A hybrid unsupervised-Deep learning tandem for electrooculography time series analysis
| Autor: | Francisco García-Lagos, Ruxandra Stoean, Luis Velázquez-Pérez, Roberto Becerra-García, Gonzalo Joya, Miguel Atencia, Rodolfo García-Bermúdez, Catalin Stoean |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
Time Factors
Eye Movements Computer science Physiology Visual System Sensory Physiology 02 engineering and technology Task (project management) Machine Learning Electrocardiography Animal Cells 0202 electrical engineering electronic engineering information engineering Medicine and Health Sciences Cluster Analysis Neurons 0303 health sciences Multidisciplinary Movement Disorders medicine.diagnostic_test Artificial neural network Applied Mathematics Simulation and Modeling Neurodegenerative Diseases Electrooculography Sensory Systems Identification (information) Bioassays and Physiological Analysis Databases as Topic Neurology Physical Sciences Unsupervised learning Medicine 020201 artificial intelligence & image processing Cellular Types Algorithms Research Article Computer and Information Sciences Similarity (geometry) Neural Networks Science Research and Analysis Methods 03 medical and health sciences Deep Learning Artificial Intelligence medicine Saccades Humans Cluster analysis 030304 developmental biology business.industry Deep learning Electrophysiological Techniques Biology and Life Sciences Pattern recognition Cell Biology Cellular Neuroscience Ataxia Artificial intelligence Neural Networks Computer Cardiac Electrophysiology business Photic Stimulation Mathematics Unsupervised Machine Learning Neuroscience |
| Zdroj: | PLoS ONE PLoS ONE, Vol 15, Iss 7, p e0236401 (2020) |
| ISSN: | 1932-6203 |
| Popis: | Medical data are often tricky to get mined for patterns even by the generally demonstrated successful modern methodologies of deep learning. This paper puts forward such a medical classification task, where patient registers of two of the categories are sometimes hard to be distinguished because of samples showing characteristics of both labels in turn in several repetitions of the screening procedure. To this end, the current research appoints a pre-processing clustering step (through self-organizing maps) to group the data based on shape similarity and relabel it accordingly. Subsequently, a deep learning approach (a tandem of convolutional and long short-term memory networks) performs the training classification phase on the 'cleaned' samples. The dual methodology was applied for the computational diagnosis of electrooculography tests within spino-cerebral ataxia of type 2. The accuracy obtained for the discrimination into three classes was of 78.24%. The improvement that this duo brings over the deep learner alone does not stem from significantly higher accuracy results when the performance is considered for all classes. The major finding of this combination is that half of the presymptomatic cases were correctly found, in opposition to the single deep model, where this category was sacrificed by the learner in favor of a good accuracy overall. A high accuracy in general is desirable for any medical task, however the correct identification of cases before the symptoms become evident is more important. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|