Using Angio Image Sequence For Coronary Vessel Segmentation


TENEKECİ M. E. , PEHLİVAN H. , GÜMÜŞÇÜ A., Karadag K.

26th IEEE Signal Processing and Communications Applications Conference (SIU), İzmir, Türkiye, 2 - 05 Mayıs 2018 identifier identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Cilt numarası:
  • Doi Numarası: 10.1109/siu.2018.8404754
  • Basıldığı Şehir: İzmir
  • Basıldığı Ülke: Türkiye

Özet

Determination of abnormalities in coronary vessels and vessel segmentation in X-ray coronary angiography has been a challenging problem in recent years. This paper presents a new approach to increase segmentation success in angiography images. With the proposed method, the vessel location due to the change in brightness are detected in the angiography image sequences. In Angiography images Frangi Filter for vessel identification and the OTSU algorithm for automatic threshold detection in image thresholding that was tried to vessel segmentation. The contribution of the proposed method to the success of segmentation is shown wiht the experiments. Sensitivity, specifity and accuracy measures are used to analyze the segmentation success. The results are compared with the manual segmentation of the angio by the specialist. In the experiments, it is seen that the proposed method increases at least 3% of the segmantation success. In the best example, the accuracy has increased to 91,67 deni 93,81.