László Lefkovits

Brain Tumor Segmentation with Optimized Random Forest

In this paper we propose and tune a discriminative model based on Random Forest (RF) to accomplish brain tumor segmentation in multimodal MR images. The objective of tuning is meant to establish the optimal parameter values and the most significant constraints of the discriminative model. During the building of the RF classifier, the algorithm evaluates the importance of variables, the proximities between data instances and the generalized error. These three properties of RF are employed to optimize the segmentation framework. At the beginning the RF is tuned for variable importance evaluation, and after that it is used to optimize the segmentation framework. The framework was tested on unseen test images from BRATS. The results obtained are similar to the best ones presented in previous BRATS Challenges.

Automatic Brain Tumor Segmentation in Multispectral MRI Volumetric Records

The aim of this study was to establish a multi-stage fuzzy c-means (FCM) framework for the automatic and accurate detection of brain tumors from multimodal 3D magnetic resonance image data. The proposed algorithm uses prior information at two points of the execution: (1) the clusters of voxels produced by FCM are classified as possibly tumorous and non-tumorous based on data extracted from train volumes; (2) the choice of FCM parameters (e.g. number of clusters, fuzzy exponent) is supported by train data as well. FCM is applied in two stages: the first stage eliminates the most part of non-tumorous tissues from further processing, while the second stage is intended to accurately extract the tumor tissue clusters. The algorithm was tested on 13 selected volumes from the BRATS 2012 database. The achieved accuracy is generally characterized by a Dice score in the range of 0.7 to 0.9. Tests have revealed that increasing the size of the train data set slightly improves the overall accuracy.

Automatic Detection and Segmentation of Brain Tumor Using Random Forest Approach

Early detection is the key of success in the treatment of tumors. Establishing methods that can identify the presence and position of tumors in their early stage is a current great challenge in medical imaging. This study proposes a machine learning solution based on binary decision trees and random forest technique, aiming at the detection and accurate segmentation of brain tumors from multispectral volumetric MRI records. The training and testing of the proposed method uses twelve selected volumes from the BRATS 2012/13 database. Image volumes were preprocessed to extend the feature set with local information of each voxel. Intending to enhance the segmentation accuracy, each detected tumor pixel is validated or discarded according to a criterion based on neighborhood information. A detailed preliminary investigation is carried out in order to identify and enhance the capabilities of random forests trained with information originating from single image records. The achieved accuracy is generally characterized by a Dice score up to 0.9. Recommendation are formulated for the future development of a complex, random forest based tumor detection and segmentation system.

Patch based descriptors for iris recognition

In recent years, local texture analysis methods have gained increasing attention in many areas of image processing and computer vision. The current paper deals with iris features extraction, based on dense descriptors. A dense descriptor captures the local details, pixel by pixel over the complete image. Three different techniques were employed: Local Binary Pattern, Local Phase Quantization and Differential Excitation in order to provide both spatial and frequency information. To evaluate the proposed system, experiments were performed on the UPOL database, by using a linear SVM classification scheme. The results show that the iris micro-texture patterns such as crypts, furrows or pigment spots can be well characterized by patched based descriptors.

Automatic Brain Tumor Segmentation in multispectral MRI volumes using a fuzzy c-means cascade algorithm

The aim of this study was to establish a multi-stage fuzzy c-means (FCM) framework for the automatic and accurate detection of brain tumors from multimodal 3D magnetic resonance image data. The proposed algorithm uses prior information at two points of the execution: (1) the clusters of voxels produced by FCM are classified as possibly tumorous and non-tumorous based on data extracted from train volumes; (2) the choice of FCM parameters (e.g. number of clusters, fuzzy exponent) is supported by train data as well. FCM is applied in two stages: the first stage eliminates the most part of non-tumorous tissues from further processing, while the second stage is intended to accurately extract the tumor tissue clusters. The algorithm was tested on six selected volumes from the BRATS 2012 database. The achieved accuracy is generally characterized by a Dice score in the range of 0.7 to 0.9. Tests have revealed that increasing the size of the train data set slightly improves the overall accuracy.

