A Multi-Institute Evaluation and Analysis Framework on a Standardized Dataset: Findings From The International Workshop on Osteoarthritis Imaging Knee MRI Segmentation Challenge
Publikation: Bidrag til tidsskrift › Konferenceabstrakt i tidsskrift › Forskning › fagfællebedømt
Standard
A Multi-Institute Evaluation and Analysis Framework on a Standardized Dataset : Findings From The International Workshop on Osteoarthritis Imaging Knee MRI Segmentation Challenge. / Desai, Arjun D.; Caliva, Francesco; Iriondo, Claudia; Khosravan, Naji; Mortazi, Aliasghar; Jambawalikar, Sachin; Torigian, Drew; Ellerman, Jutta; Akcakaya, Mehmet; Bagci, Ulas; Tibrewala, Radhika; Flament, Io; O`Brien, Matthew; Majumdar, Sharmila; Perslev, Mathias; Pai, Akshay; Igel, Christian; Dam, Erik B.; Gaj, Sibaji; Yang, Mingrui; Nakamura, Kunio; Li, Xiaojuan; Deniz, Cem M.; Juras, Vladimir; Regatte, Ravinder; Gold, Garry E.; Hargreaves, Brian A.; Pedoia, Valentina; Chaudhari, Akshay S.
I: Osteoarthritis and Cartilage Open, Bind 28, Nr. Suppl. 1, 2020, s. 5304-5305.Publikation: Bidrag til tidsskrift › Konferenceabstrakt i tidsskrift › Forskning › fagfællebedømt
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - ABST
T1 - A Multi-Institute Evaluation and Analysis Framework on a Standardized Dataset
AU - Desai, Arjun D.
AU - Caliva, Francesco
AU - Iriondo, Claudia
AU - Khosravan, Naji
AU - Mortazi, Aliasghar
AU - Jambawalikar, Sachin
AU - Torigian, Drew
AU - Ellerman, Jutta
AU - Akcakaya, Mehmet
AU - Bagci, Ulas
AU - Tibrewala, Radhika
AU - Flament, Io
AU - O`Brien, Matthew
AU - Majumdar, Sharmila
AU - Perslev, Mathias
AU - Pai, Akshay
AU - Igel, Christian
AU - Dam, Erik B.
AU - Gaj, Sibaji
AU - Yang, Mingrui
AU - Nakamura, Kunio
AU - Li, Xiaojuan
AU - Deniz, Cem M.
AU - Juras, Vladimir
AU - Regatte, Ravinder
AU - Gold, Garry E.
AU - Hargreaves, Brian A.
AU - Pedoia, Valentina
AU - Chaudhari, Akshay S.
N1 - Submitted to Radiology: Artificial Intelligence
PY - 2020
Y1 - 2020
N2 - Purpose: To organize a knee MRI segmentation challenge for characterizing the semantic and clinical efficacy of automatic segmentation methods relevant for monitoring osteoarthritis progression. Methods: A dataset partition consisting of 3D knee MRI from 88 subjects at two timepoints with ground-truth articular (femoral, tibial, patellar) cartilage and meniscus segmentations was standardized. Challenge submissions and a majority-vote ensemble were evaluated using Dice score, average symmetric surface distance, volumetric overlap error, and coefficient of variation on a hold-out test set. Similarities in network segmentations were evaluated using pairwise Dice correlations. Articular cartilage thickness was computed per-scan and longitudinally. Correlation between thickness error and segmentation metrics was measured using Pearson's coefficient. Two empirical upper bounds for ensemble performance were computed using combinations of model outputs that consolidated true positives and true negatives. Results: Six teams (T1-T6) submitted entries for the challenge. No significant differences were observed across all segmentation metrics for all tissues (p=1.0) among the four top-performing networks (T2, T3, T4, T6). Dice correlations between network pairs were high (>0.85). Per-scan thickness errors were negligible among T1-T4 (p=0.99) and longitudinal changes showed minimal bias (<0.03mm). Low correlations (<0.41) were observed between segmentation metrics and thickness error. The majority-vote ensemble was comparable to top performing networks (p=1.0). Empirical upper bound performances were similar for both combinations (p=1.0). Conclusion: Diverse networks learned to segment the knee similarly where high segmentation accuracy did not correlate to cartilage thickness accuracy. Voting ensembles did not outperform individual networks but may help regularize individual models.
AB - Purpose: To organize a knee MRI segmentation challenge for characterizing the semantic and clinical efficacy of automatic segmentation methods relevant for monitoring osteoarthritis progression. Methods: A dataset partition consisting of 3D knee MRI from 88 subjects at two timepoints with ground-truth articular (femoral, tibial, patellar) cartilage and meniscus segmentations was standardized. Challenge submissions and a majority-vote ensemble were evaluated using Dice score, average symmetric surface distance, volumetric overlap error, and coefficient of variation on a hold-out test set. Similarities in network segmentations were evaluated using pairwise Dice correlations. Articular cartilage thickness was computed per-scan and longitudinally. Correlation between thickness error and segmentation metrics was measured using Pearson's coefficient. Two empirical upper bounds for ensemble performance were computed using combinations of model outputs that consolidated true positives and true negatives. Results: Six teams (T1-T6) submitted entries for the challenge. No significant differences were observed across all segmentation metrics for all tissues (p=1.0) among the four top-performing networks (T2, T3, T4, T6). Dice correlations between network pairs were high (>0.85). Per-scan thickness errors were negligible among T1-T4 (p=0.99) and longitudinal changes showed minimal bias (<0.03mm). Low correlations (<0.41) were observed between segmentation metrics and thickness error. The majority-vote ensemble was comparable to top performing networks (p=1.0). Empirical upper bound performances were similar for both combinations (p=1.0). Conclusion: Diverse networks learned to segment the knee similarly where high segmentation accuracy did not correlate to cartilage thickness accuracy. Voting ensembles did not outperform individual networks but may help regularize individual models.
KW - eess.IV
KW - cs.CV
M3 - Conference abstract in journal
VL - 28
SP - 5304
EP - 5305
JO - Osteoarthritis and Cartilage Open
JF - Osteoarthritis and Cartilage Open
SN - 2665-9131
IS - Suppl. 1
Y2 - 30 April 2020 through 3 May 2020
ER -
ID: 255780231