Joan Massich

Lesion segmentation in breast sonography

Sonography is gaining popularity as an adjunct screening technique for assessing abnormalities in the breast This is particularly true in cases where the subject has dense breast tissue, wherein widespread techniques like Digital Mammography (DM) fail to produce reliable outcomes This article proposes a novel and fully automatic methodology for breast lesion segmentation in B-mode Ultra-Sound (US) images by utilizing region, boundary and shape information to cope up with the inherent artifacts present in US images The proposed approach has been evaluated using a set of sonographic images with accompanying expert-provided ground truth.

Classification of SD-OCT Volumes Using Local Binary Patterns: Experimental Validation for DME Detection

This paper addresses the problem of automatic classification of Spectral Domain OCT (SD-OCT) data for automatic identification of patients with DME versus normal subjects. Optical Coherence Tomography (OCT) has been a valuable diagnostic tool for DME, which is among the most common causes of irreversible vision loss in individuals with diabetes. Here, a classification framework with five distinctive steps is proposed and we present an extensive study of each step. Our method considers combination of various preprocessing steps in conjunction with Local Binary Patterns (LBP) features and different mapping strategies. Using linear and nonlinear classifiers, we tested the developed framework on a balanced cohort of 32 patients. Experimental results show that the proposed method outperforms the previous studies by achieving a Sensitivity (SE) and a Specificity (SP) of 81.2% and 93.7%, respectively. Our study concludes that the 3D features and high-level representation of 2D features using patches achieve the best results. However, the effects of preprocessing are inconsistent with different classifiers and feature configurations.

An anomaly detection approach for the identification of DME patients using spectral domain optical coherence tomography images

This paper proposes a method for automatic classification of spectral domain OCT data for the identification of patients with retinal diseases such as Diabetic Macular Edema (DME). We address this issue as an anomaly detection problem and propose a method that not only allows the classification of the OCT volume, but also allows the identification of the individual diseased B-scans inside the volume. Our approach is based on modeling the appearance of normal OCT images with a Gaussian Mixture Model (GMM) and detecting abnormal OCT images as outliers. The classification of an OCT volume is based on the number of detected outliers. Experimental results with two different datasets show that the proposed method achieves a sensitivity and a specificity of 80% and 93% on the first dataset, and 100% and 80% on the second one. Moreover, the experiments show that the proposed method achieves better classification performance than other recently published works.

Normalization of T2W-MRI prostate images using Rician a priori

Prostate cancer is reported to be the second most frequently diagnosed cancer of men in the world. In practise, diagnosis can be affected by multiple factors which reduces the chance to detect the potential lesions. In the last decades, new imaging techniques mainly based on MRI are developed in conjunction with Computer-Aided Diagnosis (CAD) systems to help radiologists for such diagnosis. CAD systems are usually designed as a sequential process consisting of four stages: pre-processing, segmentation, registration and classification. As a pre-processing, image normalization is a critical and important step of the chain in order to design a robust classifier and overcome the inter-patients intensity variations. However, little attention has been dedicated to the normalization of T2W-Magnetic Resonance Imaging (MRI) prostate images. In this paper, we propose two methods to normalize T2W-MRI prostate images: (i) based on a Rician a priori and (ii) based on a Square-Root Slope Function (SRSF) representation which does not make any assumption regarding the Probability Density Function (PDF) of the data. A comparison with the state-of-the-art methods is also provided. The normalization of the data is assessed by comparing the alignment of the patient PDFs in both qualitative and quantitative manners. In both evaluation, the normalization using Rician a priori outperforms the other state-of-the-art methods.

A boosting approach for prostate cancer detection using multi-parametric MRI

Prostate cancer has been reported as the second most frequently diagnosed men cancers in the world. In the last decades, new imaging techniques based on MRI have been developed in order to improve the diagnosis task of radiologists. In practise, diagnosis can be affected by multiple factors reducing the chance to detect potential lesions. Computer-aided detection and computer-aided diagnosis have been designed to answer to these needs and provide help to radiologists in their daily duties. In this study, we proposed an automatic method to detect prostate cancer from a per voxel manner using 3T multi-parametric Magnetic Resonance Imaging (MRI) and a gradient boosting classifier. The best performances are obtained using all multi-parametric information as well as zonal information. The sensitivity and specificity obtained are 94:7% and 93:0%, respectively and an Area Under Curve (AUC) of 0:968.

