Abstract
Background: Manual hippocampus segmentation on structural magnetic resonance images (MRI) is labor intensive and time consuming. This paper presents a semi-automatic hippocampus segmentation method that decreases segmentation time without compromising accuracy. Methods: We present a method that reconstructs sparse delineations into full hippocampal surfaces meshes with a smooth surface reconstruction technique. From fully manual segmented hippocampi in ten subjects with about 20 slice contours, we simulated sparse delineations ranging from 4-10 contours to simulate decreased contouring time by at least half. We compared the original hippocampi with reconstructed hippocampi as well as automatic segmentations obtained from each subjects’ T1 weighted MRI using FSL-FIRST and FreeSurfer. We computed Dice overlap indices, percentage volume differences (PVD) and intra-correlation coefficients (ICC) with manual hippocampus segmentations. Results: For the hippocampi reconstructed from 4 to 10 contours, we obtained high mean dice overlaps, low mean PVDs and high ICCs in the range of 81(±0.03)-91(±0.01)%, 6.85(±5.33)-1.98(±1.63)% and 0.970–0.997 respectively. Reconstructed hippocampi agreed consistently better with manual segmentations than automatic segmentation methods, even when 5 contours were used. Conclusions: We were able to reconstruct hippocampi from a minimum number of contours and maintained high accuracy results that were consistently better than automatic methods. We next need to test this method on a larger scale and validate reproducibility and robustness.
| Original language | English |
|---|---|
| Title of host publication | EMBEC and NBC 2017 - Joint Conference of the European Medical and Biological Engineering Conference EMBEC 2017 and the Nordic-Baltic Conference on Biomedical Engineering and Medical Physics, NBC 2017 |
| Publisher | Springer Verlag |
| Pages | 956-959 |
| Number of pages | 4 |
| Volume | 65 |
| ISBN (Print) | 9789811051210 |
| DOIs | |
| Publication status | Published - 2017 |
| Event | Joint Conference of the European Medical and Biological Engineering Conference, EMBEC 2017 and Nordic-Baltic Conference on Biomedical Engineering and Medical Physics, NBC 2107 - Tampere, Finland Duration: 11 Jun 2017 → 15 Jun 2017 |
Conference
| Conference | Joint Conference of the European Medical and Biological Engineering Conference, EMBEC 2017 and Nordic-Baltic Conference on Biomedical Engineering and Medical Physics, NBC 2107 |
|---|---|
| Country | Finland |
| City | Tampere |
| Period | 11/06/2017 → 15/06/2017 |
Cite this
}
Semi-automatic hippocampus delineation algorithm using surface fairing. / Bartel, Fabian; Vrenken, H.; van Herk, M.; de Ruiter, M. B.; Belderbos, J.; Hulshof, J.; de Munck, J. C.
EMBEC and NBC 2017 - Joint Conference of the European Medical and Biological Engineering Conference EMBEC 2017 and the Nordic-Baltic Conference on Biomedical Engineering and Medical Physics, NBC 2017. Vol. 65 Springer Verlag, 2017. p. 956-959.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review
TY - GEN
T1 - Semi-automatic hippocampus delineation algorithm using surface fairing
AU - Bartel, Fabian
AU - Vrenken, H.
AU - van Herk, M.
AU - de Ruiter, M. B.
AU - Belderbos, J.
AU - Hulshof, J.
AU - de Munck, J. C.
PY - 2017
Y1 - 2017
N2 - Background: Manual hippocampus segmentation on structural magnetic resonance images (MRI) is labor intensive and time consuming. This paper presents a semi-automatic hippocampus segmentation method that decreases segmentation time without compromising accuracy. Methods: We present a method that reconstructs sparse delineations into full hippocampal surfaces meshes with a smooth surface reconstruction technique. From fully manual segmented hippocampi in ten subjects with about 20 slice contours, we simulated sparse delineations ranging from 4-10 contours to simulate decreased contouring time by at least half. We compared the original hippocampi with reconstructed hippocampi as well as automatic segmentations obtained from each subjects’ T1 weighted MRI using FSL-FIRST and FreeSurfer. We computed Dice overlap indices, percentage volume differences (PVD) and intra-correlation coefficients (ICC) with manual hippocampus segmentations. Results: For the hippocampi reconstructed from 4 to 10 contours, we obtained high mean dice overlaps, low mean PVDs and high ICCs in the range of 81(±0.03)-91(±0.01)%, 6.85(±5.33)-1.98(±1.63)% and 0.970–0.997 respectively. Reconstructed hippocampi agreed consistently better with manual segmentations than automatic segmentation methods, even when 5 contours were used. Conclusions: We were able to reconstruct hippocampi from a minimum number of contours and maintained high accuracy results that were consistently better than automatic methods. We next need to test this method on a larger scale and validate reproducibility and robustness.
AB - Background: Manual hippocampus segmentation on structural magnetic resonance images (MRI) is labor intensive and time consuming. This paper presents a semi-automatic hippocampus segmentation method that decreases segmentation time without compromising accuracy. Methods: We present a method that reconstructs sparse delineations into full hippocampal surfaces meshes with a smooth surface reconstruction technique. From fully manual segmented hippocampi in ten subjects with about 20 slice contours, we simulated sparse delineations ranging from 4-10 contours to simulate decreased contouring time by at least half. We compared the original hippocampi with reconstructed hippocampi as well as automatic segmentations obtained from each subjects’ T1 weighted MRI using FSL-FIRST and FreeSurfer. We computed Dice overlap indices, percentage volume differences (PVD) and intra-correlation coefficients (ICC) with manual hippocampus segmentations. Results: For the hippocampi reconstructed from 4 to 10 contours, we obtained high mean dice overlaps, low mean PVDs and high ICCs in the range of 81(±0.03)-91(±0.01)%, 6.85(±5.33)-1.98(±1.63)% and 0.970–0.997 respectively. Reconstructed hippocampi agreed consistently better with manual segmentations than automatic segmentation methods, even when 5 contours were used. Conclusions: We were able to reconstruct hippocampi from a minimum number of contours and maintained high accuracy results that were consistently better than automatic methods. We next need to test this method on a larger scale and validate reproducibility and robustness.
KW - FreeSurfer
KW - FSL-FIRST
KW - Hippocampus
KW - Semi-automatic delineation
KW - Surface fairing
UR - http://www.scopus.com/inward/record.url?scp=85021704991&partnerID=8YFLogxK
U2 - 10.1007/978-981-10-5122-7_239
DO - 10.1007/978-981-10-5122-7_239
M3 - Conference contribution
SN - 9789811051210
VL - 65
SP - 956
EP - 959
BT - EMBEC and NBC 2017 - Joint Conference of the European Medical and Biological Engineering Conference EMBEC 2017 and the Nordic-Baltic Conference on Biomedical Engineering and Medical Physics, NBC 2017
PB - Springer Verlag
ER -