TY - JOUR
T1 - Development of a Deep-Learning Algorithm for Small Bowel-Lesion Detection and a Study of the Improvement in the False-Positive Rate
AU - Hosoe, Naoki
AU - Horie, Tomofumi
AU - Tojo, Anna
AU - Sakurai, Hinako
AU - Hayashi, Yukie
AU - Kamiya, Kenji Jose Luis Limpias
AU - Sujino, Tomohisa
AU - Takabayashi, Kaoru
AU - Ogata, Haruhiko
AU - Kanai, Takanori
N1 - Publisher Copyright:
© 2022 by the authors. Licensee MDPI, Basel, Switzerland.
PY - 2022/7/1
Y1 - 2022/7/1
N2 - Deep learning has recently been gaining attention as a promising technology to improve the identification of lesions, and deep-learning algorithms for lesion detection have been actively developed in small-bowel capsule endoscopy (SBCE). We developed a detection algorithm for abnormal findings by deep learning (convolutional neural network) the SBCE imaging data of 30 cases with abnormal findings. To enable the detection of a wide variety of abnormal findings, the training data were balanced to include all major findings identified in SBCE (bleeding, angiodysplasia, ulceration, and neoplastic lesions). To reduce the false-positive rate, “findings that may be responsible for hemorrhage” and “findings that may require therapeutic intervention” were extracted from the images of abnormal findings and added to the training dataset. For the performance evaluation, the sensitivity and the specificity were calculated using 271 detectable findings in 35 cases. The sensitivity was calculated using 68,494 images of non-abnormal findings. The sensitivity and specificity were 93.4% and 97.8%, respectively. The average number of images detected by the algorithm as having abnormal findings was 7514. We developed an image-reading support system using deep learning for SBCE and obtained a good detection performance.
AB - Deep learning has recently been gaining attention as a promising technology to improve the identification of lesions, and deep-learning algorithms for lesion detection have been actively developed in small-bowel capsule endoscopy (SBCE). We developed a detection algorithm for abnormal findings by deep learning (convolutional neural network) the SBCE imaging data of 30 cases with abnormal findings. To enable the detection of a wide variety of abnormal findings, the training data were balanced to include all major findings identified in SBCE (bleeding, angiodysplasia, ulceration, and neoplastic lesions). To reduce the false-positive rate, “findings that may be responsible for hemorrhage” and “findings that may require therapeutic intervention” were extracted from the images of abnormal findings and added to the training dataset. For the performance evaluation, the sensitivity and the specificity were calculated using 271 detectable findings in 35 cases. The sensitivity was calculated using 68,494 images of non-abnormal findings. The sensitivity and specificity were 93.4% and 97.8%, respectively. The average number of images detected by the algorithm as having abnormal findings was 7514. We developed an image-reading support system using deep learning for SBCE and obtained a good detection performance.
KW - angioectasia
KW - deep learning
KW - obscure gastrointestinal bleeding
KW - tumor
KW - video-capsule endoscopy
UR - http://www.scopus.com/inward/record.url?scp=85132831603&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85132831603&partnerID=8YFLogxK
U2 - 10.3390/jcm11133682
DO - 10.3390/jcm11133682
M3 - Article
AN - SCOPUS:85132831603
SN - 2077-0383
VL - 11
JO - Journal of Clinical Medicine
JF - Journal of Clinical Medicine
IS - 13
M1 - 3682
ER -