SIFT Texture Description for Understanding Breast Ultrasound Images

Texture is a powerful cue for describing structures that show a high degree of similarity in their image intensity patterns. This paper describes the use of Self-Invariant Feature Transform (SIFT), both as low-level and high-level descriptors, applied to differentiate the tissues present in breast US images. For the low-level texture descriptors case, SIFT descriptors are extracted from a regular grid. The high-level texture descriptor is build as a Bag-of-Features (BoF) of SIFT descriptors. Experimental results are provided showing the validity of the proposed approach for describing the tissues in breast US images.

Tackling the Problem of Data Imbalancing for Melanoma Classification

Malignant melanoma is the most dangerous type of skin cancer, yet melanoma is the most treatable kind of cancer when diagnosed at an early stage. In this regard, Computer-Aided Diagnosis systems based on machine learning have been developed to discern melanoma lesions from benign and dysplastic nevi in dermoscopic images. Similar to a large range of real world applications encountered in machine learning, melanoma classification faces the challenge of imbalanced data, where the percentage of melanoma cases in comparison with benign and dysplastic cases is far less. This article analyzes the impact of data balancing strategies at the training step. Subsequently, Over-Sampling (OS) and Under-Sampling (US) are extensively compared in both feature and data space, revealing that NearMiss-2 (NM2) outperform other methods achieving Sensitivity (SE) and Specificity (SP) of 91.2% and 81.7%, respectively. More generally, the reported results highlight that methods based on US or combination of OS and US in feature space outperform the others.

Classification of melanoma lesions using sparse coded features and random forests

Malignant melanoma is the most dangerous type of skin cancer, yet it is the most treatable kind of cancer, conditioned by its early diagnosis which is a challenging task for clinicians and dermatologists. In this regard, CAD systems based on machine learning and image processing techniques are developed to differentiate melanoma lesions from benign and dysplastic nevi using dermoscopic images. Generally, these frameworks are composed of sequential processes: pre-processing, segmentation, and classification. This architecture faces mainly two challenges: (i) each process is complex with the need to tune a set of parameters, and is specific to a given dataset; (ii) the performance of each process depends on the previous one, and the errors are accumulated throughout the framework. In this paper, we propose a framework for melanoma classification based on sparse coding which does not rely on any pre-processing or lesion segmentation. Our framework uses Random Forests classifier and sparse representation of three features: SIFT, Hue and Opponent angle histograms, and RGB intensities. The experiments are carried out on the public PH2 dataset using a 10-fold cross-validation. The results show that SIFT sparse-coded feature achieves the highest performance with sensitivity and specificity of 100% and 90.3% respectively, with a dictionary size of 800 atoms and a sparsity level of 2. Furthermore, the descriptor based on RGB intensities achieves similar results with sensitivity and specificity of 100% and 71.3%, respectively for a smaller dictionary size of 100 atoms. In conclusion, dictionary learning techniques encode strong structures of dermoscopic images and provide discriminant descriptors.

Automatic seed placement for breast lesion segmentation on US images

Breast lesion boundaries have been mostly extracted by using conventional approaches as a previous step in the development of computer-aided diagnosis systems. Among these, region growing is a frequently used segmentation method. To make the segmentation completely automatic, most of the region growing methods incorporate automatic selection of the seed points. This paper proposes a new automatic seed placement algorithm for breast lesion segmentation on ultrasound images by means of assigning the probability of belonging to a lesion for every pixel depending on intensity, texture and geometrical constraints. The proposal has been evaluated using a set of sonographic breast images with accompanying expert-provided ground truth, and successfully compared to other existing algorithms.

Nondestructive testing based on scanning-from-heating approach: application to nonthrough defect detection and fiber orientation assessment

Abstract. Today, industries ensure the quality of their manufactured products through computer vision techniques and nonconventional imaging. Three-dimensional (3-D) scanners and nondestructive testing (NDT) systems are commonly used independently for such applications. Furthermore, these approaches combined constitute hybrid systems, providing a 3-D reconstruction and NDT analysis. These systems, however, suffer from drawbacks such as errors during the data fusion and higher cost for manufacturers. In an attempt to solve these problems, a single active thermography system based on scanning-from-heating is proposed in this paper. In addition to 3-D digitization of the object, our contributions are twofold: (1) the nonthrough defect detection for a homogeneous metallic object and (2) fiber orientation assessment for a long fiber composite material. The experiments on steel and aluminum plates show that our method achieves the detection of nonthrough defects. Additionally, the estimation of the fiber orientation is evaluated on carbon-fiber composite material.