A texture-based 3D region growing approach for segmentation of ICA through the skull base in CTA

BOZKURT F., KÖSE C., SARI A.

MULTIMEDIA TOOLS AND APPLICATIONS, cilt.79, ss.33253-33278, 2020 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 79
  • Basım Tarihi: 2020
  • Doi Numarası: 10.1007/s11042-020-09690-z
  • Dergi Adı: MULTIMEDIA TOOLS AND APPLICATIONS
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, FRANCIS, ABI/INFORM, Applied Science & Technology Source, Compendex, Computer & Applied Sciences, INSPEC, zbMATH
  • Sayfa Sayıları: ss.33253-33278
  • Anahtar Kelimeler: 3D vessel segmentation, Carotid artery segmentation, Texture-based region growing, Skull base, Computed tomography angiography, LUMEN SEGMENTATION, CEREBRAL-ARTERIES, MRI SEGMENTATION, IMAGES, ALGORITHMS, FEATURES
  • Atatürk Üniversitesi Adresli: Evet

Özet

The internal carotid artery (ICA) segmentation is a complicated task at skull base in computed tomography angiography (CTA) images. The ICA enters into from skull cavity and shows close proximity to bone and surrounding soft tissues. For this reason, there exists a robust intensity overlap between vessels, bone and other surrounding tissues. Thus, these similar objects are not separated properly in images only according to the intensity level. In this paper, a texture-based 3D region growing approach is proposed and applied to the ICA through the skull base. The main contribution of this study is that the method does not ask for an extra computed tomography scan for bone masking. Moreover, the method dynamically sets the segmentation parameters according to texture knowledge. The proposed method was evaluated by the experiments on 15 actual clinical data. The performance evaluations were performed by comparing the automatic outputs with manual segmentations which are done by two radiologist observers. As a result, dice similarity rate of 89% was achieved together with 99% accuracy and 0.32 mm mean surface distance (Msd) for ICA segmentation through the skull base. The results show that the average overlap for the observers are similar. The proposed texture-based approach decreases significantly explosions, over-segmentations and increases rate of area overlap, sensitivity, precision at skull base. Therefore, the method is clinically useful and has potential to segment carotid arteries at skull base efficiently.