To study the similarity between queries and retrieved masses, we design an interactive CBIR (Content-based Image Retrieval) CADx (Computer-aided Diagnosis) system using relevance feedback for the characterization of breast masses in ultrasound (US) images based on radiologists' visual similarity assessment. The CADx system retrieves masses that are similar to query masses from a reference library based on six computer-extracted features that describe the texture, width-to-height, and posterior shadowing of the mass. The k-NN retrieval with Euclidean distance similarity measure and the Rocchio relevance feedback algorithm (RRF) are used. To train the RRF parameters, the similarities of 1891 image pairs from 62 (31 malignant and 31 benign) masses are rated by 3 MQSA (Mammography Quality Standards Act) radiologists using a 9-point scale (9=most similar). The best RRF parameters are chosen based on 3 observer experiments. For testing, 100 independent query masses (49 malignant and 51 benign) and 121 reference masses on 230 (79 malignant and 151 benign) images were collected. Three radiologists rated the similarity between the query masses and the computer-retrieved masses. Average similarity ratings without and with RRF were 5.39 and 5.64 for the training set and 5.78 and 6.02 for the test set, respectively. Average AUC values without and with RRF were, respectively, 0.86±0.03 and 0.87±0.03 for the training set and 0.91±0.03 and 0.90±0.03 for the test set. On average, masses retrieved using the CBIR system were moderately similar to the query masses based on radiologists' similarity assessments. RRF improved the similarity of the retrieved masses.
Hyun-chong CHO
Kangwon National University
Lubomir HADJIISKI
University of Michigan
Berkman SAHINER
US Food and Drug Administration (FDA)
Heang-Ping CHAN
University of Michigan
Chintana PARAMAGUL
University of Michigan
Mark HELVIE
University of Michigan
Alexis V. NEES
University of Michigan
Hyun Chin CHO
Gyeongsang National University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Hyun-chong CHO, Lubomir HADJIISKI, Berkman SAHINER, Heang-Ping CHAN, Chintana PARAMAGUL, Mark HELVIE, Alexis V. NEES, Hyun Chin CHO, "A Similarity Study of Interactive Content-Based Image Retrieval Scheme for Classification of Breast Lesions" in IEICE TRANSACTIONS on Information,
vol. E99-D, no. 6, pp. 1663-1670, June 2016, doi: 10.1587/transinf.2015EDP7207.
Abstract: To study the similarity between queries and retrieved masses, we design an interactive CBIR (Content-based Image Retrieval) CADx (Computer-aided Diagnosis) system using relevance feedback for the characterization of breast masses in ultrasound (US) images based on radiologists' visual similarity assessment. The CADx system retrieves masses that are similar to query masses from a reference library based on six computer-extracted features that describe the texture, width-to-height, and posterior shadowing of the mass. The k-NN retrieval with Euclidean distance similarity measure and the Rocchio relevance feedback algorithm (RRF) are used. To train the RRF parameters, the similarities of 1891 image pairs from 62 (31 malignant and 31 benign) masses are rated by 3 MQSA (Mammography Quality Standards Act) radiologists using a 9-point scale (9=most similar). The best RRF parameters are chosen based on 3 observer experiments. For testing, 100 independent query masses (49 malignant and 51 benign) and 121 reference masses on 230 (79 malignant and 151 benign) images were collected. Three radiologists rated the similarity between the query masses and the computer-retrieved masses. Average similarity ratings without and with RRF were 5.39 and 5.64 for the training set and 5.78 and 6.02 for the test set, respectively. Average AUC values without and with RRF were, respectively, 0.86±0.03 and 0.87±0.03 for the training set and 0.91±0.03 and 0.90±0.03 for the test set. On average, masses retrieved using the CBIR system were moderately similar to the query masses based on radiologists' similarity assessments. RRF improved the similarity of the retrieved masses.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2015EDP7207/_p
Copy
@ARTICLE{e99-d_6_1663,
