A 2-step deep learning method with domain adaptation for multi-centre, multi-vendor and multi-disease cardiac magnetic resonance segmentation
Segmentation of anatomical structures from Cardiac Magnetic Resonance (CMR) is central to the non-invasive quantitative assessment of cardiac function and structure. Anatomical variability, imaging heterogeneity and cardiac dynamics challenge the automation of this task. Deep learning (DL) approache...
| Main Authors: | , , , , |
|---|---|
| Format: | Conference item |
| Language: | English |
| Published: |
Springer
2021
|
| _version_ | 1797086415548317696 |
|---|---|
| author | Corral Acero, J Sundaresan, V Dinsdale, N Grau, V Jenkinson, M |
| author_facet | Corral Acero, J Sundaresan, V Dinsdale, N Grau, V Jenkinson, M |
| author_sort | Corral Acero, J |
| collection | OXFORD |
| description | Segmentation of anatomical structures from Cardiac Magnetic Resonance (CMR) is central to the non-invasive quantitative assessment of cardiac function and structure. Anatomical variability, imaging heterogeneity and cardiac dynamics challenge the automation of this task. Deep learning (DL) approaches have taken over the field of automatic segmentation in recent years, however they are limited by data availability and the additional variability introduced by differences in scanners and protocols. In this work, we propose a 2-step fully automated pipeline to segment CMR images, based on DL encoder-decoder frameworks, and we explore two domain adaptation techniques, domain adversarial training and iterative domain unlearning, to overcome the imaging heterogeneity limitations. We evaluate our methods on the MICCAI 2020 Multi-Centre, Multi-Vendor & Multi-Disease Cardiac Image Segmentation Challenge training and validation datasets. The results show the improvement in performance produced by domain adaptation models, especially among the seen vendors. Finally, we build an ensemble of baseline and domain adapted networks, that reported state-of-art mean Dice scores of 0.912, 0.857 and 0.861 for left ventricle (LV) cavity, LV myocardium and right ventricle cavity, respectively, on the externally validated Challenge dataset, including several unseen vendors, centers and cardiac pathologies.
|
| first_indexed | 2024-03-07T02:21:44Z |
| format | Conference item |
| id | oxford-uuid:a428a71b-810e-45ad-a0a9-5bced9343378 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-07T02:21:44Z |
| publishDate | 2021 |
| publisher | Springer |
| record_format | dspace |
| spelling | oxford-uuid:a428a71b-810e-45ad-a0a9-5bced93433782022-03-27T02:31:59ZA 2-step deep learning method with domain adaptation for multi-centre, multi-vendor and multi-disease cardiac magnetic resonance segmentationConference itemhttp://purl.org/coar/resource_type/c_5794uuid:a428a71b-810e-45ad-a0a9-5bced9343378EnglishSymplectic ElementsSpringer2021Corral Acero, JSundaresan, VDinsdale, NGrau, VJenkinson, MSegmentation of anatomical structures from Cardiac Magnetic Resonance (CMR) is central to the non-invasive quantitative assessment of cardiac function and structure. Anatomical variability, imaging heterogeneity and cardiac dynamics challenge the automation of this task. Deep learning (DL) approaches have taken over the field of automatic segmentation in recent years, however they are limited by data availability and the additional variability introduced by differences in scanners and protocols. In this work, we propose a 2-step fully automated pipeline to segment CMR images, based on DL encoder-decoder frameworks, and we explore two domain adaptation techniques, domain adversarial training and iterative domain unlearning, to overcome the imaging heterogeneity limitations. We evaluate our methods on the MICCAI 2020 Multi-Centre, Multi-Vendor & Multi-Disease Cardiac Image Segmentation Challenge training and validation datasets. The results show the improvement in performance produced by domain adaptation models, especially among the seen vendors. Finally, we build an ensemble of baseline and domain adapted networks, that reported state-of-art mean Dice scores of 0.912, 0.857 and 0.861 for left ventricle (LV) cavity, LV myocardium and right ventricle cavity, respectively, on the externally validated Challenge dataset, including several unseen vendors, centers and cardiac pathologies. |
| spellingShingle | Corral Acero, J Sundaresan, V Dinsdale, N Grau, V Jenkinson, M A 2-step deep learning method with domain adaptation for multi-centre, multi-vendor and multi-disease cardiac magnetic resonance segmentation |
| title | A 2-step deep learning method with domain adaptation for multi-centre, multi-vendor and multi-disease cardiac magnetic resonance segmentation |
| title_full | A 2-step deep learning method with domain adaptation for multi-centre, multi-vendor and multi-disease cardiac magnetic resonance segmentation |
| title_fullStr | A 2-step deep learning method with domain adaptation for multi-centre, multi-vendor and multi-disease cardiac magnetic resonance segmentation |
| title_full_unstemmed | A 2-step deep learning method with domain adaptation for multi-centre, multi-vendor and multi-disease cardiac magnetic resonance segmentation |
| title_short | A 2-step deep learning method with domain adaptation for multi-centre, multi-vendor and multi-disease cardiac magnetic resonance segmentation |
| title_sort | 2 step deep learning method with domain adaptation for multi centre multi vendor and multi disease cardiac magnetic resonance segmentation |
| work_keys_str_mv | AT corralaceroj a2stepdeeplearningmethodwithdomainadaptationformulticentremultivendorandmultidiseasecardiacmagneticresonancesegmentation AT sundaresanv a2stepdeeplearningmethodwithdomainadaptationformulticentremultivendorandmultidiseasecardiacmagneticresonancesegmentation AT dinsdalen a2stepdeeplearningmethodwithdomainadaptationformulticentremultivendorandmultidiseasecardiacmagneticresonancesegmentation AT grauv a2stepdeeplearningmethodwithdomainadaptationformulticentremultivendorandmultidiseasecardiacmagneticresonancesegmentation AT jenkinsonm a2stepdeeplearningmethodwithdomainadaptationformulticentremultivendorandmultidiseasecardiacmagneticresonancesegmentation AT corralaceroj 2stepdeeplearningmethodwithdomainadaptationformulticentremultivendorandmultidiseasecardiacmagneticresonancesegmentation AT sundaresanv 2stepdeeplearningmethodwithdomainadaptationformulticentremultivendorandmultidiseasecardiacmagneticresonancesegmentation AT dinsdalen 2stepdeeplearningmethodwithdomainadaptationformulticentremultivendorandmultidiseasecardiacmagneticresonancesegmentation AT grauv 2stepdeeplearningmethodwithdomainadaptationformulticentremultivendorandmultidiseasecardiacmagneticresonancesegmentation AT jenkinsonm 2stepdeeplearningmethodwithdomainadaptationformulticentremultivendorandmultidiseasecardiacmagneticresonancesegmentation |