The systematic assessment of completeness of public metadata accompanying omics studies in the Gene Expression Omnibus data repository.

Background: Recent advances in high-throughput sequencing technologies have enabled the collection and sharing of a massive amount of omics data, along with its associated metadata-descriptive information that contextualizes the data, including phenotypic traits and experimental design. Enhancing metadata availability is critical to ensure data reusability and reproducibility and to facilitate novel biomedical discoveries through effective data reuse. Yet, incomplete metadata accompanying public omics data may hinder reproducibility and reusability and limit secondary analyses.

Wearable privacy.

Wearable technology has a promising potential to transform users' lives by continuously collecting data and providing convenient services on demand. Yet, there is also a large potential to breach users' privacy compromising the confidentiality of sensitive data. The lack of privacy regulations is caused by a limited understanding of how to control data collection, access and sharing.

Wearable biosensing devices for mental health, wellness, and stress management.

Data gathering for diagnostic purposes often relies on psychological instruments and validated tests applied individually through in person interviews. Such an approach is limited since it relies on a subjective perception of the individual as well as their abilities to recall information concerning their behaviors, thoughts, and feelings. Thus, the accuracy of the assessment tends to be unreliable and prone to bias, stigma, as well as subjective interpretations.

Inter-modality feature prediction through multimodal fusion for 3D shape defect detection.

3D shape defect detection plays an important role in autonomous industrial inspection. However, accurate detection of anomalies remains challenging due to the complexity of multimodal sensor data, especially when both color and structural information are required. In this work, we propose a lightweight inter-modality feature prediction framework that effectively utilizes multimodal fused features from the inputs of RGB, depth and point clouds for efficient 3D shape defect detection.

ACG-SFE: Adaptive cluster-guided simple, fast, and efficient feature selection for high-dimensional microarray data in binary classification.

Advances in data collection have resulted in an exponential growth of high-dimensional microarray datasets for binary classification in bioinformatics and medical diagnostics. These datasets generally possess many features but relatively few samples, resulting in challenges associated with the "curse of dimensionality", such as feature redundancy and an elevated risk of overfitting. While traditional feature selection approaches, such as filter-based and wrapper-based approaches, can help to reduce dimensionality, they often struggle to capture feature interactions while adequately preserving model generalization.

Oral-Anatomical Knowledge-Informed Semi-Supervised Learning for 3D Dental CBCT Segmentation and Lesion Detection.

Cone beam computed tomography (CBCT) is a widely-used imaging modality in dental healthcare. It is an important task to segment each 3D CBCT image, which involves labeling lesions, bone, teeth, and restorative material on a voxel-by-voxel basis, as it aids in lesion detection, diagnosis, and treatment planning. The current clinical practice relies on manual segmentation, which is labor-intensive and demands considerable expertise.

Adaptive individualized gene pair signatures distinguishing melanoma and predicting response to immune checkpoint blockade.

Distinguishing similar cancer subtypes and predicting responses to immune checkpoint blockade (ICB) are critical for improving clinical outcomes. However, existing gene expression signatures often suffer from batch effects and poor generalizability across cohorts. To address these limitations, we propose adaptive individualized gene pair signatures (AIGPS), a robust method that adaptively quantifies gene pair reversals and selects informative features using machine learning.

Multimodal deep learning for predicting protein ubiquitination sites.

Motivation: Ubiquitination is a crucial post-translational modification that regulates various biological functions, including protein degradation, signal transduction, and cellular homeostasis. Accurate identification of ubiquitination sites is essential for understanding these mechanisms, yet existing prediction tools often lack generalizability across diverse datasets. To address this limitation, we developed Multimodal Ubiquitination Predictor, a deep learning-based approach capable of predicting ubiquitination sites across general, human-specific, and plant-specific datasets.

SelfReg-UNet: Self-Regularized UNet for Medical Image Segmentation.

