Deep Learning Based Prediction of Pulmonary Hypertension in Newborns Using Echocardiograms
Pulmonary hypertension (PH) in newborns and infants is a complex condition associated with several pulmonary, cardiac, and systemic diseases contributing to morbidity and mortality. Thus, accurate and early detection of PH and the classification of its severity is crucial for appropriate and success...
| Main Authors: | , , , , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
Springer US
2024
|
| Online Access: | https://hdl.handle.net/1721.1/153498 |
| _version_ | 1824458136615911424 |
|---|---|
| author | Ragnarsdottir, Hanna Ozkan, Ece Michel, Holger Chin-Cheong, Kieran Manduchi, Laura Wellmann, Sven Vogt, Julia E. |
| author2 | Massachusetts Institute of Technology. Department of Brain and Cognitive Sciences |
| author_facet | Massachusetts Institute of Technology. Department of Brain and Cognitive Sciences Ragnarsdottir, Hanna Ozkan, Ece Michel, Holger Chin-Cheong, Kieran Manduchi, Laura Wellmann, Sven Vogt, Julia E. |
| author_sort | Ragnarsdottir, Hanna |
| collection | MIT |
| description | Pulmonary hypertension (PH) in newborns and infants is a complex condition associated with several pulmonary, cardiac, and systemic diseases contributing to morbidity and mortality. Thus, accurate and early detection of PH and the classification of its severity is crucial for appropriate and successful management. Using echocardiography, the primary diagnostic tool in pediatrics, human assessment is both time-consuming and expertise-demanding, raising the need for an automated approach. Little effort has been directed towards automatic assessment of PH using echocardiography, and the few proposed methods only focus on binary PH classification on the adult population. In this work, we present an explainable multi-view video-based deep learning approach to predict and classify the severity of PH for a cohort of 270 newborns using echocardiograms. We use spatio-temporal convolutional architectures for the prediction of PH from each view, and aggregate the predictions of the different views using majority voting. Our results show a mean F1-score of 0.84 for severity prediction and 0.92 for binary detection using 10-fold cross-validation and 0.63 for severity prediction and 0.78 for binary detection on the held-out test set. We complement our predictions with saliency maps and show that the learned model focuses on clinically relevant cardiac structures, motivating its usage in clinical practice. To the best of our knowledge, this is the first work for an automated assessment of PH in newborns using echocardiograms. |
| first_indexed | 2024-09-23T12:12:19Z |
| format | Article |
| id | mit-1721.1/153498 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2025-02-19T04:21:06Z |
| publishDate | 2024 |
| publisher | Springer US |
| record_format | dspace |
| spelling | mit-1721.1/1534982024-11-13T21:43:59Z Deep Learning Based Prediction of Pulmonary Hypertension in Newborns Using Echocardiograms Ragnarsdottir, Hanna Ozkan, Ece Michel, Holger Chin-Cheong, Kieran Manduchi, Laura Wellmann, Sven Vogt, Julia E. Massachusetts Institute of Technology. Department of Brain and Cognitive Sciences Pulmonary hypertension (PH) in newborns and infants is a complex condition associated with several pulmonary, cardiac, and systemic diseases contributing to morbidity and mortality. Thus, accurate and early detection of PH and the classification of its severity is crucial for appropriate and successful management. Using echocardiography, the primary diagnostic tool in pediatrics, human assessment is both time-consuming and expertise-demanding, raising the need for an automated approach. Little effort has been directed towards automatic assessment of PH using echocardiography, and the few proposed methods only focus on binary PH classification on the adult population. In this work, we present an explainable multi-view video-based deep learning approach to predict and classify the severity of PH for a cohort of 270 newborns using echocardiograms. We use spatio-temporal convolutional architectures for the prediction of PH from each view, and aggregate the predictions of the different views using majority voting. Our results show a mean F1-score of 0.84 for severity prediction and 0.92 for binary detection using 10-fold cross-validation and 0.63 for severity prediction and 0.78 for binary detection on the held-out test set. We complement our predictions with saliency maps and show that the learned model focuses on clinically relevant cardiac structures, motivating its usage in clinical practice. To the best of our knowledge, this is the first work for an automated assessment of PH in newborns using echocardiograms. 2024-02-12T16:24:47Z 2024-02-12T16:24:47Z 2024-02-06 2024-02-11T04:14:43Z Article http://purl.org/eprint/type/JournalArticle https://hdl.handle.net/1721.1/153498 Ragnarsdottir, H., Ozkan, E., Michel, H. et al. Deep Learning Based Prediction of Pulmonary Hypertension in Newborns Using Echocardiograms. Int J Comput Vis (2024). PUBLISHER_CC en https://doi.org/10.1007/s11263-024-01996-x Creative Commons Attribution https://creativecommons.org/licenses/by/4.0/ The Author(s) application/pdf Springer US Springer US |
| spellingShingle | Ragnarsdottir, Hanna Ozkan, Ece Michel, Holger Chin-Cheong, Kieran Manduchi, Laura Wellmann, Sven Vogt, Julia E. Deep Learning Based Prediction of Pulmonary Hypertension in Newborns Using Echocardiograms |
| title | Deep Learning Based Prediction of Pulmonary Hypertension in Newborns Using Echocardiograms |
| title_full | Deep Learning Based Prediction of Pulmonary Hypertension in Newborns Using Echocardiograms |
| title_fullStr | Deep Learning Based Prediction of Pulmonary Hypertension in Newborns Using Echocardiograms |
| title_full_unstemmed | Deep Learning Based Prediction of Pulmonary Hypertension in Newborns Using Echocardiograms |
| title_short | Deep Learning Based Prediction of Pulmonary Hypertension in Newborns Using Echocardiograms |
| title_sort | deep learning based prediction of pulmonary hypertension in newborns using echocardiograms |
| url | https://hdl.handle.net/1721.1/153498 |
| work_keys_str_mv | AT ragnarsdottirhanna deeplearningbasedpredictionofpulmonaryhypertensioninnewbornsusingechocardiograms AT ozkanece deeplearningbasedpredictionofpulmonaryhypertensioninnewbornsusingechocardiograms AT michelholger deeplearningbasedpredictionofpulmonaryhypertensioninnewbornsusingechocardiograms AT chincheongkieran deeplearningbasedpredictionofpulmonaryhypertensioninnewbornsusingechocardiograms AT manduchilaura deeplearningbasedpredictionofpulmonaryhypertensioninnewbornsusingechocardiograms AT wellmannsven deeplearningbasedpredictionofpulmonaryhypertensioninnewbornsusingechocardiograms AT vogtjuliae deeplearningbasedpredictionofpulmonaryhypertensioninnewbornsusingechocardiograms |