Automatic Algorithm for Segmentation of Atherosclerotic Carotid Plaque

Lilla Bonanno, Fabrizio Sottile, Rosella Ciurleo, Giuseppe Di Lorenzo, Daniele Bruschetta, Alessia Bramanti, Giorgio Ascenti, Placido Bramanti, Silvia Marino

Research output: Contribution to journalArticlepeer-review


Background: Carotid atherosclerosis is one of the major causes of stroke. The determination of the intima-media thickness, the identification of carotid atherosclerotic plaque, and the classification of the different stenoses are considered as important parameters for the assessment of atherosclerotic diseases. The aim of this work is to segment the plaques and to allow a better comprehension of carotid atherosclerosis. Methods: We considered 44 subjects, 22 with and 22 without the presence of plaques in the carotid axis, and we applied the snake algorithm. Results: The resulting interclass correlation coefficients (ICCs) were significant for all 3 parameters (mean echogenicity: ICC1 = .78 [95%CI: .55-0.90]; perimeter: ICC2 = .81 [95%CI: .61-0.92]; area: ICC3 = .89 [95%CI: .75-0.95]). The diagnostic accuracy was 82%, with an appropriate cutoff value of 224.5, sensitivity of 79%, and specificity of 85%. Conclusions: In this study, we developed an automatic method to identify the carotid plaque. Our results showed that an automatic system of image segmentation could be used to identify, characterize, and measure atherosclerotic carotid plaques.

Original languageEnglish
Pages (from-to)1-6
Number of pages6
JournalJournal of Stroke and Cerebrovascular Diseases
Publication statusAccepted/In press - Jul 25 2016


  • Automatic segmentation
  • Carotid atherosclerosis
  • Snake algorithm
  • Ultrasound image

ASJC Scopus subject areas

  • Surgery
  • Rehabilitation
  • Clinical Neurology
  • Cardiology and Cardiovascular Medicine

Fingerprint Dive into the research topics of 'Automatic Algorithm for Segmentation of Atherosclerotic Carotid Plaque'. Together they form a unique fingerprint.

Cite this