João Dallyson Sousa de Almeida

Computational methodology for automatic detection of strabismus in digital images through Hirschberg test

Strabismus is a pathology that affects about 4% of the population, causing aesthetic problems, reversible at any age; however, problems that can also cause irreversible muscular alterations, and alter the vision mechanism. The Hirschberg test is one of the exams used to detect this pathology. The application of high technology resources to help diagnose and treat ophthalmological conditions is, lamentably, not commonly found in the sub-specialty of strabismus. This work presents a methodology for automatic detection of strabismus in digital images through the Hirschberg test. For such, the work was organized into four stages: (1) finding the region of the eyes; (2) determining the precise location of the eyes; (3) locating the limbus and brightness; and (4) identifying strabismus. The methodology has produced results on the range of 100% sensibility, 91.3% specificity and 94% for the correct identification of strabismus, ensuring the efficiency of its geostatistical functions for the extraction of eye texture and for the calculation of the alignment between the eyes on digital images obtained from the Hirschberg test.

Texture based on geostatistic for glaucoma diagnosis from fundus eye image

Glaucoma is an ocular disorder that can permanently damage patient vision. Initially, it reduces the visual field, and may cause blindness. Effective methods for early detection is crucial for avoiding significant damages of the patient vision. The use of CAD (Computer-Aided Detection) and CADx (Computer-Aided Diagnosis) systems has contributed to increase the chances of detection and precise diagnoses, assisting experts’ decision making on treatment regarding glaucoma. This paper proposes a method that analyzes the texture of the optical disk image region to diagnose glaucoma. Such analysis is done using the Local Binary Pattern (LBP) to represent the optic disk region, and geostatistical functions to describe texture patterns. The obtained texture features are used for classification based on Support Vector Machine. The proposed method presented as best results a sensitivity of 95%, accuracy of 91% and specificity of 88% in the diagnosis of glaucoma. The method has proved to be promising in assisting glaucoma diagnosis.

Computer-Aided Methodology for Syndromic Strabismus Diagnosis

Strabismus is a pathology that affects approximately 4 % of the population, causing aesthetic problems reversible at any age and irreversible sensory alterations that modify the vision mechanism. The Hirschberg test is one type of examination for detecting this pathology. Computer-aided detection/diagnosis is being used with relative success to aid health professionals. Nevertheless, the routine use of high-tech devices for aiding ophthalmological diagnosis and therapy is not a reality within the subspecialty of strabismus. Thus, this work presents a methodology to aid in diagnosis of syndromic strabismus through digital imaging. Two hundred images belonging to 40 patients previously diagnosed by an specialist were tested. The method was demonstrated to be 88 % accurate in esotropias identification (ET), 100 % for exotropias (XT), 80.33 % for hypertropias (HT), and 83.33 % for hypotropias (HoT). The overall average error was 5.6Δ and 3.83Δ for horizontal and vertical deviations, respectively, against the measures presented by the specialist.

Surgical planning for horizontal strabismus using Support Vector Regression

Strabismus is a pathology which affects about 4% of the population, causing esthetic problems (reversible at any age) and irreversible sensory disorders, altering the vision mechanism. Many techniques can be applied to settle the muscular balance, thus eliminating strabismus. However, when the conservative treatment is not enough, the surgical treatment is adopted, applying recoils or resections to the ocular muscles affected. The factors involved in the surgical strategy in cases of strabismus are complex, demanding both theoretical knowledge and experience from the surgeon. So, the present work proposes a methodology based on Support Vector Regression to help the physician with decision related to horizontal strabismus surgeries. The efficiency of the method at the indication of the surgical plan was evaluated through the average difference between the values that it provided and the values indicated by the specialists. In the planning of medial rectus muscles surgeries, the average error was 0.5mm for recoil and 0.7 for resection. For lateral rectus muscles, the mean error was 0.6 for recoil and 0.8 for resection. The results are promising and prove the feasibility of the use of Support Vector Regression in the indication of strabismus surgeries.

Sign Language Recognition Based on 3D Convolutional Neural Networks.

The inclusion of disabled people is still a recurring problem throughout the world. For the hearing impaired, the barrier imposed by the sign language spoken by a small part of the population imposes limitations that interfere in the quality of life of these people. The popularization or even automation of sign language recognition can take their lives to a higher level. Understanding the importance of sign language recognition for the hearing impaired we propose a 3D CNN architecture for the recognition of 64 classes of gestures from Argentinian Sign Language (LSA64). We demonstrate the efficiency of the method when compared to traditional methods based on hand-crafted features and that its results outperform most deep learning-based work reaching 93.9% of accuracy.

