Automated Characterization of Atheromatous Plaque in Intravascular Ultrasound Images Using Neuro Fuzzy Classifier

Abstract

Abstract-The medical imaging field has grown significantly in recent years and demands high accuracy since it deals with human life. The idea is to reduce human error as much as possible by assisting physicians and radiologists with some automatic techniques. The use of artificial intelligent techniques has shown great potential in this field. Hence, in this paper the neuro fuzzy classifier is applied for the automated characterization of atheromatous plaque to identify the fibrotic, lipidic and calcified tissues in Intravascular Ultrasound images (IVUS) which is designed using sixteen inputs, corresponds to sixteen pixels of instantaneous scanning matrix, one output that tells whether the pixel under consideration is Fibrotic, Lipidic, Calcified or Normal pixel. The classification performance was evaluated in terms of sensitivity, specificity and accuracy and the results confirmed that the proposed system has potential in detecting the respective plaque with the average accuracy of 98.9%.

Recent Trends in Artificial Intelligence-Assisted Coronary Atherosclerotic Plaque Characterization

“…A high specificity of 98.5% was achieved for automated segmentation of calcified plaque using a Bayesian classifier for a dataset that comprised 60 IVUS images from seven patients [ 80 ]. IVUS outperformed other modalities with an average accuracy of 98.9% for the classification of atheromatous plaque, which confirmed its clinical utility in plaque characterization [ 15 ]. Furthermore, a PPV of 96.69% was achieved in coronary and carotid plaque characterization using the SVM classifier implemented on 2685 IVUS images obtained from 15 patients [ 99 ].…”

“…It is also dependent on image quality, which is easily affected by speckle noise. These reasons motivate the development of computer aided diagnosis (CAD) systems for automated image processing as well as coronary plaque identification and characterization on both invasive [ 15 ] and noninvasive image readouts [ 16 ]. With the help of CAD, image quality can be improved, which helps in the accurate interpretation of results.…”

Systematic mapping study on diagnosis of vulnerable plaque

“…The penetrating nature of IVUS imaging method visualizes the blood flow in the artery wall. IVUS image analyses the internal morphology of coronary artery by means of the microscopic visualization of the tissue structures [74]. Development of IVUS imaging system provides assessment of plaque for in vivo histological image during the surgery operation.…”

“…Their method evaluated the consistency of tissues behind the calcified arc. Selvathi, Emimal [74] proposed a neuro-fuzzy model for automatic characterization of atheromatous plaque. Neuro fuzzy classifier was trained by means of the intensity values of overlapping window and classified each pixel into lipidic, fibrotic, calcified, or normal pixel.…”

“…Selvathi, Emimal [74] 2012 Neuro-fuzzy • Integration of K-means clustering and neuro-fuzzy method to reduce the uncertainty.…”