Grupo de Ingeniería Biomédica

DDFU-Net: A Deep Decoder-Focused U-Net Model for Retinal Lesion Segmentation

2026-03-26T20:05:16Z

Early detection of retinal lesions helps to avoid visual loss or blindness. The main lesions associated with eye diseases include soft exudates, hard exudates, microaneurysms, and hemorrhages. However, the segmentation of these four kinds of lesions is difficult and time-consuming due to their uncertainty in size, contrast, and high inter-class similarity. To address these issues, this study presents Deep Decoder-Focused U-Net (DDFU-Net), an asymmetric dense U-Net model for automatic and accurate multi-lesion segmentation using fundus images. Our approach simultaneously segments all four kinds of retinal lesions after proving that multi-task learning yields better results than single-task learning. DDFU-Net incorporates an asymmetric design with five dense blocks in the encoder and seven dense blocks in the decoder. This design enhances feature extraction while ensuring a more refined reconstruction of lesion boundaries, particularly for small and complex structures. By allocating more layers to the decoder, the model improves segmentation accuracy by gradually restoring spatial details lost during down-sampling, mitigating over-compression, and enhancing fine-grained feature preservation. Comprehensive experiments on IDRiD and DDR datasets well demonstrate the superiority of our approach, which outperforms state-of-the-art segmentation methods. Specifically, DDFU-Net achieved a mean Area Under the Precision-Recall Curve of 54.86%, a mean Intersection Over Union of 36.96%, and mean Dice scores of 52.24% on the DDR test set. On the IDRiD test set, it achieved 66.69%, 57.31%, and 69.93%, respectively. The asymmetric structure outperforms traditional symmetric U-Nets by capturing more detailed features during encoding while reducing complexity during decoding. The proposed method can be useful to aid in the diagnosis of eye diseases, reducing the workload of specialists and improving the attention to patients.

Variability and task-responsiveness of electrophysiological dynamics: Scale-free stability and oscillatory flexibility

2026-02-13T20:01:13Z

Cortical oscillations and scale-free neural activity are thought to influence a variety of cognitive functions, but their differential relationships to neural stability and flexibility has never been investigated. Based on the existing literature, we hypothesize that scale-free and oscillatory processes in the brain exhibit different trade-offs between stability and flexibility; specifically, cortical oscillations may reflect variable, task-responsive aspects of brain activity, while scale-free activity is proposed to reflect a more stable and task-unresponsive aspect. We test this hypothesis using data from two large-scale MEG studies (HCP: n = 89; CamCAN: n = 195), operationalizing stability and flexibility by task-responsiveness and spontaneous intra-subject variability in resting state. We demonstrate that the power-law exponent of scale-free activity is a highly stable parameter, which responds little to external cognitive demands and shows minimal spontaneous fluctuations over time. In contrast, oscillatory power, particularly in the alpha range (8–13 Hz), responds strongly to tasks and exhibits comparatively large spontaneous fluctuations over time. In sum, our data support differential roles for oscillatory and scale-free activity in the brain with respect to neural stability and flexibility. This result carries implications for criticality-based theories of scale-free activity, state-trait models of variability, and homeostatic views of the brain with regulated variables vs. effectors.

Altered brain dynamics index levels of arousal in complete locked-in syndrome

2026-02-13T20:01:11Z

Complete locked-in syndrome (CLIS) resulting from late-stage amyotrophic lateral sclerosis (ALS) is characterised by loss of motor function and eye movements. The absence of behavioural indicators of consciousness makes the search for neuronal correlates as possible biomarkers clinically and ethically urgent. EEG-based measures of brain dynamics such as power-law exponent (PLE) and Lempel-Ziv complexity (LZC) have been shown to have explanatory power for consciousness and may provide such neuronal indices for patients with CLIS. Here, we validated PLE and LZC (calculated in a dynamic way) as benchmarks of a wide range of arousal states across different reference states of consciousness (e.g., awake, sleep stages, ketamine, sevoflurane). We show a tendency toward high PLE and low LZC, with high intra-subject fluctuations and inter-subject variability in a cohort of CLIS patients with values graded along different arousal states as in our reference data sets. In conclusion, changes in brain dynamics indicate altered arousal in CLIS. Specifically, PLE and LZC are potentially relevant biomarkers to identify or diagnose the arousal level in CLIS and to determine the optimal time point for treatment, including communication attempts.

