TY - GEN
T1 - Deep Selection
T2 - 23rd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2020
AU - Hachiuma, Ryo
AU - Shimizu, Tomohiro
AU - Saito, Hideo
AU - Kajita, Hiroki
AU - Takatsume, Yoshifumi
N1 - Funding Information:
Acknowledgement. This research was funded by JST-Mirai Program Grant Number JPMJMI19B2, ROIS NII Open Collaborative Research 2020-20S0404, SCOPE of the Ministry of Internal Affairs and Communications and Saitama Prefecture Leading-edge Industry Design Project, Japan. We would like to thank the reviewers for their valuable comments.
Publisher Copyright:
© 2020, Springer Nature Switzerland AG.
PY - 2020
Y1 - 2020
N2 - Recording surgery in operating rooms is an essential task for education and evaluation of medical treatment. However, recording the desired targets, such as the surgery field, surgical tools, or doctor’s hands, is difficult because the targets are heavily occluded during surgery. We use a recording system in which multiple cameras are embedded in the surgical lamp, and we assume that at least one camera is recording the target without occlusion at any given time. As the embedded cameras obtain multiple video sequences, we address the task of selecting the camera with the best view of the surgery. Unlike the conventional method, which selects the camera based on the area size of the surgery field, we propose a deep neural network that predicts the camera selection probability from multiple video sequences by learning the supervision of the expert annotation. We created a dataset in which six different types of plastic surgery are recorded, and we provided the annotation of camera switching. Our experiments show that our approach successfully switched between cameras and outperformed three baseline methods.
AB - Recording surgery in operating rooms is an essential task for education and evaluation of medical treatment. However, recording the desired targets, such as the surgery field, surgical tools, or doctor’s hands, is difficult because the targets are heavily occluded during surgery. We use a recording system in which multiple cameras are embedded in the surgical lamp, and we assume that at least one camera is recording the target without occlusion at any given time. As the embedded cameras obtain multiple video sequences, we address the task of selecting the camera with the best view of the surgery. Unlike the conventional method, which selects the camera based on the area size of the surgery field, we propose a deep neural network that predicts the camera selection probability from multiple video sequences by learning the supervision of the expert annotation. We created a dataset in which six different types of plastic surgery are recorded, and we provided the annotation of camera switching. Our experiments show that our approach successfully switched between cameras and outperformed three baseline methods.
KW - Camera selection
KW - Deep neural network
KW - Surgery recording
UR - http://www.scopus.com/inward/record.url?scp=85092719981&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85092719981&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-59716-0_40
DO - 10.1007/978-3-030-59716-0_40
M3 - Conference contribution
AN - SCOPUS:85092719981
SN - 9783030597153
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 419
EP - 428
BT - Medical Image Computing and Computer Assisted Intervention – MICCAI 2020 - 23rd International Conference, Proceedings
A2 - Martel, Anne L.
A2 - Abolmaesumi, Purang
A2 - Stoyanov, Danail
A2 - Mateus, Diana
A2 - Zuluaga, Maria A.
A2 - Zhou, S. Kevin
A2 - Racoceanu, Daniel
A2 - Joskowicz, Leo
PB - Springer Science and Business Media Deutschland GmbH
Y2 - 4 October 2020 through 8 October 2020
ER -