Multivendor Evaluation of Automated MRI Postprocessing of Biventricular Size and Function for Children With and Without Congenital Heart Defects.
artificial intelligence
automated segmentation
automatic segmentation
congenital heart disease
pediatric
reproducibility
volumes and masses
Journal
Journal of magnetic resonance imaging : JMRI
ISSN: 1522-2586
Titre abrégé: J Magn Reson Imaging
Pays: United States
ID NLM: 9105850
Informations de publication
Date de publication:
09 2023
09 2023
Historique:
revised:
05
12
2022
received:
04
08
2022
accepted:
05
12
2022
medline:
8
8
2023
pubmed:
28
12
2022
entrez:
27
12
2022
Statut:
ppublish
Résumé
Manually segmenting cardiac structures is time-consuming and produces variability in MRI assessments. Automated segmentation could solve this. However, current software is developed for adults without congenital heart defects (CHD). To evaluate automated segmentation of left ventricle (LV) and right ventricle (RV) for pediatric MRI studies. Retrospective comparative study. Twenty children per group of: healthy children, LV-CHD, tetralogy of Fallot (ToF), and univentricular CHD, aged 11.7 [8.9-16.0], 14.2 [10.6-15.7], 14.6 [11.6-16.4], and 12.2 [10.2-14.9] years, respectively. Balanced steady-state free precession at 1.5 T. Biventricular volumes and masses were calculated from a short-axis stack of images, which were segmented manually and using two fully automated software suites (Medis Suite 3.2, Medis, Leiden, the Netherlands and SuiteHeart 5.0, Neosoft LLC, Pewaukee, USA). Fully automated segmentations were manually adjusted to provide two further sets of segmentations. Fully automated and adjusted automated segmentation were compared to manual segmentation. Segmentation times and reproducibility for each method were assessed. Bland Altman analysis and intraclass correlation coefficients (ICC) were used to compare volumes and masses between methods. Postprocessing times were compared by paired t-tests. Fully automated methods provided good segmentation (ICC > 0.90 compared to manual segmentation) for the LV in the healthy and left-sided CHD groups (eg LV-EDV difference for healthy children 1.4 ± 11.5 mL, ICC: 0.97, for Medis and 3.0 ± 12.2 mL, ICC: 0.96 for SuiteHeart). Both automated methods gave larger errors (ICC: 0.62-0.94) for the RV in these populations, and for all structures in the ToF and univentricular CHD groups. Adjusted automated segmentation agreed well with manual segmentation (ICC: 0.71-1.00), improved reproducibility and reduced segmentation time in all patient groups, compared to manual segmentation. Fully automated segmentation eliminates observer variability but may produce large errors compared to manual segmentation. Manual adjustments reduce these errors, improve reproducibility, and reduce postprocessing times compared to manual segmentation. Adjusted automated segmentation is reasonable in children with and without CHD. 3. Stage 2.
Sections du résumé
BACKGROUND
Manually segmenting cardiac structures is time-consuming and produces variability in MRI assessments. Automated segmentation could solve this. However, current software is developed for adults without congenital heart defects (CHD).
PURPOSE
To evaluate automated segmentation of left ventricle (LV) and right ventricle (RV) for pediatric MRI studies.
STUDY TYPE
Retrospective comparative study.
POPULATION
Twenty children per group of: healthy children, LV-CHD, tetralogy of Fallot (ToF), and univentricular CHD, aged 11.7 [8.9-16.0], 14.2 [10.6-15.7], 14.6 [11.6-16.4], and 12.2 [10.2-14.9] years, respectively.
SEQUENCE/FIELD STRENGTH
Balanced steady-state free precession at 1.5 T.
ASSESSMENT
Biventricular volumes and masses were calculated from a short-axis stack of images, which were segmented manually and using two fully automated software suites (Medis Suite 3.2, Medis, Leiden, the Netherlands and SuiteHeart 5.0, Neosoft LLC, Pewaukee, USA). Fully automated segmentations were manually adjusted to provide two further sets of segmentations. Fully automated and adjusted automated segmentation were compared to manual segmentation. Segmentation times and reproducibility for each method were assessed.
STATISTICAL TESTS
Bland Altman analysis and intraclass correlation coefficients (ICC) were used to compare volumes and masses between methods. Postprocessing times were compared by paired t-tests.
RESULTS
Fully automated methods provided good segmentation (ICC > 0.90 compared to manual segmentation) for the LV in the healthy and left-sided CHD groups (eg LV-EDV difference for healthy children 1.4 ± 11.5 mL, ICC: 0.97, for Medis and 3.0 ± 12.2 mL, ICC: 0.96 for SuiteHeart). Both automated methods gave larger errors (ICC: 0.62-0.94) for the RV in these populations, and for all structures in the ToF and univentricular CHD groups. Adjusted automated segmentation agreed well with manual segmentation (ICC: 0.71-1.00), improved reproducibility and reduced segmentation time in all patient groups, compared to manual segmentation.
