Roberto Hornero

Retinopathy Online Challenge: Automatic Detection of Microaneurysms in Digital Color Fundus Photographs

The detection of microaneurysms in digital color fundus photographs is a critical first step in automated screening for diabetic retinopathy (DR), a common complication of diabetes. To accomplish this detection numerous methods have been published in the past but none of these was compared with each other on the same data. In this work we present the results of the first international microaneurysm detection competition, organized in the context of the Retinopathy Online Challenge (ROC), a multiyear online competition for various aspects of DR detection. For this competition, we compare the results of five different methods, produced by five different teams of researchers on the same set of data. The evaluation was performed in a uniform manner using an algorithm presented in this work. The set of data used for the competition consisted of 50 training images with available reference standard and 50 test images where the reference standard was withheld by the organizers (M. Niemeijer, B. van Ginneken, and M. D. AbrA¿moff). The results obtained on the test data was submitted through a website after which standardized evaluation software was used to determine the performance of each of the methods. A human expert detected microaneurysms in the test set to allow comparison with the performance of the automatic methods. The overall results show that microaneurysm detection is a challenging task for both the automatic methods as well as the human expert. There is room for improvement as the best performing system does not reach the performance of the human expert. The data associated with the ROC microaneurysm detection competition will remain publicly available and the website will continue accepting submissions.

Interpretation of the Lempel-Ziv Complexity Measure in the Context of Biomedical Signal Analysis

Lempel-Ziv complexity (LZ) and derived LZ algorithms have been extensively used to solve information theoretic problems such as coding and lossless data compression. In recent years, LZ has been widely used in biomedical applications to estimate the complexity of discrete-time signals. Despite its popularity as a complexity measure for biosignal analysis, the question of LZ interpretability and its relationship to other signal parameters and to other metrics has not been previously addressed. We have carried out an investigation aimed at gaining a better understanding of the LZ complexity itself, especially regarding its interpretability as a biomedical signal analysis technique. Our results indicate that LZ is particularly useful as a scalar metric to estimate the bandwidth of random processes and the harmonic variability in quasi-periodic signals

Analysis of electroencephalograms in Alzheimer's disease patients with multiscale entropy

The aim of this study was to analyse the electroencephalogram (EEG) background activity of Alzheimers disease (AD) patients using multiscale entropy (MSE). MSE is a recently developed method that quantifies the regularity of a signal on different time scales. These time scales are inspected by means of several coarse-grained sequences formed from the analysed signals. We recorded the EEGs from 19 scalp electrodes in 11 AD patients and 11 age-matched controls and estimated the MSE profile for each epoch of the EEG recordings. The shape of the MSE profiles reveals the EEG complexity, and it suggests that the EEG contains information in deeper scales than the smallest one. Moreover, the results showed that the EEG background activity is less complex in AD patients than control subjects. We found significant differences between both subject groups at electrodes F3, F7, Fp1, Fp2, T5, T6, P3, P4, O1 and O2 (p-value < 0.01, Students t-test). These findings indicate that the EEG complexity analysis performed on deeper time scales by MSE may be a useful tool in order to increase our knowledge of AD.

Retinal image analysis based on mixture models to detect hard exudates.

Diabetic Retinopathy is one of the leading causes of blindness in developed countries. Hard exudates have been found to be one of the most prevalent earliest clinical signs of retinopathy. Thus, automatic detection of hard exudates from retinal images is clinically significant. In this study, an automatic method to detect hard exudates is proposed. The algorithm is based on mixture models to dynamically threshold the images in order to separate exudates from background. A postprocessing technique, based on edge detection, is applied to distinguish hard exudates from cotton wool spots and other artefacts. We prospectively assessed the algorithm performance using a database of 80 retinal images with variable colour, brightness, and quality. The algorithm obtained a sensitivity of 90.2% and a positive predictive value of 96.8% using a lesion-based criterion. The image-based classification accuracy is also evaluated obtaining a sensitivity of 100% and a specificity of 90%.

Nonlinear analysis of electroencephalogram and magnetoencephalogram recordings in patients with Alzheimer's disease

The aim of the present study is to show the usefulness of nonlinear methods to analyse the electroencephalogram (EEG) and magnetoencephalogram (MEG) in patients with Alzheimers disease (AD). The following nonlinear methods have been applied to study the EEG and MEG background activity in AD patients and control subjects: approximate entropy, sample entropy, multiscale entropy, auto-mutual information and Lempel–Ziv complexity. We discuss why these nonlinear methods are appropriate to analyse the EEG and MEG. Furthermore, the performance of all these methods has been compared when applied to the same databases of EEG and MEG recordings. Our results show that EEG and MEG background activities in AD patients are less complex and more regular than in healthy control subjects. In line with previous studies, our work suggests that nonlinear analysis techniques could be useful in AD diagnosis.