Intrinsic neural timescales: temporal integration and segregation

2026-02-13T20:01:10Z

We are continuously bombarded by external inputs of various timescales from the environment. How does the brain process this multitude of timescales? Recent resting state studies show a hierarchy of intrinsic neural timescales (INT) with a shorter duration in unimodal regions (e.g., visual cortex and auditory cortex) and a longer duration in transmodal regions (e.g., default mode network). This unimodal-transmodal hierarchy is present across acquisition modalities [electroencephalogram (EEG)/magnetoencephalogram (MEG) and fMRI] and can be found in different species and during a variety of different task states. Together, this suggests that the hierarchy of INT is central to the temporal integration (combining successive stimuli) and segregation (separating successive stimuli) of external inputs from the environment, leading to temporal segmentation and prediction in perception and cognition.

Overcoming Rest–Task Divide—Abnormal Temporospatial Dynamics and Its Cognition in Schizophrenia

2026-02-13T20:01:10Z

Schizophrenia is a complex psychiatric disorder exhibiting alterations in spontaneous and task-related cerebral activity whose relation (termed “state dependence”) remains unclear. For unraveling their relationship, we review recent electroencephalographic (and a few functional magnetic resonance imaging) studies in schizophrenia that assess and compare both rest/prestimulus and task states, ie, rest/prestimulus–task modulation. Results report reduced neural differentiation of task-related activity from rest/prestimulus activity across different regions, neural measures, cognitive domains, and imaging modalities. Together, the findings show reduced rest/prestimulus–task modulation, which is mediated by abnormal temporospatial dynamics of the spontaneous activity. Abnormal temporospatial dynamics, in turn, may lead to abnormal prediction, ie, predictive coding, which mediates cognitive changes and psychopathological symptoms, including confusion of internally and externally oriented cognition. In conclusion, reduced rest/prestimulus–task modulation in schizophrenia provides novel insight into the neuronal mechanisms that connect task-related changes to cognitive abnormalities and psychopathological symptoms.

The brain and its time: intrinsic neural timescales are key for input processing

2026-02-13T20:01:08Z

We process and integrate multiple timescales into one meaningful whole. Recent evidence suggests that the brain displays a complex multiscale temporal organization. Different regions exhibit different timescales as described by the concept of intrinsic neural timescales (INT); however, their function and neural mechanisms remains unclear. We review recent literature on INT and propose that they are key for input processing. Specifically, they are shared across different species, i.e., input sharing. This suggests a role of INT in encoding inputs through matching the inputs’ stochastics with the ongoing temporal statistics of the brain’s neural activity, i.e., input encoding. Following simulation and empirical data, we point out input integration versus segregation and input sampling as key temporal mechanisms of input processing. This deeply grounds the brain within its environmental and evolutionary context. It carries major implications in understanding mental features and psychiatric disorders, as well as going beyond the brain in integrating timescales into artificial intelligence.

Heart rate variability as a potential biomarker of pediatric obstructive sleep apnea resolution

2026-03-25T08:43:48Z

Abstract Study Objectives Pediatric obstructive sleep apnea (OSA) affects cardiac autonomic regulation, altering heart rate variability (HRV). Although changes in classical HRV parameters occur after OSA treatment, they have not been evaluated as reporters of OSA resolution. Specific frequency bands (named BW1, BW2, and BWRes) have been recently identified in OSA. We hypothesized that changes with treatment in these spectral bands can reliably identify changes in OSA severity and reflect OSA resolution. Methods Four hundred and four OSA children (5–9.9 years) from the prospective Childhood Adenotonsillectomy Trial were included; 206 underwent early adenotonsillectomy (eAT), while 198 underwent watchful waiting with supportive care (WWSC). HRV changes from baseline to follow-up were computed for classical and OSA-related frequency bands. Causal mediation analysis was conducted to evaluate how treatment influences HRV through mediators such as OSA resolution and changes in disease severity. Disease resolution was initially assessed by considering only obstructive events, and was followed by adding central apneas to the analyses. Results Treatment, regardless of eAT or WWSC, affects HRV activity, mainly in the specific frequency band BW2 (0.028–0.074 Hz). Furthermore, only changes in BW2 were specifically attributable to all OSA resolution mediators. HRV activity in BW2 also showed statistically significant differences between resolved and non-resolved OSA. Conclusions OSA treatment affects HRV activity in terms of change in severity and disease resolution, especially in OSA-related BW2 frequency band. This band allowed to differentiate HRV activity between children with and without resolution, so we propose BW2 as potential biomarker of pediatric OSA resolution.