DATA CONCLUSION
Fully automated segmentation eliminates observer variability but may produce large errors compared to manual segmentation. Manual adjustments reduce these errors, improve reproducibility, and reduce postprocessing times compared to manual segmentation. Adjusted automated segmentation is reasonable in children with and without CHD.
EVIDENCE LEVEL
3.
TECHNICAL EFFICACY
Stage 2.
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
794-804Commentaires et corrections
Type : CommentIn
Informations de copyright
© 2022 The Authors. Journal of Magnetic Resonance Imaging published by Wiley Periodicals LLC on behalf of International Society for Magnetic Resonance in Medicine.
Références
Leiner T, Bogaert J, Friedrich MG, et al. SCMR Position Paper (2020) on clinical indications for cardiovascular magnetic resonance. J Cardiovasc Magn Reson. 2020;22(1):76.
Valsangiacomo Buechel ER, Grosse-Wortmann L, Fratz S, et al. Indications for cardiovascular magnetic resonance in children with congenital and acquired heart disease: An expert consensus paper of the Imaging Working Group of the AEPC and the Cardiovascular Magnetic Resonance Section of the EACVI. Eur Heart J Cardiovasc Imaging. 2015;16(3):281-297.
Hedstrom E, Ishida M, Sepulveda-Martinez A, et al. The effect of initial teaching on evaluation of left ventricular volumes by cardiovascular magnetic resonance imaging: Comparison between complete and intermediate beginners and experienced observers. BMC Med Imaging. 2017;17(1):33.
Beerbaum P, Barth P, Kropf S, et al. Cardiac function by MRI in congenital heart disease: Impact of consensus training on interinstitutional variance. J Magn Reson Imaging. 2009;30(5):956-966.
Luijnenburg SE, Robbers-Visser D, Moelker A, Vliegen HW, Mulder BJ, Helbing WA. Intra-observer and interobserver variability of biventricular function, volumes and mass in patients with congenital heart disease measured by CMR imaging. Int J Cardiovasc Imaging. 2010;26(1):57-64.
Shu L-Q, Sun Y-K, Tan L-H, Shu Q, Chang AC. Application of artificial intelligence in pediatrics: Past, present and future. World J Pediatr 2019;15(2):105-108.
Queirós S, Barbosa D, Engvall J, et al. Multi-centre validation of an automatic algorithm for fast 4D myocardial segmentation in cine CMR datasets. Eur Heart J 2016;17(10):1118-1127.
Hautvast GL, Salton CJ, Chuang ML, Breeuwer M, O'Donnell CJ, Manning WJ. Accurate computer-aided quantification of left ventricular parameters: Experience in 1555 cardiac magnetic resonance studies from the Framingham Heart Study. Magn Reson Med. 2012;67(5):1478-1486.
Backhaus SJ, Staab W, Steinmetz M, et al. Fully automated quantification of biventricular volumes and function in cardiovascular magnetic resonance: Applicability to clinical routine settings. J Cardiovasc Magn Reson. 2019;21(1):24.
Petersen SE, Aung N, Sanghvi MM, et al. Reference ranges for cardiac structure and function using cardiovascular magnetic resonance (CMR) in Caucasians from the UK Biobank population cohort. J Cardiovasc Magn Reson. 2017;19(1):18.
Tan LK, Liew YM, Lim E, McLaughlin RA. Convolutional neural network regression for short-axis left ventricle segmentation in cardiac cine MR sequences. Med Image Anal 2017;39:78-86.
Bai W, Sinclair M, Tarroni G, et al. Automated cardiovascular magnetic resonance image analysis with fully convolutional networks. J Cardiovasc Magn Reson 2018;20(1):65.
Arafati A, Hu P, Finn JP, et al. Artificial intelligence in pediatric and adult congenital cardiac MRI: An unmet clinical need. Cardiovasc Diagn Ther 2019;9(S2):S310-S325.
Dewey FE, Rosenthal D, Murphy DJ Jr, Froelicher VF, Ashley EA. Does size matter? Clinical applications of scaling cardiac size and function for body size. Circulation. 2008;117(17):2279-2287.
van der Ven JPG, Sadighy Z, Valsangiacomo Buechel ER, et al. Multicentre reference values for cardiac magnetic resonance imaging derived ventricular size and function for children aged 0-18 years. Eur Heart J Cardiovasc Imaging. 2020;21(1):102-113.
van den Berg J, Hop WC, Strengers JL, et al. Clinical condition at mid-to-late follow-up after transatrial-transpulmonary repair of tetralogy of Fallot. J Thorac Cardiovasc Surg. 2007;133(2):470-477.
