Automated localization and quality control of the aorta in cine CMR can significantly accelerate processing of the UK Biobank population data.
<h4>Introduction</h4>Aortic distensibility can be calculated using semi-automated methods to segment the aortic lumen on cine CMR (Cardiovascular Magnetic Resonance) images. However, these methods require visual quality control and manual localization of the region of interest (ROI) of a...
Main Authors: | , , , , , , , , , , , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Public Library of Science (PLoS)
2019-01-01
|
Series: | PLoS ONE |
Online Access: | https://doi.org/10.1371/journal.pone.0212272 |
id |
doaj-4cafb6e716724cfd9ee8f55f7386c1c9 |
---|---|
record_format |
Article |
spelling |
doaj-4cafb6e716724cfd9ee8f55f7386c1c92021-03-04T12:39:03ZengPublic Library of Science (PLoS)PLoS ONE1932-62032019-01-01142e021227210.1371/journal.pone.0212272Automated localization and quality control of the aorta in cine CMR can significantly accelerate processing of the UK Biobank population data.Luca BiasiolliEvan HannElena LukaschukValentina CarapellaJose M PaivaNay AungJennifer J RaynerKonrad WerysKenneth FungHenrike PuchtaMihir M SanghviNiall O MoonRoss J ThomsonKatharine E ThomasMatthew D RobsonVicente GrauSteffen E PetersenStefan NeubauerStefan K Piechnik<h4>Introduction</h4>Aortic distensibility can be calculated using semi-automated methods to segment the aortic lumen on cine CMR (Cardiovascular Magnetic Resonance) images. However, these methods require visual quality control and manual localization of the region of interest (ROI) of ascending (AA) and proximal descending (PDA) aorta, which limit the analysis in large-scale population-based studies. Using 5100 scans from UK Biobank, this study sought to develop and validate a fully automated method to 1) detect and locate the ROIs of AA and PDA, and 2) provide a quality control mechanism.<h4>Methods</h4>The automated AA and PDA detection-localization algorithm followed these steps: 1) foreground segmentation; 2) detection of candidate ROIs by Circular Hough Transform (CHT); 3) spatial, histogram and shape feature extraction for candidate ROIs; 4) AA and PDA detection using Random Forest (RF); 5) quality control based on RF detection probability. To provide the ground truth, overall image quality (IQ = 0-3 from poor to good) and aortic locations were visually assessed by 13 observers. The automated algorithm was trained on 1200 scans and Dice Similarity Coefficient (DSC) was used to calculate the agreement between ground truth and automatically detected ROIs.<h4>Results</h4>The automated algorithm was tested on 3900 scans. Detection accuracy was 99.4% for AA and 99.8% for PDA. Aorta localization showed excellent agreement with the ground truth, with DSC ≥ 0.9 in 94.8% of AA (DSC = 0.97 ± 0.04) and 99.5% of PDA cases (DSC = 0.98 ± 0.03). AA×PDA detection probabilities could discriminate scans with IQ ≥ 1 from those severely corrupted by artefacts (AUC = 90.6%). If scans with detection probability < 0.75 were excluded (350 scans), the algorithm was able to correctly detect and localize AA and PDA in all the remaining 3550 scans (100% accuracy).<h4>Conclusion</h4>The proposed method for automated AA and PDA localization was extremely accurate and the automatically derived detection probabilities provided a robust mechanism to detect low quality scans for further human review. Applying the proposed localization and quality control techniques promises at least a ten-fold reduction in human involvement without sacrificing any accuracy.https://doi.org/10.1371/journal.pone.0212272 |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Luca Biasiolli Evan Hann Elena Lukaschuk Valentina Carapella Jose M Paiva Nay Aung Jennifer J Rayner Konrad Werys Kenneth Fung Henrike Puchta Mihir M Sanghvi Niall O Moon Ross J Thomson Katharine E Thomas Matthew D Robson Vicente Grau Steffen E Petersen Stefan Neubauer Stefan K Piechnik |
spellingShingle |
Luca Biasiolli Evan Hann Elena Lukaschuk Valentina Carapella Jose M Paiva Nay Aung Jennifer J Rayner Konrad Werys Kenneth Fung Henrike Puchta Mihir M Sanghvi Niall O Moon Ross J Thomson Katharine E Thomas Matthew D Robson Vicente Grau Steffen E Petersen Stefan Neubauer Stefan K Piechnik Automated localization and quality control of the aorta in cine CMR can significantly accelerate processing of the UK Biobank population data. PLoS ONE |
author_facet |
Luca Biasiolli Evan Hann Elena Lukaschuk Valentina Carapella Jose M Paiva Nay Aung Jennifer J Rayner Konrad Werys Kenneth Fung Henrike Puchta Mihir M Sanghvi Niall O Moon Ross J Thomson Katharine E Thomas Matthew D Robson Vicente Grau Steffen E Petersen Stefan Neubauer Stefan K Piechnik |
author_sort |
Luca Biasiolli |
title |
Automated localization and quality control of the aorta in cine CMR can significantly accelerate processing of the UK Biobank population data. |
title_short |
Automated localization and quality control of the aorta in cine CMR can significantly accelerate processing of the UK Biobank population data. |
title_full |
Automated localization and quality control of the aorta in cine CMR can significantly accelerate processing of the UK Biobank population data. |
title_fullStr |
Automated localization and quality control of the aorta in cine CMR can significantly accelerate processing of the UK Biobank population data. |
title_full_unstemmed |