Heart rate variability spectrum characteristics in children with sleep apnea

2026-03-25T08:35:44Z

Background Classic spectral analysis of heart rate variability (HRV) in pediatric sleep apnea–hypopnea syndrome (SAHS) traditionally evaluates the very low frequency (VLF: 0–0.04 Hz), low frequency (LF: 0.04–0.15 Hz), and high frequency (HF: 0.15–0.40 Hz) bands. However, specific SAHS-related frequency bands have not been explored. Methods One thousand seven hundred and thirty-eight HRV overnight recordings from two pediatric databases (0–13 years) were evaluated. The first one (981 children) served as training set to define new HRV pediatric SAHS-related frequency bands. The associated relative power (RP) were computed in the test set, the Childhood Adenotonsillectomy Trial database (CHAT, 757 children). Their relationships with polysomnographic variables and diagnostic ability were assessed. Results Two new specific spectral bands of pediatric SAHS within 0–0.15 Hz were related to duration of apneic events, number of awakenings, and wakefulness after sleep onset (WASO), while an adaptive individual-specific new band from HF was related to oxyhemoglobin desaturations, arousals, and WASO. Furthermore, these new spectral bands showed improved diagnostic ability than classic HRV. Conclusions Novel spectral bands provide improved characterization of pediatric SAHS. These findings may pioneer a better understanding of the effects of SAHS on cardiac function and potentially serve as detection biomarkers.

ECG-based convolutional neural network in pediatric obstructive sleep apnea diagnosis

2026-03-17T12:55:11Z

Obstructive sleep apnea (OSA) is a prevalent respiratory condition in children and is characterized by partial or complete obstruction of the upper airway during sleep. The respiratory events in OSA induce transient alterations of the cardiovascular system that ultimately can lead to increased cardiovascular risk in affected children. Therefore, a timely and accurate diagnosis is of utmost importance. However, polysomnography (PSG), the standard diagnostic test for pediatric OSA, is complex, uncomfortable, costly, and relatively inaccessible, particularly in low-resource environments, thereby resulting in substantial underdiagnosis. Here, we propose a novel deep-learning approach to simplify the diagnosis of pediatric OSA using raw electrocardiogram tracing (ECG). Specifically, a new convolutional neural network (CNN)-based regression model was implemented to automatically predict pediatric OSA by estimating its severity based on the apnea-hypopnea index (AHI) and deriving 4 OSA severity categories. For this purpose, overnight ECGs from 1,610 PSG recordings obtained from the Childhood Adenotonsillectomy Trial (CHAT) database were used. The database was randomly divided into approximately 60%, 20%, and 20% for training, validation, and testing, respectively. The diagnostic performance of the proposed CNN model largely outperformed the most accurate previous algorithms that relied on ECG-derived features (4-class Cohen's kappa coefficient of 0.373 versus 0.166). Specifically, for AHI cutoff values of 1, 5, and 10 events/hour, the binary classification achieved sensitivities of 84.19%, 76.67%, and 53.66%; specificities of 46.15%, 91.39%, and 98.06%; and accuracies of 75.92%, 86.96%, and 91.97%, respectively. Therefore, pediatric OSA can be readily identified by our proposed CNN model, which provides a simpler, faster, and more accessible diagnostic test that can be implemented in clinical practice.