Luijnenburg SE, Helbing WA, Moelker A, et al. 5-year serial follow-up of clinical condition and ventricular function in patients after repair of tetralogy of Fallot. Int J Cardiol. 2013;169(6):439-444.
van den Bosch E, Cuypers J, Luijnenburg SE, et al. Ventricular response to dobutamine stress cardiac magnetic resonance imaging is associated with adverse outcome during 8-year follow-up in patients with repaired Tetralogy of Fallot. Eur Heart J Cardiovasc Imaging. 2020;21(9):1039-1046.
van den Bosch E, Bossers SSM, Kamphuis VP, et al. Associations between blood biomarkers, cardiac function, and adverse outcome in a Young Fontan cohort. J Am Heart Assoc. 2021;10(5):e015022.
Bossers SS, Kapusta L, Kuipers IM, et al. Ventricular function and cardiac reserve in contemporary Fontan patients. Int J Cardiol. 2015;196:73-80.
Robbers-Visser D, Boersma E, Helbing WA. Normal biventricular function, volumes, and mass in children aged 8 to 17 years. J Magn Reson Imaging. 2009;29(3):552-559.
Sarikouch S, Peters B, Gutberlet M, et al. Sex-specific pediatric percentiles for ventricular size and mass as reference values for cardiac MRI: Assessment by steady-state free-precession and phase-contrast MRI flow. Circ Cardiovasc Imaging. 2010;3(1):65-76.
Buechel EV, Kaiser T, Jackson C, Schmitz A, Kellenberger CJ. Normal right- and left ventricular volumes and myocardial mass in children measured by steady state free precession cardiovascular magnetic resonance. J Cardiovasc Magn Reson. 2009;11:19.
Robbers-Visser D, Jan Ten Harkel D, Kapusta L, et al. Usefulness of cardiac magnetic resonance imaging combined with low-dose dobutamine stress to detect an abnormal ventricular stress response in children and young adults after Fontan operation at young age. Am J Cardiol. 2008;101(11):1657-1662.
Schulz-Menger J, Bluemke DA, Bremerich J, et al. Standardized image interpretation and post processing in cardiovascular magnetic resonance: Society for Cardiovascular Magnetic Resonance (SCMR) board of trustees task force on standardized post processing. J Cardiovasc Magn Reson. 2013;15:35.
Van Der Ven JPG, Sadighy Z, Valsangiacomo Buechel ER, et al. Multicentre reference values for cardiac magnetic resonance imaging derived ventricular size and function for children aged 0-18 years. Eur Heart J 2020;21(1):102-113.
van der Ven JPG, Alsaied T, Juggan S, et al. Atrial function in Fontan patients assessed by CMR: Relation with exercise capacity and long-term outcomes. Int J Cardiol. 2020;312:56-61.
Shrout PE, Fleiss JL. Intraclass correlations: Uses in assessing rater reliability. Psychol Bull. 1979;86(2):420-428.
Liljequist D, Elfving B, Skavberg RK. Intraclass correlation - A discussion and demonstration of basic features. PLOS one. 2019;14(7):e0219854.
Hammon M, Janka R, Dankerl P, et al. Pediatric cardiac MRI: Automated left-ventricular volumes and function analysis and effects of manual adjustments. Pediatr Radiol. 2015;45(5):651-657.
Kaku K, Takeuchi M, Otani K, et al. Age- and gender-dependency of left ventricular geometry assessed with real-time three-dimensional transthoracic echocardiography. J Am Soc Echocardiogr. 2011;24(5):541-547.
Winter MM, Bernink FJ, Groenink M, et al. Evaluating the systemic right ventricle by CMR: The importance of consistent and reproducible delineation of the cavity. J Cardiovasc Magn Reson. 2008;10:40.
Kawel-Boehm N, Maceira A, Valsangiacomo-Buechel ER, et al. Normal values for cardiovascular magnetic resonance in adults and children. J Cardiovasc Magn Reson. 2015;17:29.
Dice LR. Measures of the amount of ecologic association between species. Ecology 1945;26(3):297-302.
Rockafellar RT, Wets RJ-B. Variational analysis. Berlin Heidelberg: Springer-Verlag; 1998.
Gnanappa GK, Rashid I, Celermajer D, Ayer J, Puranik R. Reproducibility of cardiac magnetic resonance imaging (CMRI)-derived right ventricular parameters in repaired Tetralogy of Fallot (ToF). Heart, Lung Circul 2018;27(3):381-385.