Deep Learning Supplants Visual Analysis by Experienced Operators for the Diagnosis of Cardiac Amyloidosis by Cine-CMR
Background: Diagnosing cardiac amyloidosis (CA) from cine-CMR (cardiac magnetic resonance) alone is not reliable. In this study, we tested if a convolutional neural network (CNN) could outperform the visual diagnosis of experienced operators. Method: 119 patients with cardiac amyloidosis and 122 pat...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-12-01
|
| Series: | Diagnostics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-4418/12/1/69 |
| _version_ | 1797494760954396672 |
|---|---|
| author | Philippe Germain Armine Vardazaryan Nicolas Padoy Aissam Labani Catherine Roy Thomas Hellmut Schindler Soraya El Ghannudi |
| author_facet | Philippe Germain Armine Vardazaryan Nicolas Padoy Aissam Labani Catherine Roy Thomas Hellmut Schindler Soraya El Ghannudi |
| author_sort | Philippe Germain |
| collection | DOAJ |
| description | Background: Diagnosing cardiac amyloidosis (CA) from cine-CMR (cardiac magnetic resonance) alone is not reliable. In this study, we tested if a convolutional neural network (CNN) could outperform the visual diagnosis of experienced operators. Method: 119 patients with cardiac amyloidosis and 122 patients with left ventricular hypertrophy (LVH) of other origins were retrospectively selected. Diastolic and systolic cine-CMR images were preprocessed and labeled. A dual-input visual geometry group (VGG ) model was used for binary image classification. All images belonging to the same patient were distributed in the same set. Accuracy and area under the curve (AUC) were calculated per frame and per patient from a 40% held-out test set. Results were compared to a visual analysis assessed by three experienced operators. Results: frame-based comparisons between humans and a CNN provided an accuracy of 0.605 vs. 0.746 (<i>p</i> < 0.0008) and an AUC of 0.630 vs. 0.824 (<i>p</i> < 0.0001). Patient-based comparisons provided an accuracy of 0.660 vs. 0.825 (<i>p</i> < 0.008) and an AUC of 0.727 vs. 0.895 (<i>p</i> < 0.002). Conclusion: based on cine-CMR images alone, a CNN is able to discriminate cardiac amyloidosis from LVH of other origins better than experienced human operators (15 to 20 points more in absolute value for accuracy and AUC), demonstrating a unique capability to identify what the eyes cannot see through classical radiological analysis. |
| first_indexed | 2024-03-10T01:39:01Z |
| format | Article |
| id | doaj.art-03aed893a8724d8baa91af368f19b1c6 |
| institution | Directory Open Access Journal |
| issn | 2075-4418 |
| language | English |
| last_indexed | 2024-03-10T01:39:01Z |
| publishDate | 2021-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Diagnostics |
| spelling | doaj.art-03aed893a8724d8baa91af368f19b1c62023-11-23T13:27:41ZengMDPI AGDiagnostics2075-44182021-12-011216910.3390/diagnostics12010069Deep Learning Supplants Visual Analysis by Experienced Operators for the Diagnosis of Cardiac Amyloidosis by Cine-CMRPhilippe Germain0Armine Vardazaryan1Nicolas Padoy2Aissam Labani3Catherine Roy4Thomas Hellmut Schindler5Soraya El Ghannudi6Department of Radiology, Nouvel Hopital Civil, University Hospital, 67000 Strasbourg, FranceICube, University of Strasbourg, CNRS, 67000 Strasbourg, FranceICube, University of Strasbourg, CNRS, 67000 Strasbourg, FranceDepartment of Radiology, Nouvel Hopital Civil, University Hospital, 67000 Strasbourg, FranceDepartment of Radiology, Nouvel Hopital Civil, University Hospital, 67000 Strasbourg, FranceMallinckrodt Institute of Radiology, Division of Nuclear Medicine, Washington University School of Medicine, Saint Louis, MO 63110, USADepartment of Radiology, Nouvel Hopital Civil, University Hospital, 67000 Strasbourg, FranceBackground: Diagnosing cardiac amyloidosis (CA) from cine-CMR (cardiac magnetic