Usefulness of recurrence plots from airflow recordings to aid in paediatric sleep apnoea diagnosis

2026-03-05T08:26:38Z

Background and objective: In-laboratory overnight polysomnography (PSG) is the gold standard method to diagnose the Sleep Apnoea-Hypopnoea Syndrome (SAHS). PSG is a complex, expensive, labour-intensive and time-consuming test. Consequently, simplified diagnostic methods are desirable. We propose the analysis of the airflow (AF) signal by means of recurrence plots (RP) features. The main goal of our study was to evaluate the utility of the information from RPs of the AF signals to detect paediatric SAHS at different levels of severity. In addition, we also evaluated the complementarity with the 3% oxygen desaturation index (ODI3). Methods: 946 AF and blood oxygen saturation (SpO2) recordings from children ages 0–13 years were used. The population under study was randomly split into training (60%) and test (40%) sets. RP was computed and 9 RP features were extracted from each AF recording. ODI3 was also calculated from each SpO2 recording. A feature selection stage was conducted in the training group by means of the fast correlation-based filter (FCBF) methodology to obtain a relevant and non-redundant optimum feature subset. A multi-layer perceptron neural network with Bayesian approach (BY-MLP), trained with these optimum features, was used to estimate the apnoea–hypopnoea index (AHI). Results: 8 of the RP features showed statistically significant differences (p-value <0.01) among the SAHS severity groups. FCBF selected the maximum length of the diagonal lines from RP, as well as the ODI3. Using these optimum features, the BY-MLP model achieved 83.2%, 78.5%, and 91.0% accuracy in the test group for the AHI thresholds 1, 5, and 10 events/h, respectively. Moreover, this model reached a negative likelihood ratio of 0.1 for 1 event/h and a positive likelihood ratio of 13.7 for 10 events/h. Conclusions: RP analysis enables extraction of useful SAHS-related information from overnight AF paediatric recordings. Moreover, it provides complementary information to the widely-used clinical variable ODI3. Thus, RP applied to AF signals can be used along with ODI3 to help in paediatric SAHS diagnosis, particularly to either confirm the absence of SAHS or the presence of severe SAHS.

Bispectral analysis of overnight airflow to improve the pediatric sleep apnea diagnosis

2026-03-24T09:46:49Z

Pediatric Obstructive Sleep Apnea (OSA) is a respiratory disease whose diagnosis is performed through overnight polysomnography (PSG). Since it is a complex, time-consuming, expensive, and labor-intensive test, simpler alternatives are being intensively sought. In this study, bispectral analysis of overnight airflow (AF) signal is proposed as a potential approach to replace PSG when indicated. Thus, our objective was to characterize AF through bispectrum, and assess its performance to diagnose pediatric OSA. This characterization was conducted using 13 bispectral features from 946 AF signals. The oxygen desaturation index ≥3% (ODI3), a common clinical measure of OSA severity, was also obtained to evaluate its complementarity to the AF bispectral analysis. The fast correlation-based filter (FCBF) and a multi-layer perceptron (MLP) were used for subsequent automatic feature selection and pattern recognition stages. FCBF selected 3 bispectral features and ODI3, which were used to train a MLP model with ability to estimate apnea-hypopnea index (AHI). The model reached 82.16%, 82.49%, and 90.15% accuracies for the common AHI cut-offs 1, 5, and 10 events/h, respectively. The different bispectral approaches used to characterize AF in children provided complementary information. Accordingly, bispectral analysis showed that the occurrence of apneic events decreases the non-gaussianity and non-linear interaction of the AF harmonic components, as well as the regularity of the respiratory patterns. Moreover, the bispectral information from AF also showed complementarity with ODI3. Our findings suggest that AF bispectral analysis may serve as a useful tool to simplify the diagnosis of pediatric OSA, particularly for children with moderate-to-severe OSA.

