Automated Detection of Presymptomatic Conditions in Spinocerebellar Ataxia Type 2 Using Monte Carlo Dropout and Deep Neural Network Techniques with Electrooculogram Signals
Application of deep learning (DL) to the field of healthcare is aiding clinicians to make an accurate diagnosis. DL provides reliable results for image processing and sensor interpretation problems most of the time. However, model uncertainty should also be thoroughly quantified. This paper therefor...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-05-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/20/11/3032 |
| _version_ | 1827716167786037248 |
|---|---|
| author | Catalin Stoean Ruxandra Stoean Miguel Atencia Moloud Abdar Luis Velázquez-Pérez Abbas Khosravi Saeid Nahavandi U. Rajendra Acharya Gonzalo Joya |
| author_facet | Catalin Stoean Ruxandra Stoean Miguel Atencia Moloud Abdar Luis Velázquez-Pérez Abbas Khosravi Saeid Nahavandi U. Rajendra Acharya Gonzalo Joya |
| author_sort | Catalin Stoean |
| collection | DOAJ |
| description | Application of deep learning (DL) to the field of healthcare is aiding clinicians to make an accurate diagnosis. DL provides reliable results for image processing and sensor interpretation problems most of the time. However, model uncertainty should also be thoroughly quantified. This paper therefore addresses the employment of Monte Carlo dropout within the DL structure to automatically discriminate presymptomatic signs of spinocerebellar ataxia type 2 in saccadic samples obtained from electrooculograms. The current work goes beyond the common incorporation of this special type of dropout into deep neural networks and uses the uncertainty derived from the validation samples to construct a decision tree at the register level of the patients. The decision tree built from the uncertainty estimates obtained a classification accuracy of 81.18% in automatically discriminating control, presymptomatic and sick classes. This paper proposes a novel method to address both uncertainty quantification and explainability to develop reliable healthcare support systems. |
| first_indexed | 2024-03-10T19:33:46Z |
| format | Article |
| id | doaj.art-7640803b83bf419eb16a27c703639419 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T19:33:46Z |
| publishDate | 2020-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-7640803b83bf419eb16a27c7036394192023-11-20T01:52:44ZengMDPI AGSensors1424-82202020-05-012011303210.3390/s20113032Automated Detection of Presymptomatic Conditions in Spinocerebellar Ataxia Type 2 Using Monte Carlo Dropout and Deep Neural Network Techniques with Electrooculogram SignalsCatalin Stoean0Ruxandra Stoean1Miguel Atencia2Moloud Abdar3Luis Velázquez-Pérez4Abbas Khosravi5Saeid Nahavandi6U. Rajendra Acharya7Gonzalo Joya8Romanian Institute of Science and Technology, 400022 Cluj-Napoca, RomaniaRomanian Institute of Science and Technology, 400022 Cluj-Napoca, RomaniaDepartment of Applied Mathematics, Universidad de Málaga, 29071 Málaga, SpainInstitute for Intelligent Systems Research and Innovation (IISRI), Deakin University, Geelong 3216, AustraliaCuban Academy of Sciences, La Habana 10100, CubaInstitute for Intelligent Systems Research and Innovation (IISRI), Deakin University, Geelong 3216, AustraliaInstitute for Intelligent Systems Research and Innovation (IISRI), Deakin University, Geelong 3216, AustraliaDepartment of Electronics and Computer Engineering, Ngee Ann Polytechnic, Clementi 599489, SingaporeDepartment of Electronic Technology, Universidad de Málaga, 29071 Málaga, SpainApplication of deep learning (DL) to the field of healthcare is aiding clinicians to make an accurate diagnosis. DL provides reliable results for image processing and sensor interpretation problems most of the time. However, model uncertainty should also be thoroughly quantified. This paper therefore addresses the employment of Monte Carlo dropout within the DL structure to automatically discriminate presymptomatic signs of spinocerebellar ataxia type 2 in saccadic