TY - GEN
T1 - Initial estimation of landmark location for automated cephalometric analysis using template matching method
AU - Sari, Indira Pramita
AU - Widayati, Retno
AU - Priaminiarti, Menik
AU - Danudirdjo, Donny
AU - Mengko, Tati Latifah
N1 - Publisher Copyright:
© 2015 IEEE.
PY - 2016/2/8
Y1 - 2016/2/8
N2 - An initial estimator algorithm, aims to localize the position of anatomical landmarks in lateral cephalogram, is presented. This proposed algorithm explores the use of affine transform and template matching to detect three essential reference points: two cephalostat projections (ear rods and nasal-positioner) and menton. The detection of cephalostat projections using template matching method are conducted in the first step. The second step, in which the initial menton positions and small expectation windows for menton are computed, is implemented using affine transform. The exact position of menton is then located within the expectation window using template matching method with sum of squared differences-based similarity measurement in the final step. Using a ten-fold cross validation scheme, we test the algorithm on 106 direct-digital lateral cephalograms. The experimental results show that the initial estimator can efficiently limit the search areas. The obtained detection rate within 4 mm accuracy window is 92.5 percent.
AB - An initial estimator algorithm, aims to localize the position of anatomical landmarks in lateral cephalogram, is presented. This proposed algorithm explores the use of affine transform and template matching to detect three essential reference points: two cephalostat projections (ear rods and nasal-positioner) and menton. The detection of cephalostat projections using template matching method are conducted in the first step. The second step, in which the initial menton positions and small expectation windows for menton are computed, is implemented using affine transform. The exact position of menton is then located within the expectation window using template matching method with sum of squared differences-based similarity measurement in the final step. Using a ten-fold cross validation scheme, we test the algorithm on 106 direct-digital lateral cephalograms. The experimental results show that the initial estimator can efficiently limit the search areas. The obtained detection rate within 4 mm accuracy window is 92.5 percent.
KW - Cephalometry
KW - affine transform
KW - landmarks
KW - template matching
UR - http://www.scopus.com/inward/record.url?scp=84963958961&partnerID=8YFLogxK
U2 - 10.1109/ICICI-BME.2015.7401355
DO - 10.1109/ICICI-BME.2015.7401355
M3 - Conference contribution
AN - SCOPUS:84963958961
T3 - Proceedings - 2015 4th International Conference on Instrumentation, Communications, Information Technology and Biomedical Engineering, ICICI-BME 2015
SP - 159
EP - 162
BT - Proceedings - 2015 4th International Conference on Instrumentation, Communications, Information Technology and Biomedical Engineering, ICICI-BME 2015
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 4th International Conference on Instrumentation, Communications, Information Technology and Biomedical Engineering, ICICI-BME 2015
Y2 - 2 November 2015 through 3 November 2015
ER -