVM) or shallow neural networks have been placed in psychiatry and neuroscience. While classification and prediction tasks with Deep Neural Networks DNN with vast amount of medical images already finds their way in medical applications, processing of signals is not addressed in such density yet. In this paper we will address the use of DL in the domain of psychiatry.