Neural network based detection of hard exudates in retinal images

Diabetic retinopathy (DR) is an important cause of visual impairment in developed countries. Automatic recognition of DR lesions in fundus images can contribute to the diagnosis of the disease. The aim of this study is to automatically detect one of these lesions, hard exudates (EXs), in order to help ophthalmologists in the diagnosis and follow-up of the disease. We propose an algorithm which includes a neural network (NN) classifier for this task. Three NN classifiers were investigated: multilayer perceptron (MLP), radial basis function (RBF) and support vector machine (SVM). Our database was composed of 117 images with variable colour, brightness, and quality. 50 of them (from DR patients) were used to train the NN classifiers and 67 (40 from DR patients and 27 from healthy retinas) to test the method. Using a lesion-based criterion, we achieved a mean sensitivity (SE(l)) of 88.14% and a mean positive predictive value (PPV(l)) of 80.72% for MLP. With RBF we obtained SE(l)=88.49% and PPV(l)=77.41%, while we reached SE(l)=87.61% and PPV(l)=83.51% using SVM. With an image-based criterion, a mean sensitivity (SE(i)) of 100%, a mean specificity (SP(i)) of 92.59% and a mean accuracy (AC(i)) of 97.01% were obtained with MLP. Using RBF we achieved SE(i)=100%, SP(i)=81.48% and AC(i)=92.54%. With SVM the image-based results were SE(i)=100%, SP(i)=77.78% and AC(i)=91.04%.

Retinal image analysis to detect and quantify lesions associated with diabetic retinopathy

An automatic method to detect hard exudates, a lesion associated with diabetic retinopathy, is proposed. The algorithm found on their color, using a statistical classification, and their sharp edges, applying an edge detector, to localize them. A sensitivity of 79.62% with a mean number of 3 false positives per image is obtained in a database of 20 retinal image with variable color, brightness and quality. In that way, we evaluate the robustness of the method in order to make adequate to a clinical environment. Further efforts will be done to improve its performance.

Multivariate Analysis of Blood Oxygen Saturation Recordings in Obstructive Sleep Apnea Diagnosis

This study focuses on the analysis of blood oxygen saturation (SaO2) from nocturnal pulse oximetry (NPO) to help in the diagnosis of the obstructive sleep apnea (OSA) syndrome. A population of 148 patients suspected of suffering from OSA syndrome was studied. A wide set of 16 features was used to characterize changes in the SaO2 profile during the night. Our feature set included common statistics in the time and frequency domains, conventional spectral characteristics from the power spectral density (PSD) function, and nonlinear features. We performed feature selection by means of a step-forward logistic regression (LR) approach with leave-one-out cross-validation. Second- and fourth-order statistical moments in the time domain (M2t and M4t), the relative power in the 0.014-0.033 Hz frequency band (PR), and the Lempel-Ziv complexity (LZC) were automatically selected. 92.0% sensitivity, 85.4% specificity, and 89.7% accuracy were obtained. The optimum feature set significantly improved the diagnostic ability of each feature individually. Furthermore, our results outperformed classic oximetric indexes commonly used by physicians. We conclude that simultaneous analysis in the time and frequency domains by means of statistical moments, spectral and nonlinear features could provide complementary information from NPO to improve OSA diagnosis.

Utility of Approximate Entropy From Overnight Pulse Oximetry Data in the Diagnosis of the Obstructive Sleep Apnea Syndrome

Approximate entropy (ApEn) is a family of statistics introduced as a quantification of regularity in time series without any a priori knowledge about the system generating them. The aim of this preliminary study was to assess whether a time series analysis of arterial oxygen saturation (SaO2) signals from overnight pulse oximetry by means of ApEn could yield essential information on the diagnosis of obstructive sleep apnea (OSA) syndrome. We analyzed SaO2 signals from 187 subjects: 111 with a positive diagnosis of OSA and 76 with a negative diagnosis of OSA. We divided our data in a training set (44 patients with OSA Positive and 30 patients with OSA Negative) and a test set (67 patients with OSA Positive and 46 patients with OSA Negative). The training set was used for algorithm development and optimum threshold selection. Results showed that recurrence of apnea events in patients with OSA determined a significant increase in ApEn values. This method was assessed prospectively using the test dataset, where we obtained 82.09% sensitivity and 86.96% specificity. We conclude that ApEn analysis of SaO2 from pulse oximetric recording could be useful in the study of OSA

Nonlinear characteristics of blood oxygen saturation from nocturnal oximetry for obstructive sleep apnoea detection.

Nocturnal oximetry is an attractive option for the diagnosis of obstructive sleep apnoea (OSA) syndrome because of its simplicity and low cost compared to polysomnography (PSG). The present study assesses nonlinear analysis of blood oxygen saturation (SaO(2)) from nocturnal oximetry as a diagnostic test to discriminate between OSA positive and OSA negative patients. A sample of 187 referred outpatients, clinically suspected of having OSA, was studied using nocturnal oximetry performed simultaneously with complete PSG. A positive OSA diagnosis was found for 111 cases, while the remaining 76 cases were classified as OSA negative. The following oximetric indices were obtained: cumulative time spent below a saturation of 90% (CT90), oxygen desaturation indices of 4% (ODI4), 3% (ODI3) and 2% (ODI2) and the delta index (Delta index). SaO(2) records were subsequently processed applying two nonlinear methods: central tendency measure (CTM) and Lempel-Ziv (LZ) complexity. Significant differences (p < 0.01) were found between OSA positive and OSA negative patients. Using CTM we obtained a sensitivity of 90.1% and a specificity of 82.9%, while with LZ the sensitivity was 86.5% and the specificity was 77.6%. CTM and LZ accuracies were higher than those provided by ODI4, ODI3, ODI2 and CT90. The results suggest that nonlinear analysis of SaO(2) signals from nocturnal oximetry could yield useful information in OSA diagnosis.