Introduction: This paper presents the problem of automatic measurement of the\r\niridocorneal angle in tomographic images of the anterior segment of the eye. It\r\nincludes the results of the comparison of well-known methods for measuring the\r\niridocorneal angle with new methods, proposed in this paper. All these methods\r\nconcern tomographic image analysis and processing.\r\nMaterial and method: In total, approximately 100�000 tomographic images (from\r\nabout 6�000 patients) were analysed. They were obtained using two devices: SOCT\r\nCopernicus (Optopol Tech. SA, Zawiercie, Poland) and Visante OCT (Carl Zeiss\r\nMeditec, Inc, Dublin, California, USA). The patients, aged 12 to 78 years with varying\r\ndegrees of the iridocorneal angle pathology, were from the region of Silesia, Poland.\r\nThe images were in DICOM or RAW formats and analysed in the software developed\r\nby the authors for the purposes of this study.\r\nResults: The results indicate that the measurement method proposed by the\r\nauthors, which is based on the calculation of the minimum distance between the iris\r\nand the cornea in the adopted area, is the most accurate. For this method sensitivity\r\nwas 0.88, specificity 0.89 and the area under the Receiver Operating Characteristic\r\ncurve (AUC) was 0.88. The other known methods for measuring the iridocorneal\r\nangle gave worse results, that is, for example, for the measurement of the distance\r\nbetween the iris and the cornea AUC = 0.87, sensitivity = 0.86 and specificity = 0.71.\r\nFor another well-known method of measuring the iridocorneal angle AUC = 0.77,\r\nsensitivity = 0.82 and specificity = 0.61.\r\nConclusions: The study proved that the proposed method of measuring the\r\nminimum distance between the iris and the cornea within the adopted area is the\r\nmost effective in the classification of the iridocorneal angle in patients with a high\r\ndegree of pathology of all the compared measurement methods based on\r\ntomographic images. However, it requires fully automated measurement.
Loading....