Final published version
Other version
Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Conference contribution/Paper › peer-review
Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Conference contribution/Paper › peer-review
}
TY - GEN
T1 - Automatic detection and identification of retinal vessel junctions in colour fundus photography
AU - Pratt, Harry
AU - Williams, Bryan M.
AU - Ku, Jae
AU - Coenen, Frans
AU - Zheng, Yalin
PY - 2017/6/22
Y1 - 2017/6/22
N2 - The quantitative analysis of retinal blood vessels is important for the management of vascular disease and tackling problems such as locating blood clots. Such tasks are hampered by the inability to accurately trace back problems along vessels to the source. This is due to the unresolved challenge of distinguishing automatically between vessel branchings and vessel crossings. In this paper, we present a new technique for tackling this challenging problem by developing a convolutional neural network approach for first locating vessel junctions and then classifying them as either branchings or crossings. We achieve a high accuracy of 94% for junction detection and 88% for classification. Combined with work in segmentation, this method has the potential to facilitate automated localisation of blood clots and other disease symptoms leading to improved management of eye disease through aiding or replacing a clinicians diagnosis.
AB - The quantitative analysis of retinal blood vessels is important for the management of vascular disease and tackling problems such as locating blood clots. Such tasks are hampered by the inability to accurately trace back problems along vessels to the source. This is due to the unresolved challenge of distinguishing automatically between vessel branchings and vessel crossings. In this paper, we present a new technique for tackling this challenging problem by developing a convolutional neural network approach for first locating vessel junctions and then classifying them as either branchings or crossings. We achieve a high accuracy of 94% for junction detection and 88% for classification. Combined with work in segmentation, this method has the potential to facilitate automated localisation of blood clots and other disease symptoms leading to improved management of eye disease through aiding or replacing a clinicians diagnosis.
KW - Convolutional neural networks
KW - Retinal imaging
KW - Retinal vessels fundus photography
KW - Vessel classification
U2 - 10.1007/978-3-319-60964-5_3
DO - 10.1007/978-3-319-60964-5_3
M3 - Conference contribution/Paper
AN - SCOPUS:85022230686
SN - 9783319609638
T3 - Communications in Computer and Information Science
SP - 27
EP - 37
BT - Medical Image Understanding and Analysis - 21st Annual Conference, MIUA 2017, Proceedings
A2 - Gonzalez-Castro, Victor
A2 - Valdes Hernandez, Maria
PB - Springer-Verlag
T2 - 21st Annual Conference on Medical Image Understanding and Analysis, MIUA 2017
Y2 - 11 July 2017 through 13 July 2017
ER -