samples obtained from electrooculograms. The current work goes beyond the common incorporation of this special type of dropout into deep neural networks and uses the uncertainty derived from the validation samples to construct a decision tree at the register level of the patients. The decision tree built from the uncertainty estimates obtained a classification accuracy of 81.18% in automatically discriminating control, presymptomatic and sick classes. This paper proposes a novel method to address both uncertainty quantification and explainability to develop reliable healthcare support systems.https://www.mdpi.com/1424-8220/20/11/3032deep learningmedicinesensor dataelectrooculogramuncertainty quantificationMonte Carlo dropout |
| spellingShingle | Catalin Stoean Ruxandra Stoean Miguel Atencia Moloud Abdar Luis Velázquez-Pérez Abbas Khosravi Saeid Nahavandi U. Rajendra Acharya Gonzalo Joya Automated Detection of Presymptomatic Conditions in Spinocerebellar Ataxia Type 2 Using Monte Carlo Dropout and Deep Neural Network Techniques with Electrooculogram Signals Sensors deep learning medicine sensor data electrooculogram uncertainty quantification Monte Carlo dropout |
| title | Automated Detection of Presymptomatic Conditions in Spinocerebellar Ataxia Type 2 Using Monte Carlo Dropout and Deep Neural Network Techniques with Electrooculogram Signals |
| title_full | Automated Detection of Presymptomatic Conditions in Spinocerebellar Ataxia Type 2 Using Monte Carlo Dropout and Deep Neural Network Techniques with Electrooculogram Signals |
| title_fullStr | Automated Detection of Presymptomatic Conditions in Spinocerebellar Ataxia Type 2 Using Monte Carlo Dropout and Deep Neural Network Techniques with Electrooculogram Signals |
| title_full_unstemmed | Automated Detection of Presymptomatic Conditions in Spinocerebellar Ataxia Type 2 Using Monte Carlo Dropout and Deep Neural Network Techniques with Electrooculogram Signals |
| title_short | Automated Detection of Presymptomatic Conditions in Spinocerebellar Ataxia Type 2 Using Monte Carlo Dropout and Deep Neural Network Techniques with Electrooculogram Signals |
| title_sort | automated detection of presymptomatic conditions in spinocerebellar ataxia type 2 using monte carlo dropout and deep neural network techniques with electrooculogram signals |
| topic | deep learning medicine sensor data electrooculogram uncertainty quantification Monte Carlo dropout |
| url | https://www.mdpi.com/1424-8220/20/11/3032 |
| work_keys_str_mv | AT catalinstoean automateddetectionofpresymptomaticconditionsinspinocerebellarataxiatype2usingmontecarlodropoutanddeepneuralnetworktechniqueswithelectrooculogramsignals AT ruxandrastoean automateddetectionofpresymptomaticconditionsinspinocerebellarataxiatype2usingmontecarlodropoutanddeepneuralnetworktechniqueswithelectrooculogramsignals AT miguelatencia automateddetectionofpresymptomaticconditionsinspinocerebellarataxiatype2usingmontecarlodropoutanddeepneuralnetworktechniqueswithelectrooculogramsignals AT moloudabdar automateddetectionofpresymptomaticconditionsinspinocerebellarataxiatype2usingmontecarlodropoutanddeepneuralnetworktechniqueswithelectrooculogramsignals AT luisvelazquezperez automateddetectionofpresymptomaticconditionsinspinocerebellarataxiatype2usingmontecarlodropoutanddeepneuralnetworktechniqueswithelectrooculogramsignals AT abbaskhosravi automateddetectionofpresymptomaticconditionsinspinocerebellarataxiatype2usingmontecarlodropoutanddeepneuralnetworktechniqueswithelectrooculogramsignals AT saeidnahavandi automateddetectionofpresymptomaticconditionsinspinocerebellarataxiatype2usingmontecarlodropoutanddeepneuralnetworktechniqueswithelectrooculogramsignals AT urajendraacharya automateddetectionofpresymptomaticconditionsinspinocerebellarataxiatype2usingmontecarlodropoutanddeepneuralnetworktechniqueswithelectrooculogramsignals AT gonzalojoya automateddetectionofpresymptomaticconditionsinspinocerebellarataxiatype2usingmontecarlodropoutanddeepneuralnetworktechniqueswithelectrooculogramsignals |