Kamil Říha

Novel Method for Localization of Common Carotid Artery Transverse Section in Ultrasound Images Using Modified Viola-Jones Detector

This article describes a novel method for highly accurate and effective localization of the transverse section of the carotis comunis artery in ultrasound images. The method has a high success rate, approximately 97%. Unlike analytical methods based on geometric descriptions of the object sought, the method proposed here can cover a large area of shape variation of the artery under study, which normally occurs during examinations as a result of the pressure on the examined tissue, tilt of the probe, setup of the sonographic device, and other factors. This method shows great promise in automating the process of determining circulatory system parameters in the non-invasive clinical diagnostics of cardiovascular diseases. The method employs a Viola-Jones detector that has been specially adapted for efficient detection of transverse sections of the carotid artery. This algorithm is trained on a set of labeled images using the AdaBoost algorithm, Haar-like features and the Matthews coefficient. The training algorithm of the artery detector was modified using evolutionary algorithms. The method for training a cascade of classifiers achieves on a small number of positive and negative training data samples (about 500 images) a high success rate in a computational time that allows implementation of the detector in real time. Testing was performed on images of different patients for whom different ultrasonic instruments were used under different conditions (settings) so that the algorithm developed is applicable in general radiologic practice.

Analysis of Carotid Artery Transverse Sections in Long Ultrasound Video Sequences

Examination of the common carotid artery (CCA) based on an ultrasound video sequence is an effective method for detecting cardiovascular diseases. Here, we propose a video processing method for the automated geometric analysis of CCA transverse sections. By explicitly compensating the parasitic phenomena of global movement and feature drift, our method enables a reliable and accurate estimation of the movement of the arterial wall based on ultrasound sequences of arbitrary length and in situations where state-of-the-art methods fail or are very inaccurate. The method uses a modified Viola-Jones detector and the Hough transform to localize the artery in the image. Then it identifies dominant scatterers, also known as interest points (IPs), whose positions are tracked by means of the pyramidal Lucas-Kanade method. Robustness to global movement and feature drift is achieved by a detection of global movement and subsequent IP re-initialization, as well as an adaptive removal and addition of IPs. The performance of the proposed method is evaluated using simulated and real ultrasound video sequences. Using the Harris detector for IP detection, we obtained an overall root-mean-square error, averaged over all the simulated sequences, of 2.16 ± 1.18 px. The computational complexity of our method is compatible with real-time operation; the runtime is about 30-70 ms/frame for sequences with a spatial resolution of up to 490 × 490 px. We expect that in future clinical practice, our method will be instrumental for non-invasive early-stage diagnosis of atherosclerosis and other cardiovascular diseases.

Manifold regularized multiple kernel learning with Hellinger distance

The aim of this paper is to solve the problem of unsupervised manifold regularization being used under supervised classification circumstance. This paper not only considers that the manifold information of data can provide useful information but also proposes a supervised method to compute the Laplacian graph by using the label information and the Hellinger distance for a comprehensive evaluation of the similarity of data samples. Meanwhile, multi-source or complex data is increasing nowadays. It is desirable to learn from several kernels that are adaptable and flexible to deal with this type of data. Therefore, our classifier is based on multiple kernel learning, and the proposed approach to supervised classification is a multiple kernel model with manifold regularization to incorporate intrinsic geometrical information. Finally, a classifier that minimizes the testing error and considers the geometrical structure of data is put forward. The results of experiments with other methods show the effectiveness of the proposed model and computing the inner potential geometrical information is useful for classification.

Towards an Optimal Interest Point Detector for Measurements in Ultrasound Images

Abstract This paper focuses on the comparison of different interest point detectors and their utilization for measurements in ultrasound (US) images. Certain medical examinations are based on speckle tracking which strongly relies on features that can be reliably tracked frame to frame. Only significant features (interest points) resistant to noise and brightness changes within US images are suitable for accurate long-lasting tracking. We compare three interest point detectors - Harris-Laplace, Difference of Gaussian (DoG) and Fast Hessian - and identify the most suitable one for use in US images on the basis of an objective criterion. Repeatability rate is assumed to be an objective quality measure for comparison. We have measured repeatability in images corrupted by different types of noise (speckle noise, Gaussian noise) and for changes in brightness. The Harris-Laplace detector outperformed its competitors and seems to be a sound option when choosing a suitable interest point detector for US images. However, it has to be noted that Fast Hessian and DoG detectors achieved better results in terms of processing speed.