Decision Trees for Glaucoma Screening Based on the Asymmetry of the Retinal Nerve Fiber Layer in Optical Coherence Tomography.

Purpose: The aim of this study was to analyze the relevance of asymmetry features between both eyes of the same patient for glaucoma screening using optical coherence tomography. Methods: Spectral-domain optical coherence tomography was used to estimate the thickness of the peripapillary retinal nerve fiber layer in both eyes of the patients in the study. These measurements were collected in a dataset from healthy and glaucoma patients. Several metrics for asymmetry in the retinal nerve fiber layer thickness between the two eyes were then proposed. These metrics were evaluated using the dataset by performing a statistical analysis to assess their significance as relevant features in the diagnosis of glaucoma. Finally, the usefulness of these asymmetry features was demonstrated by designing supervised machine learning models that can be used for the early diagnosis of glaucoma. Results: Machine learning models were designed and optimized, specifically decision trees, based on the values of proposed asymmetry metrics. The use of these models on the dataset provided good classification of the patients (accuracy 88%, sensitivity 70%, specificity 93% and precision 75%). Conclusions: The obtained machine learning models based on retinal nerve fiber layer asymmetry are simple but effective methods which offer a good trade-off in classification of patients and simplicity. The fast binary classification relies on a few asymmetry values of the retinal nerve fiber layer thickness, allowing their use in the daily clinical practice for glaucoma screening.

Improving Glaucoma Diagnosis Assembling Deep Networks and Voting Schemes.

Glaucoma is a group of eye conditions that damage the optic nerve, the health of which is vital for good eyesight. This damage is often caused by higher-than-normal pressure in the eye. In the past few years, the applications of artificial intelligence and data science have increased rapidly in medicine especially in imaging applications. In particular, deep learning tools have been successfully applied obtaining, in some cases, results superior to those obtained by humans. In this article, we present a soft novel ensemble model based on the K-NN algorithm, that combines the probability of class membership obtained by several deep learning models. In this research, three models of different nature (CNN, CapsNets and Convolutional Autoencoders) have been selected searching for diversity. The latent space of these models are combined using the local information provided by the true sample labels and the K-NN algorithm is applied to determine the final decision. The results obtained on two different datasets of retinal images show that the proposed ensemble model improves the diagnosis capabilities for both the individual models and the state-of-the-art results.

PAPILA: Dataset with fundus images and clinical data of both eyes of the same patient for glaucoma assessment.

Glaucoma is one of the ophthalmological diseases that frequently causes loss of vision in today's society. Previous studies assess which anatomical parameters of the optic nerve can be predictive of glaucomatous damage, but to date there is no test that by itself has sufficient sensitivity and specificity to diagnose this disease. This work provides a public dataset with medical data and fundus images of both eyes of the same patient. Segmentations of the cup and optic disc, as well as the labeling of the patients based on the evaluation of clinical data are also provided. The dataset has been tested with a neural network to classify healthy and glaucoma patients. Specifically, the ResNet-50 has been used as the basis to classify patients using information from each eye independently as well as using the joint information from both eyes of each patient. Results provide the baseline metrics, with the aim of promoting research in the early detection of glaucoma based on the joint analysis of both eyes of the same patient.

Bioinspired Architecture Selection for Multitask Learning.

Faced with a new concept to learn, our brain does not work in isolation. It uses all previously learned knowledge. In addition, the brain is able to isolate the knowledge that does not benefit us, and to use what is actually useful. In machine learning, we do not usually benefit from the knowledge of other learned tasks. However, there is a methodology called Multitask Learning (MTL), which is based on the idea that learning a task along with other related tasks produces a transfer of information between them, what can be advantageous for learning the first one. This paper presents a new method to completely design MTL architectures, by including the selection of the most helpful subtasks for the learning of the main task, and the optimal network connections. In this sense, the proposed method realizes a complete design of the MTL schemes. The method is simple and uses the advantages of the Extreme Learning Machine to automatically design a MTL machine, eliminating those factors that hinder, or do not benefit, the learning process of the main task. This architecture is unique and it is obtained without testing/error methodologies that increase the computational complexity. The results obtained over several real problems show the good performances of the designed networks with this method.

Automatic detection of the intima-media thickness in ultrasound images of the common carotid artery using neural networks.

