A Deep Learning Segmentation Pipeline for Cardiac T1 Mapping Using MRI Relaxation-based Synthetic Contrast Augmentation
© 2022 by the Radiological Society of North America, Inc..
Purpose: To design and evaluate an automated deep learning method for segmentation and analysis of cardiac MRI T1 maps with use of synthetic T1-weighted images for MRI relaxation-based contrast augmentation.
Materials and Methods: This retrospective study included MRI scans acquired between 2016 and 2019 from 100 patients (mean age ± SD, 55 years ± 13; 72 men) across various clinical abnormalities with use of a modified Look-Locker inversion recovery, or MOLLI, sequence to quantify native T1 (T1native), postcontrast T1 (T1post), and extracellular volume (ECV). Data were divided into training (n = 60) and internal (n = 40) test subsets. "Synthetic" T1-weighted images were generated from the T1 exponential inversion-recovery signal model at a range of optimal inversion times, yielding high blood-myocardium contrast, and were used for contrast-based image augmentation during training and testing of a convolutional neural network for myocardial segmentation. Automated segmentation, T1, and ECV were compared with experts with use of Dice similarity coefficients (DSCs), correlation coefficients, and Bland-Altman analysis. An external test dataset (n = 147) was used to assess model generalization.
Results: Internal testing showed high myocardial DSC relative to experts (0.81 ± 0.08), which was similar to interobserver DSC (0.81 ± 0.08). Automated segmental measurements strongly correlated with experts (T1native, R = 0.87; T1post, R = 0.91; ECV, R = 0.92), which were similar to interobserver correlation (T1native, R = 0.86; T1post, R = 0.94; ECV, R = 0.95). External testing showed strong DSC (0.80 ± 0.09) and T1native correlation (R = 0.88) between automatic and expert analysis.
Conclusion: This deep learning method leveraging synthetic contrast augmentation may provide accurate automated T1 and ECV analysis for cardiac MRI data acquired across different abnormalities, centers, scanners, and T1 sequences.Keywords: MRI, Cardiac, Tissue Characterization, Segmentation, Convolutional Neural Network, Deep Learning Algorithms, Machine Learning Algorithms, Supervised Learning Supplemental material is available for this article. © RSNA, 2022.
| Medienart: |
E-Artikel |
|---|
| Erscheinungsjahr: |
2022 |
|---|---|
| Erschienen: |
2022 |
| Enthalten in: |
Zur Gesamtaufnahme - volume:4 |
|---|---|
| Enthalten in: |
Radiology. Artificial intelligence - 4(2022), 6 vom: 08. Nov., Seite e210294 |
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Bhatt, Nitish [VerfasserIn] |
|---|
| Links: |
|---|
| Themen: |
Cardiac |
|---|
| Anmerkungen: |
Date Revised 21.12.2022 published: Electronic-eCollection Citation Status PubMed-not-MEDLINE |
|---|
| doi: |
10.1148/ryai.210294 |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
NLM350355614 |
|---|
| LEADER | 01000naa a22002652 4500 | ||
|---|---|---|---|
| 001 | NLM350355614 | ||
| 003 | DE-627 | ||
| 005 | 20231226044817.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 231226s2022 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1148/ryai.210294 |2 doi | |
| 028 | 5 | 2 | |a pubmed24n1167.xml |
| 035 | |a (DE-627)NLM350355614 | ||
| 035 | |a (NLM)36523641 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Bhatt, Nitish |e verfasserin |4 aut | |
| 245 | 1 | 2 | |a A Deep Learning Segmentation Pipeline for Cardiac T1 Mapping Using MRI Relaxation-based Synthetic Contrast Augmentation |
| 264 | 1 | |c 2022 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a ƒaComputermedien |b c |2 rdamedia | ||
| 338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Date Revised 21.12.2022 | ||
| 500 | |a published: Electronic-eCollection | ||