author={Hyun-chong CHO, Lubomir HADJIISKI, Berkman SAHINER, Heang-Ping CHAN, Chintana PARAMAGUL, Mark HELVIE, Alexis V. NEES, Hyun Chin CHO, },
journal={IEICE TRANSACTIONS on Information},
title={A Similarity Study of Interactive Content-Based Image Retrieval Scheme for Classification of Breast Lesions},
year={2016},
volume={E99-D},
number={6},
pages={1663-1670},
abstract={To study the similarity between queries and retrieved masses, we design an interactive CBIR (Content-based Image Retrieval) CADx (Computer-aided Diagnosis) system using relevance feedback for the characterization of breast masses in ultrasound (US) images based on radiologists' visual similarity assessment. The CADx system retrieves masses that are similar to query masses from a reference library based on six computer-extracted features that describe the texture, width-to-height, and posterior shadowing of the mass. The k-NN retrieval with Euclidean distance similarity measure and the Rocchio relevance feedback algorithm (RRF) are used. To train the RRF parameters, the similarities of 1891 image pairs from 62 (31 malignant and 31 benign) masses are rated by 3 MQSA (Mammography Quality Standards Act) radiologists using a 9-point scale (9=most similar). The best RRF parameters are chosen based on 3 observer experiments. For testing, 100 independent query masses (49 malignant and 51 benign) and 121 reference masses on 230 (79 malignant and 151 benign) images were collected. Three radiologists rated the similarity between the query masses and the computer-retrieved masses. Average similarity ratings without and with RRF were 5.39 and 5.64 for the training set and 5.78 and 6.02 for the test set, respectively. Average AUC values without and with RRF were, respectively, 0.86±0.03 and 0.87±0.03 for the training set and 0.91±0.03 and 0.90±0.03 for the test set. On average, masses retrieved using the CBIR system were moderately similar to the query masses based on radiologists' similarity assessments. RRF improved the similarity of the retrieved masses.},
keywords={},
doi={10.1587/transinf.2015EDP7207},
ISSN={1745-1361},
month={June},}
Copy
TY - JOUR
TI - A Similarity Study of Interactive Content-Based Image Retrieval Scheme for Classification of Breast Lesions
T2 - IEICE TRANSACTIONS on Information
SP - 1663
EP - 1670
AU - Hyun-chong CHO
AU - Lubomir HADJIISKI
AU - Berkman SAHINER
AU - Heang-Ping CHAN
AU - Chintana PARAMAGUL
AU - Mark HELVIE
AU - Alexis V. NEES
AU - Hyun Chin CHO
PY - 2016
DO - 10.1587/transinf.2015EDP7207
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E99-D
IS - 6
JA - IEICE TRANSACTIONS on Information
Y1 - June 2016
AB - To study the similarity between queries and retrieved masses, we design an interactive CBIR (Content-based Image Retrieval) CADx (Computer-aided Diagnosis) system using relevance feedback for the characterization of breast masses in ultrasound (US) images based on radiologists' visual similarity assessment. The CADx system retrieves masses that are similar to query masses from a reference library based on six computer-extracted features that describe the texture, width-to-height, and posterior shadowing of the mass. The k-NN retrieval with Euclidean distance similarity measure and the Rocchio relevance feedback algorithm (RRF) are used. To train the RRF parameters, the similarities of 1891 image pairs from 62 (31 malignant and 31 benign) masses are rated by 3 MQSA (Mammography Quality Standards Act) radiologists using a 9-point scale (9=most similar). The best RRF parameters are chosen based on 3 observer experiments. For testing, 100 independent query masses (49 malignant and 51 benign) and 121 reference masses on 230 (79 malignant and 151 benign) images were collected. Three radiologists rated the similarity between the query masses and the computer-retrieved masses. Average similarity ratings without and with RRF were 5.39 and 5.64 for the training set and 5.78 and 6.02 for the test set, respectively. Average AUC values without and with RRF were, respectively, 0.86±0.03 and 0.87±0.03 for the training set and 0.91±0.03 and 0.90±0.03 for the test set. On average, masses retrieved using the CBIR system were moderately similar to the query masses based on radiologists' similarity assessments. RRF improved the similarity of the retrieved masses.
ER -