Automatic diagnosis of strabismus in digital videos through cover test

BACKGROUND AND OBJECTIVE Medical image processing can contribute to the detection and diagnosis of human body anomalies, and it represents an important tool to assist in minimizing the degree of uncertainty of any diagnosis, while providing specialists with an additional source of diagnostic information. Strabismus is an anomaly that affects approximately 4% of the population. Strabismus modifies vision such that the eyes do not properly align, influencing binocular vision and depth perception. Additionally, it results in aesthetic problems, which can be reversed at any age. However, the use of low cost computational resources to assist in the diagnosis and treatment of strabismus is not yet widely available. This work presents a computational methodology to automatically diagnose strabismus through digital videos featuring a cover test using only a workstation computer to process these videos. METHODS The method proposed was validated in patients with exotropia and consists of eight steps: (1) acquisition, (2) detection of the region surrounding the eyes, (3) identification of the location of the pupil, (4) identification of the location of the limbus, (5) eye movement tracking, (6) detection of the occluder, (7) identification of evidence of the presence of strabismus, and (8) diagnosis. RESULTS To detect the presence of strabismus, the proposed method achieved a specificity value of 100%, and (2) a sensitivity value of 80%, with 93.33% accuracy in diagnosis of patients with extropia. This procedure was recognized to diagnose strabismus with an accuracy value of 87%, while acknowledging measures lower than 1Δ, and an average error in the deviation measure of 2.57Δ. CONCLUSIONS We demonstrated the feasibility of using computational resources based on image processing techniques to achieve success in diagnosing strabismus by using the cover test. Despite the promising results the proposed method must be validated in a greater volume of video including other types of strabismus.

Automatic eye detection in human faces using geostatistical functions and support vector machines

Several computational systems which depend on the precise location of the eyes have been developed in the last decades. Aware of this need, we propose a method for automatic detection of eyes in images of human faces using four geostatistical functions - semivariogram, semimadogram, covariogram and correlogram and support vector machines. The method was tested using the ORL human face database, which contains 400 images grouped in 40 persons, each having 10 different expressions. The detection obtained the following results of sensibility of 93.3%, specificity of 82.2% and accuracy of 89.4%.

Diagnosis of Non-Small Cell Lung Cancer Using Phylogenetic Diversity in Radiomics Context.

Lung cancer is the most common type of cancer and has the highest mortality rate in the world. The automatic process for the diagnosis by computer vision systems, through medical images, provides an interpretation regarding the pathology. The idea of this work is to use the texture features using phylogenetic diversity indexes, to classify Non-Small Cell Lung Cancer. This work presents the development of texture descriptors based on phylogenetic diversity indices for characterization of the nodule. The tests showed promising results of 98.47% accuracy, a Kappa index of 0.979 and an ROC of 0.999.

Segmentation of the Retinal Reflex in Brückner Test Images Using U-Net Convolutional Network.

Bruckner test is an eye exam characterized by the evaluation of brightness of red retinal reflex in pupillary area. The reflex region segmentation is important for a computational method that automatizes that examination and detects eye pathologies by the image analysis. This work presents an automatic method for retinal reflex segmentation in images of Bruckner test using the fully convolutional network U-Net. The method reaches 87.73% of Dice coefficient, 78.95% of Jaccard index, 90.63% recall and 88.03% precision.

Glaucoma diagnosis in fundus eye images using diversity indexes

Glaucoma is the second major cause of vision loss worldwide. It is usually caused by the increase in the intraocular pressure, which damages the optic nerve resulting in gradual vision loss. Glaucoma is an asymptomatic disease in the initial stages. Early detection and treatment may prevent the vision loss. The head of the optic nerve (optic disc) is examined by using fundus eye images. Computer systems have been used to provide support in glaucoma diagnosis. This work proposes a method for glaucoma diagnosis using fundus eye images. Diversity indexes, which are typically used in ecological studies, are used in this work as texture descriptors in the optic disc region. Then, a feature selection procedure is performed using genetic algorithm and support vector machines (SVM) are used to classify fundus eye images in glaucomatous or normal. The proposed method obtained promising results for glaucoma diagnosis, reaching an accuracy of 93.41%, sensitivity of 92.83% and specificity of 93.69%.