resonance) alone is not reliable. In this study, we tested if a convolutional neural network (CNN) could outperform the visual diagnosis of experienced operators. Method: 119 patients with cardiac amyloidosis and 122 patients with left ventricular hypertrophy (LVH) of other origins were retrospectively selected. Diastolic and systolic cine-CMR images were preprocessed and labeled. A dual-input visual geometry group (VGG ) model was used for binary image classification. All images belonging to the same patient were distributed in the same set. Accuracy and area under the curve (AUC) were calculated per frame and per patient from a 40% held-out test set. Results were compared to a visual analysis assessed by three experienced operators. Results: frame-based comparisons between humans and a CNN provided an accuracy of 0.605 vs. 0.746 (<i>p</i> < 0.0008) and an AUC of 0.630 vs. 0.824 (<i>p</i> < 0.0001). Patient-based comparisons provided an accuracy of 0.660 vs. 0.825 (<i>p</i> < 0.008) and an AUC of 0.727 vs. 0.895 (<i>p</i> < 0.002). Conclusion: based on cine-CMR images alone, a CNN is able to discriminate cardiac amyloidosis from LVH of other origins better than experienced human operators (15 to 20 points more in absolute value for accuracy and AUC), demonstrating a unique capability to identify what the eyes cannot see through classical radiological analysis.https://www.mdpi.com/2075-4418/12/1/69cardiac amyloidosisAL/TTR amyloidosishypertrophic cardiomyopathyleft ventricular hypertrophydeep learningconvolutional neural network |
| spellingShingle | Philippe Germain Armine Vardazaryan Nicolas Padoy Aissam Labani Catherine Roy Thomas Hellmut Schindler Soraya El Ghannudi Deep Learning Supplants Visual Analysis by Experienced Operators for the Diagnosis of Cardiac Amyloidosis by Cine-CMR Diagnostics cardiac amyloidosis AL/TTR amyloidosis hypertrophic cardiomyopathy left ventricular hypertrophy deep learning convolutional neural network |
| title | Deep Learning Supplants Visual Analysis by Experienced Operators for the Diagnosis of Cardiac Amyloidosis by Cine-CMR |
| title_full | Deep Learning Supplants Visual Analysis by Experienced Operators for the Diagnosis of Cardiac Amyloidosis by Cine-CMR |
| title_fullStr | Deep Learning Supplants Visual Analysis by Experienced Operators for the Diagnosis of Cardiac Amyloidosis by Cine-CMR |
| title_full_unstemmed | Deep Learning Supplants Visual Analysis by Experienced Operators for the Diagnosis of Cardiac Amyloidosis by Cine-CMR |
| title_short | Deep Learning Supplants Visual Analysis by Experienced Operators for the Diagnosis of Cardiac Amyloidosis by Cine-CMR |
| title_sort | deep learning supplants visual analysis by experienced operators for the diagnosis of cardiac amyloidosis by cine cmr |
| topic | cardiac amyloidosis AL/TTR amyloidosis hypertrophic cardiomyopathy left ventricular hypertrophy deep learning convolutional neural network |
| url | https://www.mdpi.com/2075-4418/12/1/69 |
| work_keys_str_mv | AT philippegermain deeplearningsupplantsvisualanalysisbyexperiencedoperatorsforthediagnosisofcardiacamyloidosisbycinecmr AT arminevardazaryan deeplearningsupplantsvisualanalysisbyexperiencedoperatorsforthediagnosisofcardiacamyloidosisbycinecmr AT nicolaspadoy deeplearningsupplantsvisualanalysisbyexperiencedoperatorsforthediagnosisofcardiacamyloidosisbycinecmr AT aissamlabani deeplearningsupplantsvisualanalysisbyexperiencedoperatorsforthediagnosisofcardiacamyloidosisbycinecmr AT catherineroy deeplearningsupplantsvisualanalysisbyexperiencedoperatorsforthediagnosisofcardiacamyloidosisbycinecmr AT thomashellmutschindler deeplearningsupplantsvisualanalysisbyexperiencedoperatorsforthediagnosisofcardiacamyloidosisbycinecmr AT sorayaelghannudi deeplearningsupplantsvisualanalysisbyexperiencedoperatorsforthediagnosisofcardiacamyloidosisbycinecmr |