Ranking Information Extracted from Uncertainty Quantification of the Prediction of a Deep Learning Model on Medical Time Series Data
Uncertainty quantification in deep learning models is especially important for the medical applications of this complex and successful type of neural architectures. One popular technique is Monte Carlo dropout that gives a sample output for a record, which can be measured statistically in terms of a...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-07-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/8/7/1078 |
| _version_ | 1797563441501700096 |
|---|---|
| author | Ruxandra Stoean Catalin Stoean Miguel Atencia Roberto Rodríguez-Labrada Gonzalo Joya |
| author_facet | Ruxandra Stoean Catalin Stoean Miguel Atencia Roberto Rodríguez-Labrada Gonzalo Joya |
| author_sort | Ruxandra Stoean |
| collection | DOAJ |
| description | Uncertainty quantification in deep learning models is especially important for the medical applications of this complex and successful type of neural architectures. One popular technique is Monte Carlo dropout that gives a sample output for a record, which can be measured statistically in terms of average probability and variance for each diagnostic class of the problem. The current paper puts forward a convolutional–long short-term memory network model with a Monte Carlo dropout layer for obtaining information regarding the model uncertainty for saccadic records of all patients. These are next used in assessing the uncertainty of the learning model at the higher level of sets of multiple records (i.e., registers) that are gathered for one patient case by the examining physician towards an accurate diagnosis. Means and standard deviations are additionally calculated for the Monte Carlo uncertainty estimates of groups of predictions. These serve as a new collection where a random forest model can perform both classification and ranking of variable importance. The approach is validated on a real-world problem of classifying electrooculography time series for an early detection of spinocerebellar ataxia 2 and reaches an accuracy of 88.59% in distinguishing between the three classes of patients. |
| first_indexed | 2024-03-10T18:43:43Z |
| format | Article |
| id | doaj.art-95cbe64b03e64dabbba9efa7b36d9e65 |
| institution | Directory Open Access Journal |
| issn | 2227-7390 |
| language | English |
| last_indexed | 2024-03-10T18:43:43Z |
| publishDate | 2020-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj.art-95cbe64b03e64dabbba9efa7b36d9e652023-11-20T05:41:37ZengMDPI AGMathematics2227-73902020-07-0187107810.3390/math8071078Ranking Information Extracted from Uncertainty Quantification of the Prediction of a Deep Learning Model on Medical Time Series DataRuxandra Stoean0Catalin Stoean1Miguel Atencia2Roberto Rodríguez-Labrada3Gonzalo Joya4Romanian 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, SpainCuban Neuroscience Center, 11600 Havana, CubaDepartment of Electronic Technology, Universidad de Málaga, 29071 Málaga, SpainUncertainty quantification in deep learning models is especially important for the medical applications of this complex and successful type of neural architectures. One popular technique is Monte Carlo dropout that gives a sample output for a record, which can be measured statistically in terms of average probability and variance for each diagnostic class of the problem. The current paper puts forward a convolutional–long short-term memory network model with a Monte Carlo dropout layer for obtaining information regarding the model uncertainty for saccadic records of all patients. These are next used in assessing the uncertainty of the learning model at the higher level of sets of multiple records (i.e., registers) that are gathered for one patient case by the examining physician towards an accurate diagnosis. Means and standard deviations are additionally calculated for the Monte Carlo uncertainty estimates of groups of predictions. These serve as a new collection where a random forest model can perform both classification and ranking of variable importance. The approach is validated on a real-world problem of classifying electrooculography time series for an early detection of spinocerebellar ataxia 2 and reaches an accuracy of 88.59% in distinguishing between the three classes of patients.https://www.mdpi.com/2227-7390/8/7/1078deep learningtime seriesuncertainty quantificationMonte Carlo dropoutrandom forest |
| spellingShingle | Ruxandra Stoean Catalin Stoean Miguel Atencia Roberto Rodríguez-Labrada Gonzalo Joya Ranking Information Extracted from Uncertainty Quantification of the Prediction of a Deep Learning Model on Medical Time Series Data Mathematics deep learning time series uncertainty quantification Monte Carlo dropout random forest |
| title | Ranking Information Extracted from Uncertainty Quantification of the Prediction of a Deep Learning Model on Medical Time Series Data |
| title_full | Ranking Information Extracted from Uncertainty Quantification of the Prediction of a Deep Learning Model on Medical Time Series Data |
| title_fullStr | Ranking Information Extracted from Uncertainty Quantification of the Prediction of a Deep Learning Model on Medical Time Series Data |
| title_full_unstemmed | Ranking Information Extracted from Uncertainty Quantification of the Prediction of a Deep Learning Model on Medical Time Series Data |
| title_short | Ranking Information Extracted from Uncertainty Quantification of the Prediction of a Deep Learning Model on Medical Time Series Data |
| title_sort | ranking information extracted from uncertainty quantification of the prediction of a deep learning model on medical time series data |
| topic | deep learning time series uncertainty quantification Monte Carlo dropout random forest |
| url | https://www.mdpi.com/2227-7390/8/7/1078 |
| work_keys_str_mv | AT ruxandrastoean rankinginformationextractedfromuncertaintyquantificationofthepredictionofadeeplearningmodelonmedicaltimeseriesdata AT catalinstoean rankinginformationextractedfromuncertaintyquantificationofthepredictionofadeeplearningmodelonmedicaltimeseriesdata AT miguelatencia rankinginformationextractedfromuncertaintyquantificationofthepredictionofadeeplearningmodelonmedicaltimeseriesdata AT robertorodriguezlabrada rankinginformationextractedfromuncertaintyquantificationofthepredictionofadeeplearningmodelonmedicaltimeseriesdata AT gonzalojoya rankinginformationextractedfromuncertaintyquantificationofthepredictionofadeeplearningmodelonmedicaltimeseriesdata |