Assignment Theory-Augmented Neural Network for Dental Arch Labeling
Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Standard
Assignment Theory-Augmented Neural Network for Dental Arch Labeling. / Dascalu, Tudor; Ibragimov, Bulat.
Medical Image Computing and Computer Assisted Intervention – MICCAI 2023 - 26th International Conference, Proceedings. ed. / Hayit Greenspan; Hayit Greenspan; Anant Madabhushi; Parvin Mousavi; Septimiu Salcudean; James Duncan; Tanveer Syeda-Mahmood; Russell Taylor. Springer, 2023. p. 295-304 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol. 14222 LNCS).Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - GEN
T1 - Assignment Theory-Augmented Neural Network for Dental Arch Labeling
AU - Dascalu, Tudor
AU - Ibragimov, Bulat
N1 - Publisher Copyright: © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.
PY - 2023
Y1 - 2023
N2 - Identifying and detecting a set of objects that conform to a structured pattern, but may also have misaligned, missing, or duplicated elements is a difficult task. Dental structures serve as a real-world example of such objects, with high variability in their shape, alignment, and number across different individuals. This study introduces an assignment theory-based approach for recognizing objects based on their positional inter-dependencies. We developed a distance-based anatomical model of teeth consisting of pair-wise displacement vectors and relative positional scores. The dental model was transformed into a cost function for a bipartite graph using a convolutional neural network (CNN). The graph connected candidate tooth labels to the correct tooth labels. We re-framed the problem of determining the optimal tooth labels for a set of candidate labels into the problem of assigning jobs to workers. This approach established a theoretical connection between our task and the field of assignment theory. To optimize the learning process for specific output requirements, we incorporated a loss term based on assignment theory into the objective function. We used the Hungarian method to assign greater importance to the costs returned on the optimal assignment path. The database used in this study consisted of 1200 dental meshes, which included separate upper and lower jaw meshes, collected from 600 patients. The testing set was generated by an indirect segmentation pipeline based on the 3D U-net architecture. To evaluate the ability of the proposed approach to handle anatomical anomalies, we introduced artificial tooth swaps, missing and double teeth. The identification accuracies of the candidate labels were 0.887 for the upper jaw and 0.888 for the lower jaw. The optimal labels predicted by our method improved the identification accuracies to 0.991 for the upper jaw and 0.992 for the lower jaw.
AB - Identifying and detecting a set of objects that conform to a structured pattern, but may also have misaligned, missing, or duplicated elements is a difficult task. Dental structures serve as a real-world example of such objects, with high variability in their shape, alignment, and number across different individuals. This study introduces an assignment theory-based approach for recognizing objects based on their positional inter-dependencies. We developed a distance-based anatomical model of teeth consisting of pair-wise displacement vectors and relative positional scores. The dental model was transformed into a cost function for a bipartite graph using a convolutional neural network (CNN). The graph connected candidate tooth labels to the correct tooth labels. We re-framed the problem of determining the optimal tooth labels for a set of candidate labels into the problem of assigning jobs to workers. This approach established a theoretical connection between our task and the field of assignment theory. To optimize the learning process for specific output requirements, we incorporated a loss term based on assignment theory into the objective function. We used the Hungarian method to assign greater importance to the costs returned on the optimal assignment path. The database used in this study consisted of 1200 dental meshes, which included separate upper and lower jaw meshes, collected from 600 patients. The testing set was generated by an indirect segmentation pipeline based on the 3D U-net architecture. To evaluate the ability of the proposed approach to handle anatomical anomalies, we introduced artificial tooth swaps, missing and double teeth. The identification accuracies of the candidate labels were 0.887 for the upper jaw and 0.888 for the lower jaw. The optimal labels predicted by our method improved the identification accuracies to 0.991 for the upper jaw and 0.992 for the lower jaw.
KW - Assignment theory
KW - Dental instance classification
KW - Multi-object recognition
UR - http://www.scopus.com/inward/record.url?scp=85174721386&partnerID=8YFLogxK
U2 - 10.1007/978-3-031-43898-1_29
DO - 10.1007/978-3-031-43898-1_29
M3 - Article in proceedings
AN - SCOPUS:85174721386
SN - 9783031438974
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 295
EP - 304
BT - Medical Image Computing and Computer Assisted Intervention – MICCAI 2023 - 26th International Conference, Proceedings
A2 - Greenspan, Hayit
A2 - Greenspan, Hayit
A2 - Madabhushi, Anant
A2 - Mousavi, Parvin
A2 - Salcudean, Septimiu
A2 - Duncan, James
A2 - Syeda-Mahmood, Tanveer
A2 - Taylor, Russell
PB - Springer
T2 - 26th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2023
Y2 - 8 October 2023 through 12 October 2023
ER -
ID: 372614591