Detection of Abnormal Cardiac Response Patterns in Cardiac Tissue Using Deep Learning
This study reports a method for the detection of mechanical signaling anomalies in cardiac tissue through the use of deep learning and the design of two anomaly detectors. In contrast to anomaly classifiers, anomaly detectors allow accurate identification of the time position of the anomaly. The fir...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-08-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/10/15/2786 |
| _version_ | 1797413056376995840 |
|---|---|
| author | Xavier Marimon Sara Traserra Marcel Jiménez Andrés Ospina Raúl Benítez |
| author_facet | Xavier Marimon Sara Traserra Marcel Jiménez Andrés Ospina Raúl Benítez |
| author_sort | Xavier Marimon |
| collection | DOAJ |
| description | This study reports a method for the detection of mechanical signaling anomalies in cardiac tissue through the use of deep learning and the design of two anomaly detectors. In contrast to anomaly classifiers, anomaly detectors allow accurate identification of the time position of the anomaly. The first detector used a recurrent neural network (RNN) of long short-term memory (LSTM) type, while the second used an autoencoder. Mechanical contraction data present several challanges, including high presence of noise due to the biological variability in the contraction response, noise introduced by the data acquisition chain and a wide variety of anomalies. Therefore, we present a robust deep-learning-based anomaly detection framework that addresses these main issues, which are difficult to address with standard unsupervised learning techniques. For the time series recording, an experimental model was designed in which signals of cardiac mechanical contraction (right and left atria) of a CD-1 mouse could be acquired in an automatic organ bath, reproducing the physiological conditions. In order to train the anomaly detection models and validate their performance, a database of synthetic signals was designed (<i>n</i> = 800 signals), including a wide range of anomalous events observed in the experimental recordings. The detector based on the LSTM neural network was the most accurate. The performance of this detector was assessed by means of experimental mechanical recordings of cardiac tissue of the right and left atria. |
| first_indexed | 2024-03-09T05:12:02Z |
| format | Article |
| id | doaj.art-7fb36a1c1b3447b9b614d77c43506208 |
| institution | Directory Open Access Journal |
| issn | 2227-7390 |
| language | English |
| last_indexed | 2024-03-09T05:12:02Z |
| publishDate | 2022-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj.art-7fb36a1c1b3447b9b614d77c435062082023-12-03T12:48:35ZengMDPI AGMathematics2227-73902022-08-011015278610.3390/math10152786Detection of Abnormal Cardiac Response Patterns in Cardiac Tissue Using Deep LearningXavier Marimon0Sara Traserra1Marcel Jiménez2Andrés Ospina3Raúl Benítez4Automatic Control Department, Universitat Politècnica de Catalunya (UPC-BarcelonaTECH), 08034 Barcelona, SpainFaculty of Veterinary Medicine, Universitat Autònoma de Barcelona, 08193 Barcelona, SpainFaculty of Veterinary Medicine, Universitat Autònoma de Barcelona, 08193 Barcelona, SpainComputer Science Department, Universitat Oberta de Catalunya (UOC), 08018 Barcelona, SpainAutomatic Control Department, Universitat Politècnica de Catalunya (UPC-BarcelonaTECH), 08034 Barcelona, SpainThis study reports a method for the detection of mechanical signaling anomalies in cardiac tissue through the use of deep learning and the design of two anomaly detectors. In contrast to anomaly classifiers, anomaly detectors allow accurate identification of the time position of the anomaly. The first detector used a recurrent neural network (RNN) of long short-term memory (LSTM) type, while the second used an autoencoder. Mechanical contraction data present several challanges, including high presence of noise due to the biological variability in the contraction response, noise introduced by the data acquisition chain and a wide variety of anomalies. Therefore, we present a robust deep-learning-based anomaly detection framework that addresses these main issues, which are difficult to address with standard unsupervised learning techniques. For the time series recording, an experimental model was designed in which signals of cardiac mechanical contraction (right and left atria) of a CD-1 mouse could be acquired in an automatic organ bath, reproducing the physiological conditions. In order to train the anomaly detection models and validate their performance, a database of synthetic signals was designed (<i>n</i> = 800 signals), including a wide range of anomalous events observed in the experimental recordings. The detector based on the LSTM neural network was the most accurate. The performance of this detector was assessed by means of experimental mechanical recordings of cardiac tissue of the right and left atria.https://www.mdpi.com/2227-7390/10/15/2786deep learningautoencodercardiac tissueelectrophysiologyelectrostimulationanomaly detection |
| spellingShingle | Xavier Marimon Sara Traserra Marcel Jiménez Andrés Ospina Raúl Benítez Detection of Abnormal Cardiac Response Patterns in Cardiac Tissue Using Deep Learning Mathematics deep learning autoencoder cardiac tissue electrophysiology electrostimulation anomaly detection |
| title | Detection of Abnormal Cardiac Response Patterns in Cardiac Tissue Using Deep Learning |
| title_full | Detection of Abnormal Cardiac Response Patterns in Cardiac Tissue Using Deep Learning |
| title_fullStr | Detection of Abnormal Cardiac Response Patterns in Cardiac Tissue Using Deep Learning |
| title_full_unstemmed | Detection of Abnormal Cardiac Response Patterns in Cardiac Tissue Using Deep Learning |
| title_short | Detection of Abnormal Cardiac Response Patterns in Cardiac Tissue Using Deep Learning |
| title_sort | detection of abnormal cardiac response patterns in cardiac tissue using deep learning |
| topic | deep learning autoencoder cardiac tissue electrophysiology electrostimulation anomaly detection |
| url | https://www.mdpi.com/2227-7390/10/15/2786 |
| work_keys_str_mv | AT xaviermarimon detectionofabnormalcardiacresponsepatternsincardiactissueusingdeeplearning AT saratraserra detectionofabnormalcardiacresponsepatternsincardiactissueusingdeeplearning AT marceljimenez detectionofabnormalcardiacresponsepatternsincardiactissueusingdeeplearning AT andresospina detectionofabnormalcardiacresponsepatternsincardiactissueusingdeeplearning AT raulbenitez detectionofabnormalcardiacresponsepatternsincardiactissueusingdeeplearning |