Combined explainable deep learning model to predict pediatric sleep apnea from ECG and SpO2

2026-03-17T07:41:51Z

Combining deep learning (DL) with eXplainable Artificial Intelligence (XAI) techniques has led to clinically applicable models that simplify the diagnosis of pediatric obstructive sleep apnea (OSA) using a restricted number of cardiorespiratory signals. However, no prior study has applied these techniques to concurrently analyze electrocardiogram (ECG) and oxygen saturation (SpO2) data. Here, we present an explainable DL approach integrating convolutional neural networks with overnight SpO2 and ECG signals to identify pediatric OSA. SHapley Additive exPlanations (SHAP) XAI technique was used to extract relevant patterns linked to pediatric OSA and explain the model decisions. Patients (n = 3,320) from the semi-public Childhood Adenotonsillectomy Trial (CHAT) and Pediatric Adenotonsillectomy Trial for Snoring (PATS), and the private University of Chicago (UofC) databases were analyzed. Performance obtained Cohen’s 4-class kappa of 0.549, 0.457, and 0.378 in CHAT, PATS, and UofC, respectively. Shapley values increased with OSA severity and highlighted the complementarity of SpO2 and ECG, with SpO2 being more relevant in moderate and severe cases and ECG in mild or no OSA cases. SHAP visualizations identified SpO2 desaturations linked to clusters of apneic events and those occurring independently. It also highlighted bradycardia-tachycardia and ECG cardiovascular risk patterns, including variations in P and T waves, PQ and QT intervals, and the QRS complex. Shapley values identified correlations between respiratory and cardiac patterns, showing that desaturations in OSA are linked to cardiac changes. Therefore, our interpretable DL approach may improve pediatric OSA diagnosis by integrating breathing information and accompanying cardiac changes, supporting its effective adoption in clinical settings

SleepECG-Net: Explainable Deep Learning Approach With ECG for Pediatric Sleep Apnea Diagnosis

2026-03-17T08:06:54Z

Obstructive sleep apnea (OSA) in children is a prevalent and serious respiratory condition linked to cardiovascular morbidity. Polysomnography, the standard diagnostic approach, faces challenges in accessibility and complexity, leading to underdiagnosis. To simplify OSA diagnosis, deep learning (DL) algorithms have been developed using cardiac signals, but they often lack interpretability. Our study introduces a novel interpretable DL approach (SleepECG-Net) for directly estimating OSA severity in at-risk children. A combination of convolutional and recurrent neural networks (CNN-RNN) was trained on overnight electrocardiogram (ECG) signals. Gradient-weighted Class Activation Mapping (Grad-CAM), an eXplainable Artificial Intelligence (XAI) algorithm, was applied to explain model decisions and extract ECG patterns relevant to pediatric OSA. Accordingly, ECG signals from the semi-public Childhood Adenotonsillectomy Trial (CHAT, n=1610) and Cleveland Family Study (CFS, n=64), and the private University of Chicago (UofC, n=981) databases were used. OSA diagnostic performance reached 4-class Cohen's Kappa of 0.410, 0.335, and 0.249 in CHAT, UofC, and CFS, respectively. The proposal demonstrated improved performance with increased severity along with heightened cardiovascular risk. XAI findings highlighted the detection of established ECG features linked to OSA, such as bradycardia-tachycardia events and delayed ECG patterns during apnea/hypopnea occurrences, focusing on clusters of events. Furthermore, Grad-CAM heatmaps identified potential ECG patterns indicating cardiovascular risk, such as P, T, and U waves, QT intervals, and QRS complex variations. Hence, SleepECG-Net approach may improve pediatric OSA diagnosis by also offering cardiac risk factor information, thereby increasing clinician confidence in automated systems, and promoting their effective adoption in clinical practice.

A subject-based association network defines new pediatric sleep apnea phenotypes with different odds of recovery after treatment

