Positive and Negative Evidence Accumulation Clustering for Sensor Fusion: An Application to Heartbeat Clustering
In this work, a new clustering algorithm especially geared towards merging data arising from multiple sensors is presented. The algorithm, called PN-EAC, is based on the ensemble clustering paradigm and it introduces the novel concept of negative evidence. PN-EAC combines both positive evidence, to...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2019-10-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/19/21/4635 |
| _version_ | 1798039177438167040 |
|---|---|
| author | David G. Márquez Paulo Félix Constantino A. García Javier Tejedor Ana L.N. Fred Abraham Otero |
| author_facet | David G. Márquez Paulo Félix Constantino A. García Javier Tejedor Ana L.N. Fred Abraham Otero |
| author_sort | David G. Márquez |
| collection | DOAJ |
| description | In this work, a new clustering algorithm especially geared towards merging data arising from multiple sensors is presented. The algorithm, called PN-EAC, is based on the ensemble clustering paradigm and it introduces the novel concept of negative evidence. PN-EAC combines both positive evidence, to gather information about the elements that should be grouped together in the final partition, and negative evidence, which has information about the elements that should not be grouped together. The algorithm has been validated in the electrocardiographic domain for heartbeat clustering, extracting positive evidence from the heartbeat morphology and negative evidence from the distances between heartbeats. The best result obtained on the MIT-BIH Arrhythmia database yielded an error of 1.44%. In the St. Petersburg Institute of Cardiological Technics 12-Lead Arrhythmia Database database (INCARTDB), an error of 0.601% was obtained when using two electrocardiogram (ECG) leads. When increasing the number of leads to 4, 6, 8, 10 and 12, the algorithm obtains better results (statistically significant) than with the previous number of leads, reaching an error of 0.338%. To the best of our knowledge, this is the first clustering algorithm that is able to process simultaneously any number of ECG leads. Our results support the use of PN-EAC to combine different sources of information and the value of the negative evidence. |
| first_indexed | 2024-04-11T21:50:18Z |
| format | Article |
| id | doaj.art-393335cde7bb472ba19096b01c75e5f5 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-11T21:50:18Z |
| publishDate | 2019-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-393335cde7bb472ba19096b01c75e5f52022-12-22T04:01:16ZengMDPI AGSensors1424-82202019-10-011921463510.3390/s19214635s19214635Positive and Negative Evidence Accumulation Clustering for Sensor Fusion: An Application to Heartbeat ClusteringDavid G. Márquez0Paulo Félix1Constantino A. García2Javier Tejedor3Ana L.N. Fred4Abraham Otero5Department of Information Technology, Escuela Politécnica Superior, Universidad San Pablo-CEU, CEU Universities, Campus Montepríncipe, Boadilla del Monte, 28668 Madrid, SpainCentro de Investigación en Tecnoloxías da Información (CiTIUS), University of Santiago de Compostela, 15782 Santiago de Compostela, SpainDepartment of Information Technology, Escuela Politécnica Superior, Universidad San Pablo-CEU, CEU Universities, Campus Montepríncipe, Boadilla del Monte, 28668 Madrid, SpainDepartment of Information Technology, Escuela Politécnica Superior, Universidad San Pablo-CEU, CEU Universities, Campus Montepríncipe, Boadilla del Monte, 28668 Madrid, SpainInstituto de Telecomunicações, Instituto Superior Técnico, 1049-001 Lisboa, PortugalDepartment of Information Technology, Escuela Politécnica Superior, Universidad San Pablo-CEU, CEU Universities, Campus Montepríncipe, Boadilla del Monte, 28668 Madrid, SpainIn this work, a new clustering algorithm especially geared towards merging data arising from multiple sensors is presented. The algorithm, called PN-EAC, is based on the ensemble clustering paradigm and it introduces the novel concept of negative evidence. PN-EAC combines both positive evidence, to gather information about the elements that should be grouped together in the final partition, and negative evidence, which has information about the elements that should not be grouped together. The algorithm has been validated in the electrocardiographic domain for heartbeat clustering, extracting positive evidence from the heartbeat morphology and negative evidence from the distances between heartbeats. The best result obtained on the MIT-BIH Arrhythmia database yielded an error of 1.44%. In the St. Petersburg Institute of Cardiological Technics 12-Lead Arrhythmia Database database (INCARTDB), an error of 0.601% was obtained when using two electrocardiogram (ECG) leads. When increasing the number of leads to 4, 6, 8, 10 and 12, the algorithm obtains better results (statistically significant) than with the previous number of leads, reaching an error of 0.338%. To the best of our knowledge, this is the first clustering algorithm that is able to process simultaneously any number of ECG leads. Our results support the use of PN-EAC to combine different sources of information and the value of the negative evidence.https://www.mdpi.com/1424-8220/19/21/4635sensor fusionclusteringevidence accumulationfusion techniquesmachine learningecgmultilead clusteringheartbeat clusteringmultimodal clustering |
| spellingShingle | David G. Márquez Paulo Félix Constantino A. García Javier Tejedor Ana L.N. Fred Abraham Otero Positive and Negative Evidence Accumulation Clustering for Sensor Fusion: An Application to Heartbeat Clustering Sensors sensor fusion clustering evidence accumulation fusion techniques machine learning ecg multilead clustering heartbeat clustering multimodal clustering |
| title | Positive and Negative Evidence Accumulation Clustering for Sensor Fusion: An Application to Heartbeat Clustering |
| title_full | Positive and Negative Evidence Accumulation Clustering for Sensor Fusion: An Application to Heartbeat Clustering |
| title_fullStr | Positive and Negative Evidence Accumulation Clustering for Sensor Fusion: An Application to Heartbeat Clustering |
| title_full_unstemmed | Positive and Negative Evidence Accumulation Clustering for Sensor Fusion: An Application to Heartbeat Clustering |
| title_short | Positive and Negative Evidence Accumulation Clustering for Sensor Fusion: An Application to Heartbeat Clustering |
| title_sort | positive and negative evidence accumulation clustering for sensor fusion an application to heartbeat clustering |
| topic | sensor fusion clustering evidence accumulation fusion techniques machine learning ecg multilead clustering heartbeat clustering multimodal clustering |
| url | https://www.mdpi.com/1424-8220/19/21/4635 |
| work_keys_str_mv | AT davidgmarquez positiveandnegativeevidenceaccumulationclusteringforsensorfusionanapplicationtoheartbeatclustering AT paulofelix positiveandnegativeevidenceaccumulationclusteringforsensorfusionanapplicationtoheartbeatclustering AT constantinoagarcia positiveandnegativeevidenceaccumulationclusteringforsensorfusionanapplicationtoheartbeatclustering AT javiertejedor positiveandnegativeevidenceaccumulationclusteringforsensorfusionanapplicationtoheartbeatclustering AT analnfred positiveandnegativeevidenceaccumulationclusteringforsensorfusionanapplicationtoheartbeatclustering AT abrahamotero positiveandnegativeevidenceaccumulationclusteringforsensorfusionanapplicationtoheartbeatclustering |