Explanatory deep learning to predict elevated pulmonary artery pressure in children with ventricular septal defects using standard chest x-rays: a novel approach
ObjectiveEarly risk assessment of pulmonary arterial hypertension (PAH) in patients with congenital heart disease (CHD) is crucial to ensure timely treatment. We hypothesize that applying artificial intelligence (AI) to chest x-rays (CXRs) could identify the future risk of PAH in patients with ventr...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2024-01-01
|
| Series: | Frontiers in Cardiovascular Medicine |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fcvm.2024.1330685/full |
| _version_ | 1797356936176336896 |
|---|---|
| author | Zhixin Li Gang Luo Zhixian Ji Sibao Wang Silin Pan |
| author_facet | Zhixin Li Gang Luo Zhixian Ji Sibao Wang Silin Pan |
| author_sort | Zhixin Li |
| collection | DOAJ |
| description | ObjectiveEarly risk assessment of pulmonary arterial hypertension (PAH) in patients with congenital heart disease (CHD) is crucial to ensure timely treatment. We hypothesize that applying artificial intelligence (AI) to chest x-rays (CXRs) could identify the future risk of PAH in patients with ventricular septal defect (VSD).MethodsA total of 831 VSD patients (161 PAH-VSD, 670 nonPAH-VSD) was retrospectively included. A residual neural networks (ResNet) was trained for classify VSD patients with different outcomes based on chest radiographs. The endpoint of this study was the occurrence of PAH in VSD children before or after surgery.ResultsIn the validation set, the AI algorithm achieved an area under the curve (AUC) of 0.82. In an independent test set, the AI algorithm significantly outperformed human observers in terms of AUC (0.81 vs. 0.65). Class Activation Mapping (CAM) images demonstrated the model's attention focused on the pulmonary artery segment.ConclusionThe preliminary findings of this study suggest that the application of artificial intelligence to chest x-rays in VSD patients can effectively identify the risk of PAH. |
| first_indexed | 2024-03-08T14:37:01Z |
| format | Article |
| id | doaj.art-29658f7c7adc494db152f903f370e3e3 |
| institution | Directory Open Access Journal |
| issn | 2297-055X |
| language | English |
| last_indexed | 2024-03-08T14:37:01Z |
| publishDate | 2024-01-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Cardiovascular Medicine |
| spelling | doaj.art-29658f7c7adc494db152f903f370e3e32024-01-12T04:25:13ZengFrontiers Media S.A.Frontiers in Cardiovascular Medicine2297-055X2024-01-011110.3389/fcvm.2024.13306851330685Explanatory deep learning to predict elevated pulmonary artery pressure in children with ventricular septal defects using standard chest x-rays: a novel approachZhixin LiGang LuoZhixian JiSibao WangSilin PanObjectiveEarly risk assessment of pulmonary arterial hypertension (PAH) in patients with congenital heart disease (CHD) is crucial to ensure timely treatment. We hypothesize that applying artificial intelligence (AI) to chest x-rays (CXRs) could identify the future risk of PAH in patients with ventricular septal defect (VSD).MethodsA total of 831 VSD patients (161 PAH-VSD, 670 nonPAH-VSD) was retrospectively included. A residual neural networks (ResNet) was trained for classify VSD patients with different outcomes based on chest radiographs. The endpoint of this study was the occurrence of PAH in VSD children before or after surgery.ResultsIn the validation set, the AI algorithm achieved an area under the curve (AUC) of 0.82. In an independent test set, the AI algorithm significantly outperformed human observers in terms of AUC (0.81 vs. 0.65). Class Activation Mapping (CAM) images demonstrated the model's attention focused on the pulmonary artery segment.ConclusionThe preliminary findings of this study suggest that the application of artificial intelligence to chest x-rays in VSD patients can effectively identify the risk of PAH.https://www.frontiersin.org/articles/10.3389/fcvm.2024.1330685/fullartificial intelligencepulmonary arterial hypertensionchest x-rayventricular septal defectdeep learning—artificial intelligence |
| spellingShingle | Zhixin Li Gang Luo Zhixian Ji Sibao Wang Silin Pan Explanatory deep learning to predict elevated pulmonary artery pressure in children with ventricular septal defects using standard chest x-rays: a novel approach Frontiers in Cardiovascular Medicine artificial intelligence pulmonary arterial hypertension chest x-ray ventricular septal defect deep learning—artificial intelligence |
| title | Explanatory deep learning to predict elevated pulmonary artery pressure in children with ventricular septal defects using standard chest x-rays: a novel approach |
| title_full | Explanatory deep learning to predict elevated pulmonary artery pressure in children with ventricular septal defects using standard chest x-rays: a novel approach |
| title_fullStr | Explanatory deep learning to predict elevated pulmonary artery pressure in children with ventricular septal defects using standard chest x-rays: a novel approach |
| title_full_unstemmed | Explanatory deep learning to predict elevated pulmonary artery pressure in children with ventricular septal defects using standard chest x-rays: a novel approach |
| title_short | Explanatory deep learning to predict elevated pulmonary artery pressure in children with ventricular septal defects using standard chest x-rays: a novel approach |
| title_sort | explanatory deep learning to predict elevated pulmonary artery pressure in children with ventricular septal defects using standard chest x rays a novel approach |
| topic | artificial intelligence pulmonary arterial hypertension chest x-ray ventricular septal defect deep learning—artificial intelligence |
| url | https://www.frontiersin.org/articles/10.3389/fcvm.2024.1330685/full |
| work_keys_str_mv | AT zhixinli explanatorydeeplearningtopredictelevatedpulmonaryarterypressureinchildrenwithventricularseptaldefectsusingstandardchestxraysanovelapproach AT gangluo explanatorydeeplearningtopredictelevatedpulmonaryarterypressureinchildrenwithventricularseptaldefectsusingstandardchestxraysanovelapproach AT zhixianji explanatorydeeplearningtopredictelevatedpulmonaryarterypressureinchildrenwithventricularseptaldefectsusingstandardchestxraysanovelapproach AT sibaowang explanatorydeeplearningtopredictelevatedpulmonaryarterypressureinchildrenwithventricularseptaldefectsusingstandardchestxraysanovelapproach AT silinpan explanatorydeeplearningtopredictelevatedpulmonaryarterypressureinchildrenwithventricularseptaldefectsusingstandardchestxraysanovelapproach |