Automated localization and quality control of the aorta in cine CMR can significantly accelerate processing of the UK Biobank population data. |
title_sort |
automated localization and quality control of the aorta in cine cmr can significantly accelerate processing of the uk biobank population data. |
publisher |
Public Library of Science (PLoS) |
series |
PLoS ONE |
issn |
1932-6203 |
publishDate |
2019-01-01 |
description |
<h4>Introduction</h4>Aortic distensibility can be calculated using semi-automated methods to segment the aortic lumen on cine CMR (Cardiovascular Magnetic Resonance) images. However, these methods require visual quality control and manual localization of the region of interest (ROI) of ascending (AA) and proximal descending (PDA) aorta, which limit the analysis in large-scale population-based studies. Using 5100 scans from UK Biobank, this study sought to develop and validate a fully automated method to 1) detect and locate the ROIs of AA and PDA, and 2) provide a quality control mechanism.<h4>Methods</h4>The automated AA and PDA detection-localization algorithm followed these steps: 1) foreground segmentation; 2) detection of candidate ROIs by Circular Hough Transform (CHT); 3) spatial, histogram and shape feature extraction for candidate ROIs; 4) AA and PDA detection using Random Forest (RF); 5) quality control based on RF detection probability. To provide the ground truth, overall image quality (IQ = 0-3 from poor to good) and aortic locations were visually assessed by 13 observers. The automated algorithm was trained on 1200 scans and Dice Similarity Coefficient (DSC) was used to calculate the agreement between ground truth and automatically detected ROIs.<h4>Results</h4>The automated algorithm was tested on 3900 scans. Detection accuracy was 99.4% for AA and 99.8% for PDA. Aorta localization showed excellent agreement with the ground truth, with DSC ≥ 0.9 in 94.8% of AA (DSC = 0.97 ± 0.04) and 99.5% of PDA cases (DSC = 0.98 ± 0.03). AA×PDA detection probabilities could discriminate scans with IQ ≥ 1 from those severely corrupted by artefacts (AUC = 90.6%). If scans with detection probability < 0.75 were excluded (350 scans), the algorithm was able to correctly detect and localize AA and PDA in all the remaining 3550 scans (100% accuracy).<h4>Conclusion</h4>The proposed method for automated AA and PDA localization was extremely accurate and the automatically derived detection probabilities provided a robust mechanism to detect low quality scans for further human review. Applying the proposed localization and quality control techniques promises at least a ten-fold reduction in human involvement without sacrificing any accuracy. |
url |
https://doi.org/10.1371/journal.pone.0212272 |
work_keys_str_mv |
AT lucabiasiolli automatedlocalizationandqualitycontroloftheaortaincinecmrcansignificantlyaccelerateprocessingoftheukbiobankpopulationdata AT evanhann automatedlocalizationandqualitycontroloftheaortaincinecmrcansignificantlyaccelerateprocessingoftheukbiobankpopulationdata AT elenalukaschuk automatedlocalizationandqualitycontroloftheaortaincinecmrcansignificantlyaccelerateprocessingoftheukbiobankpopulationdata AT valentinacarapella automatedlocalizationandqualitycontroloftheaortaincinecmrcansignificantlyaccelerateprocessingoftheukbiobankpopulationdata AT josempaiva automatedlocalizationandqualitycontroloftheaortaincinecmrcansignificantlyaccelerateprocessingoftheukbiobankpopulationdata AT nayaung automatedlocalizationandqualitycontroloftheaortaincinecmrcansignificantlyaccelerateprocessingoftheukbiobankpopulationdata AT jenniferjrayner automatedlocalizationandqualitycontroloftheaortaincinecmrcansignificantlyaccelerateprocessingoftheukbiobankpopulationdata AT konradwerys automatedlocalizationandqualitycontroloftheaortaincinecmrcansignificantlyaccelerateprocessingoftheukbiobankpopulationdata AT kennethfung automatedlocalizationandqualitycontroloftheaortaincinecmrcansignificantlyaccelerateprocessingoftheukbiobankpopulationdata AT henrikepuchta automatedlocalizationandqualitycontroloftheaortaincinecmrcansignificantlyaccelerateprocessingoftheukbiobankpopulationdata AT mihirmsanghvi automatedlocalizationandqualitycontroloftheaortaincinecmrcansignificantlyaccelerateprocessingoftheukbiobankpopulationdata AT niallomoon automatedlocalizationandqualitycontroloftheaortaincinecmrcansignificantlyaccelerateprocessingoftheukbiobankpopulationdata AT rossjthomson automatedlocalizationandqualitycontroloftheaortaincinecmrcansignificantlyaccelerateprocessingoftheukbiobankpopulationdata AT katharineethomas automatedlocalizationandqualitycontroloftheaortaincinecmrcansignificantlyaccelerateprocessingoftheukbiobankpopulationdata AT matthewdrobson automatedlocalizationandqualitycontroloftheaortaincinecmrcansignificantlyaccelerateprocessingoftheukbiobankpopulationdata AT vicentegrau automatedlocalizationandqualitycontroloftheaortaincinecmrcansignificantlyaccelerateprocessingoftheukbiobankpopulationdata AT steffenepetersen automatedlocalizationandqualitycontroloftheaortaincinecmrcansignificantlyaccelerateprocessingoftheukbiobankpopulationdata AT stefanneubauer automatedlocalizationandqualitycontroloftheaortaincinecmrcansignificantlyaccelerateprocessingoftheukbiobankpopulationdata AT stefankpiechnik automatedlocalizationandqualitycontroloftheaortaincinecmrcansignificantlyaccelerateprocessingoftheukbiobankpopulationdata |
_version_ |
1714802011716714496 |