Atherosclerosis is the leading underlying pathologic process that results in cardiovascular diseases, which represents the main cause of death and disability in the world. The atherosclerotic process is a complex degenerative condition mainly affecting the medium- and large-size arteries, which begins in childhood and may remain unnoticed during decades. The intima-media thickness (IMT) of the common carotid artery (CCA) has emerged as one of the most powerful tool for the evaluation of preclinical atherosclerosis. IMT is measured by means of B-mode ultrasound images, which is a non-invasive and relatively low-cost technique. This paper proposes an effective image segmentation method for the IMT measurement in an automatic way. With this purpose, segmentation is posed as a pattern recognition problem, and a combination of artificial neural networks has been trained to solve this task. In particular, multi-layer perceptrons trained under the scaled conjugate gradient algorithm have been used. The suggested approach is tested on a set of 60 longitudinal ultrasound images of the CCA by comparing the automatic segmentation with four manual tracings. Moreover, the intra- and inter-observer errors have also been assessed. Despite of the simplicity of our approach, several quantitative statistical evaluations have shown its accuracy and robustness.

Neural architecture design based on extreme learning machine.

Selection of the optimal neural architecture to solve a pattern classification problem entails to choose the relevant input units, the number of hidden neurons and its corresponding interconnection weights. This problem has been widely studied in many research works but their solutions usually involve excessive computational cost in most of the problems and they do not provide a unique solution. This paper proposes a new technique to efficiently design the MultiLayer Perceptron (MLP) architecture for classification using the Extreme Learning Machine (ELM) algorithm. The proposed method provides a high generalization capability and a unique solution for the architecture design. Moreover, the selected final network only retains those input connections that are relevant for the classification task. Experimental results show these advantages.

Segmentation of the common carotid artery walls based on a frequency implementation of active contours: segmentation of the common carotid artery walls.

Atherosclerosis is one of the most extended cardiovascular diseases nowadays. Although it may be unnoticed during years, it also may suddenly trigger severe illnesses such as stroke, embolisms or ischemia. Therefore, an early detection of atherosclerosis can prevent adult population from suffering more serious pathologies. The intima-media thickness (IMT) of the common carotid artery (CCA) has been used as an early and reliable indicator of atherosclerosis for years. The IMT is manually computed from ultrasound images, a process that can be repeated as many times as necessary (over different ultrasound images of the same patient), but also prone to errors. With the aim to reduce the inter-observer variability and the subjectivity of the measurement, a fully automatic computer-based method based on ultrasound image processing and a frequency-domain implementation of active contours is proposed. The images used in this work were obtained with the same ultrasound scanner (Philips iU22 Ultrasound System) but with different spatial resolutions. The proposed solution does not extract only the IMT but also the CCA diameter, which is not as relevant as the IMT to predict future atherosclerosis evolution but it is a statistically interesting piece of information for the doctors to determine the cardiovascular risk. The results of the proposed method have been validated by doctors, and these results are visually and numerically satisfactory when considering the medical measurements as ground truth, with a maximum deviation of only 3.4 pixels (0.0248 mm) for IMT.

DRR and portal image registration for automatic patient positioning in radiotherapy treatment.

Image processing turns out to be essential in the planning and verification of radiotherapy treatments. Before applying a radiotherapy treatment, a dosimetry planning must be performed. Usually, the planning is done by means of an X-ray volumetric analysis using computerized tomography, where the area to be radiated is marked out. During the treatment phase, it is necessary to place the patient under the particle accelerator exactly as considered in the dosimetry stage. Coarse alignment is achieved using fiduciary markers placed over the patient's skin as external references. Later, fine alignment is provided by comparing a digitally reconstructed radiography (DRR) from the planning stage and a portal image captured by the accelerator in the treatment stage. The preprocessing of DRR and portal images, as well as the minimization of the non-shared information between both kinds of images, is mandatory for the correct operation of the image registration algorithm. With this purpose, mathematical morphology and image processing techniques have been used. The present work describes a fully automatic method to calculate more accurately the necessary displacement of the couch to place the patient exactly at the planned position. The proposed method to achieve the correct positioning of the patient is based on advanced image registration techniques. Preliminary results show a perfect match with the displacement estimated by the physician.