| 500 | |a Citation Status PubMed-not-MEDLINE | ||
| 520 | |a © 2022 by the Radiological Society of North America, Inc. | ||
| 520 | |a Purpose: To design and evaluate an automated deep learning method for segmentation and analysis of cardiac MRI T1 maps with use of synthetic T1-weighted images for MRI relaxation-based contrast augmentation | ||
| 520 | |a Materials and Methods: This retrospective study included MRI scans acquired between 2016 and 2019 from 100 patients (mean age ± SD, 55 years ± 13; 72 men) across various clinical abnormalities with use of a modified Look-Locker inversion recovery, or MOLLI, sequence to quantify native T1 (T1native), postcontrast T1 (T1post), and extracellular volume (ECV). Data were divided into training (n = 60) and internal (n = 40) test subsets. "Synthetic" T1-weighted images were generated from the T1 exponential inversion-recovery signal model at a range of optimal inversion times, yielding high blood-myocardium contrast, and were used for contrast-based image augmentation during training and testing of a convolutional neural network for myocardial segmentation. Automated segmentation, T1, and ECV were compared with experts with use of Dice similarity coefficients (DSCs), correlation coefficients, and Bland-Altman analysis. An external test dataset (n = 147) was used to assess model generalization | ||
| 520 | |a Results: Internal testing showed high myocardial DSC relative to experts (0.81 ± 0.08), which was similar to interobserver DSC (0.81 ± 0.08). Automated segmental measurements strongly correlated with experts (T1native, R = 0.87; T1post, R = 0.91; ECV, R = 0.92), which were similar to interobserver correlation (T1native, R = 0.86; T1post, R = 0.94; ECV, R = 0.95). External testing showed strong DSC (0.80 ± 0.09) and T1native correlation (R = 0.88) between automatic and expert analysis | ||
| 520 | |a Conclusion: This deep learning method leveraging synthetic contrast augmentation may provide accurate automated T1 and ECV analysis for cardiac MRI data acquired across different abnormalities, centers, scanners, and T1 sequences.Keywords: MRI, Cardiac, Tissue Characterization, Segmentation, Convolutional Neural Network, Deep Learning Algorithms, Machine Learning Algorithms, Supervised Learning Supplemental material is available for this article. © RSNA, 2022 | ||
| 650 | 4 | |a Journal Article | |
| 650 | 4 | |a Cardiac | |
| 650 | 4 | |a Convolutional Neural Network | |
| 650 | 4 | |a Deep Learning Algorithms | |
| 650 | 4 | |a MRI | |
| 650 | 4 | |a Machine Learning Algorithms | |
| 650 | 4 | |a Segmentation | |
| 650 | 4 | |a Supervised Learning | |
| 650 | 4 | |a Tissue Characterization | |
| 700 | 1 | |a Ramanan, Venkat |e verfasserin |4 aut | |
| 700 | 1 | |a Orbach, Ady |e verfasserin |4 aut | |
| 700 | 1 | |a Biswas, Labonny |e verfasserin |4 aut | |
| 700 | 1 | |a Ng, Matthew |e verfasserin |4 aut | |
| 700 | 1 | |a Guo, Fumin |e verfasserin |4 aut | |
| 700 | 1 | |a Qi, Xiuling |e verfasserin |4 aut | |
| 700 | 1 | |a Guo, Lancia |e verfasserin |4 aut | |
| 700 | 1 | |a Jimenez-Juan, Laura |e verfasserin |4 aut | |
| 700 | 1 | |a Roifman, Idan |e verfasserin |4 aut | |
| 700 | 1 | |a Wright, Graham A |e verfasserin |4 aut | |
| 700 | 1 | |a Ghugre, Nilesh R |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Radiology. Artificial intelligence |d 2019 |g 4(2022), 6 vom: 08. Nov., Seite e210294 |w (DE-627)NLM304586358 |x 2638-6100 |7 nnns |
| 773 | 1 | 8 | |g volume:4 |g year:2022 |g number:6 |g day:08 |g month:11 |g pages:e210294 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.1148/ryai.210294 |3 Volltext |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a GBV_NLM | ||
| 951 | |a AR | ||
| 952 | |d 4 |j 2022 |e 6 |b 08 |c 11 |h e210294 | ||