Automatic Brain Tumor Segmentation in Multispectral MRI Volumes Using a Random Forest Approach

The development of automatic tumor detection and segmentation procedures enables the computers to preprocess huge sets of MRI records and draw the attention of medical staff upon suspected positive cases. This paper proposes a machine learning solution based on binary decision trees and random forest technique, trained to provide accurate segmentation of brain tumors from multispectral MRI volumes. The current version of our system was trained and tested using all 220 high-grade tumor volumes from the MICCAI BRATS 2016 database. Image records were preprocessed to attenuate the effect of relative intensities in the MRI data, and to extend the feature set with neighborhood information of each voxel. The output of the random forest is also validated for each voxel, according to labels given to neighbor voxels. The achieved accuracy is characterized by an overall mean Dice score of 80.1%, sensitivity 83.1%, and specificity 98.6%. The proposed method is likely to detect all gliomas of 2 cm diameter.

Random forest feature selection approach for image segmentation

In the field of image segmentation, discriminative models have shown promising performance. Generally, every such model begins with the extraction of numerous features from annotated images. Most authors create their discriminative model by using many features without using any selection criteria. A more reliable model can be built by using a framework that selects the important variables, from the point of view of the classification, and eliminates the unimportant once. In this article we present a framework for feature selection and data dimensionality reduction. The methodology is built around the random forest (RF) algorithm and its variable importance evaluation. In order to deal with datasets so large as to be practically unmanageable, we propose an algorithm based on RF that reduces the dimension of the database by eliminating irrelevant features. Furthermore, this framework is applied to optimize our discriminative model for brain tumor segmentation.

Half iris biometric system based on HOG and LIOP

Automatic iris recognition is becoming increasingly important technique for identity management and hence security. In the computer vision domain and mainly in the image recognition applications, the possibility to compare affined images, which could be distinguished just through small differences, is highly important. Using local image descriptors, similar images could be identified, although they are not part of the same scene or they have a changed parameter. Implemented systems show that HOG (Histogram of Oriented Gradients) and LIOP (Local Intensity Order Pattern) descriptors are promising for human recognition based on iris texture. Experimental results are reported on two public databases: UPOL and CASIA_V1.

Gaussian refinements on Gabor filter based patch descriptor

Recent research in object detection tends to put an accent not only on global object methods, but concentrates mostly on object parts and the relationship between them. One of the most widespread part-based object model was proposed by Felzenszwalb et al. [1]. Such systems can be divided into three main parts: the detection of interest points, the development of adequate local descriptors and the object model. This article deals with the most important phase: the elaboration of local descriptors. We have therefore created a patch descriptor based on two-dimensional Gabor filters. The idea of the descriptor thus developed is to select only a few from the multitude of definable Gabor filters: those most adequate for a given object part. In our previous works, we designed a response-map that played the role of the local descriptor, based on the above-mentioned filters and the GentleBoost learning algorithm [2] or the SVM classification method [3]. In this paper we propose an improvement to the filter selection process which considers not only the magnitude of the complex Gabor filter responses, but the real and imaginary parts and their statistical distribution as well. For this purpose, we have created an RBF Neural Network able to learn the statistical distribution of Gabor filter responses. This network improved the selection procedure of the most suitable filters for a given image patch. The idea of using the RBF NN has been suggested by several authors [4, 5, 6] whose systems are based on the Gaussian distribution of Gabor filters. In conclusion, we have compared the above-mentioned three methods – GentleBoost, SVM and RBFNN – and have deduced that the combination of Gaussians characterizes the patch better than just the magnitude value of the complex responses.

Biometric identification based on feature fusion with PCA and SVM

Biometric identification is gaining ground compared to traditional identification methods. Many biometric measurements may be used for secure human identification. The most reliable among them is the iris pattern because of its uniqueness, stability, unforgeability and inalterability over time. The approach presented in this paper is a fusion of different feature descriptor methods such as HOG, LIOP, LBP, used for extracting iris texture information. The classifiers obtained through the SVM and PCA methods demonstrate the effectiveness of our system applied to one and both irises. The performances measured are highly accurate and foreshadow a fusion system with a rate of identification approaching 100% on the UPOL database.