2025-12-23T20:02:40Z

Background and objectives: Timely treatment of pediatric obstructive sleep apnea (OSA) can prevent or reverse neurocognitive and cardiovascular morbidities. However, whether distinct phenotypes exist and account for divergent treatment effectiveness remains unknown. In this study, our goal is threefold: i) to define new data- driven pediatric OSA phenotypes, ii) to evaluate possible treatment effectiveness differences among them, and iii) to assess phenotypic information in predicting OSA resolution. Methods: We involved 22 sociodemographic, anthropometric, and clinical data from 464 children (5–10 years old) from the Childhood Adenotonsillectomy Trial (CHAT) database. Baseline information was used to auto- matically define pediatric OSA phenotypes using a new unsupervised subject-based association network. Follow- up data (7 months later) were used to evaluate the effects of the therapeutic intervention in terms of changes in the obstructive apnea-hypopnea index (OAHI) and the resolution of OSA (OAHI < 1 event per hour). An explainable artificial intelligence (XAI) approach was also developed to assess phenotypic information as OSA resolution predictor at baseline. Results: Our approach identified three OSA phenotypes (PHOSA1-PHOSA3), with PHOSA2 showing significantly lower odds of OSA recovery than PHOSA1 and PHOSA3 when treatment information was not considered (odds ratios, OR: 1.64 and 1.66, 95 % confidence intervals, CI: 1.03–2.62 and 1.01–2.69, respectively). The odds of OSA recovery were also significantly lower in PHOSA2 than in PHOSA3 when adenotonsillectomy was adopted as treatment (OR: 2.60, 95 % CI: 1.26–5.39). Our XAI approach identified 79.4 % (CI: 69.9–88.0 %) of children reaching OSA resolution after adenotonsillectomy, with a positive predictive value of 77.8 % (CI: 70.3 %-86.0 %). Conclusions: Our new subject-based association network successfully identified three clinically useful pediatric OSA phenotypes with different odds of therapeutic intervention effectiveness. Specifically, we found that chil- dren of any sex, >6 years old, overweight or obese, and with enlarged neck and waist circumference (PHOSA2) have less odds of recovering from OSA. Similarly, younger female children with no enlarged neck (PHOSA3) have higher odds of benefiting from adenotonsillectomy.

Reevaluating performance in c-VEP BCIs: The impact of calibration time

2025-12-17T20:03:44Z

Code-modulated visual evoked potentials (c-VEP) have demonstrated high performance in non-invasive brain- computer interfaces (BCIs). Recently, research has begun to consider practical aspects such as visual comfort, where non-binary sequences and variations in the spatial frequency of stimuli play significant roles. However, calibration requirements remain underexplored in performance comparisons. This study aims to analyze a multi- variable tradeoff crucial to the practical application of c-VEP-based BCIs: decoding accuracy, decoding speed, and calibration time. Visual comfort is retrospectively evaluated using two pre-recorded datasets. Models were trained with increasing calibration cycles and tested across varying decoding times, depicting learning and decoding curves. The datasets comprised 32 healthy subjects, and featured different stimulus paradigms: plain non-binary stimuli and checkerboard-like binary stimuli with spatial frequency variations. Results showed that all conditions achieved over 97 % grand-averaged accuracy with sufficient calibration. However, a clear tradeoff emerged between calibration duration and performance. Achieving 95 % average accuracy within a 2 s decoding window required mean calibration durations of 28.7±19.0 s for binary stimuli, or 148.7±72.3 s for non-binary stimuli. The binary checkerboard-based condition with a spatial frequency of 1.2 c/º (C016) proved to be particularly effective, achieving over 95 % accuracy within 2 s decoding window using only 7.3 s of calibration, and reporting a significant improvement in visual comfort. A minimum calibration time of 1 min was considered essential to adequately estimate the brain response, critical in template-matching paradigms. In conclusion, achieving optimal c-VEP performance requires balancing calibration duration, decoding speed and accuracy, and visual comfort.

Are intrinsic neural timescales related to sensory processing? Evidence from abnormal behavioral states

2025-12-17T14:08:24Z

The brain exhibits a complex temporal structure which translates into a hierarchy of distinct neural timescales. An open question is how these intrinsic timescales are related to sensory or motor information processing and whether these dynamics have common patterns in different behavioral states. We address these questions by investigating the brain's intrinsic timescales in healthy controls, motor (amyotrophic lateral sclerosis, locked-in syndrome), sensory (anesthesia, unresponsive wakefulness syndrome), and progressive reduction of sensory processing (from awake states over N1, N2, N3). We employed a combination of measures from EEG resting-state data: auto-correlation window (ACW), power spectral density (PSD), and power-law exponent (PLE). Prolonged neural timescales accompanied by a shift towards slower frequencies were observed in the conditions with sensory deficits, but not in conditions with motor deficits. Our results establish that the spontaneous activity's intrinsic neural timescale is related to the neural capacity that specifically supports sensory rather than motor information processing in the healthy brain.