TY - GEN
T1 - Spatiotemporal Statistical Model of Anatomical Landmarks on a Human Embryonic Brain
AU - Shinjo, Aoi
AU - Saito, Atsushi
AU - Takakuwa, Tetsuya
AU - Yamada, Shigehito
AU - Hontani, Hidekata
AU - Matsuzoe, Hiroshi
AU - Miyauchi, Shoko
AU - Morooka, Kenichi
AU - Shimizu, Akinobu
N1 - Funding Information:
Acknowledgments. This work was partly supported by MEXT/JSPS KAKENHI Grant Numbers JP26108002, JP18H03255 and JP19K20291.
Publisher Copyright:
© 2019, Springer Nature Switzerland AG.
PY - 2019
Y1 - 2019
N2 - We propose a new method for constructing a spatiotemporal statistical model of the distribution of anatomical landmarks (LMs) of a human embryo. This method exhibits potential for the quantitative assessment of the extent of anomalies and is important in the research of congenital malformations. However, a few of the LMs might not be observed at a specific developmental stage because large morphological deformations exist during the early stages of development. It is difficult for conventional statistical shape analysis methods to handle missing LMs in the training dataset. The basic concept of the proposed method is to conduct statistical analyses by predicting and completing the coordinates of the missing LMs. We demonstrated the proposed method in the context of spatiotemporal statistical modeling of 10 LMs on the brain surface using 37 embryonic subjects with Carnegie stages of 19–22. We conducted a comparative study of the spatiotemporal statistical models between four different prediction methods, and we found that deformable surface mapping was the best prediction method in terms of model generalization and specificity.
AB - We propose a new method for constructing a spatiotemporal statistical model of the distribution of anatomical landmarks (LMs) of a human embryo. This method exhibits potential for the quantitative assessment of the extent of anomalies and is important in the research of congenital malformations. However, a few of the LMs might not be observed at a specific developmental stage because large morphological deformations exist during the early stages of development. It is difficult for conventional statistical shape analysis methods to handle missing LMs in the training dataset. The basic concept of the proposed method is to conduct statistical analyses by predicting and completing the coordinates of the missing LMs. We demonstrated the proposed method in the context of spatiotemporal statistical modeling of 10 LMs on the brain surface using 37 embryonic subjects with Carnegie stages of 19–22. We conducted a comparative study of the spatiotemporal statistical models between four different prediction methods, and we found that deformable surface mapping was the best prediction method in terms of model generalization and specificity.
KW - Embryo
KW - Landmark
KW - Spatiotemporal analysis
KW - Statistical model
UR - http://www.scopus.com/inward/record.url?scp=85075751548&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85075751548&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-32689-0_10
DO - 10.1007/978-3-030-32689-0_10
M3 - Conference contribution
AN - SCOPUS:85075751548
SN - 9783030326883
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 94
EP - 103
BT - Uncertainty for Safe Utilization of Machine Learning in Medical Imaging and Clinical Image-Based Procedures - 1st International Workshop, UNSURE 2019, and 8th International Workshop, CLIP 2019, Held in Conjunction with MICCAI 2019, Proceedings
A2 - Greenspan, Hayit
A2 - Tanno, Ryutaro
A2 - Erdt, Marius
A2 - Arbel, Tal
A2 - Baumgartner, Christian
A2 - Dalca, Adrian
A2 - Sudre, Carole H.
A2 - Wells, William M.
A2 - Drechsler, Klaus
A2 - Erdt, Marius
A2 - Linguraru, Marius George
A2 - Shekhar, Raj
A2 - Oyarzun Laura, Cristina
A2 - Wesarg, Stefan
A2 - González Ballester, Miguel Ángel
PB - Springer
T2 - 1st International Workshop on Uncertainty for Safe Utilization of Machine Learning in Medical Imaging, UNSURE 2019, and the 8th International Workshop on Clinical Image-Based Procedures, CLIP 2019, held in conjunction with 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019
Y2 - 17 October 2019 through 17 October 2019
ER -