Since its introduction, UNet has been leading a variety of medical image segmentation tasks. Although numerous follow-up studies have also been dedicated to improving the performance of standard UNet, few have conducted in-depth analyses of the underlying interest pattern of UNet in medical image segmentation. In this paper, we explore the patterns learned in a UNet and observe two important factors that potentially affect its performance: (i) irrelative feature learned caused by asymmetric supervision; (ii) feature redundancy in the feature map.

Omics-driven insights into the molecular pathways driving osteoarthritis pathogenesis.

Osteoarthritis (OA) is a multifactorial, mechano-inflammatory joint disorder characterized by cartilage degradation, synovial inflammation, and subchondral bone remodeling. Despite its high prevalence and significant impact on quality of life, no disease-modifying treatments have been approved. In many other disease areas, advanced omics technologies are impacting the development of advanced therapies.

Toward a Genetic Signature of Resistance to Activity-Based Anorexia in Striatal Projecting Cortical Neurons.

Objective: Converging evidence from neuroimaging studies and genome-wide association study (GWAS) suggests the involvement of prefrontal cortex (PFC) and striatum dysfunction in the pathophysiology of anorexia nervosa (AN). However, identifying the causal role of circuit-specific genes in the development of the AN-like phenotype remains challenging and requires the combination of novel molecular tools and preclinical models.

Methods: We used the activity-based anorexia (ABA) rat model in combination with a novel viral-based translating ribosome affinity purification (TRAP) technique to identify transcriptional differences within a specific neural pathway that we have previously demonstrated to mediate pathological weight loss in ABA rats (i.

Adapformer: Adaptive channel management for multivariate time series forecasting.

In multivariate time series forecasting (MTSF), accurately modeling the intricate dependencies among multiple variables remains a significant challenge due to the inherent limitations of traditional approaches. Most existing models adopt either channel-independent (CI) or channel-dependent (CD) strategies, each presenting distinct drawbacks. CI methods fail to leverage the potential insights from inter-channel interactions, resulting in models that may not fully exploit the underlying statistical dependencies present in the data.

Latent profile analysis of healthcare-related regret coping among Master of Nursing specialist students during clinical internships: A multi-center cross-sectional study.

Background: Healthcare-related regret (HRR) is frequently encountered by healthcare professionals, even in the early clinical stages. Effective coping strategies are essential for mental well-being, professional performance, and career satisfaction. However, the specific coping mechanisms used by Master of Nursing specialist (MNS) students during clinical internships are not well understood.

Directed Structural Evolution of Nickel Nanoparticles into Atomically Dispersed Sites for Efficient CO Electroreduction.

Electrochemical CO reduction (CORR) to carbon monoxide (CO) offers a sustainable pathway for carbon utilization, yet challenges remain in terms of improving selectivity and activity. Herein, we report a Ni/NC catalyst synthesized via a milling - pyrolysis method, in which Ni particles anchored on nitrogen-doped carbon (NC) are electrochemically activated under an Ar atmosphere, leading to their structural evolution into single-atom Ni sites. After activation in Ar atmosphere, the current density nearly doubles (from ≈30 to ≈60 mA cm), and concurrently, the Faradaic efficiency of CO stays at ∼90% with the potential set to -0.

Unlocking the power of L1 regularization: A novel approach to taming overfitting in CNN for image classification.

Convolutional Neural Networks (CNNs) stand as indispensable tools in deep learning, capable of autonomously extracting crucial features from diverse data types. However, the intricacies of CNN architectures can present challenges such as overfitting and underfitting, necessitating thoughtful strategies to optimize their performance. In this work, these issues have been resolved by introducing L1 regularization in the basic architecture of CNN when it is applied for image classification.

Tailoring the Effects of Spin State and Intermediate Hydrogen Adsorption on NiPt/Ni Bridge Sites toward Robust Acidic Water Electrolysis.

Precise modulation of the electronic structure in transition metals, particularly the d-band center position and spin state, remains a critical challenge to expediting hydrogen evolution reaction (HER) kinetics. Herein, we report a NiPt/Ni-heterostructured catalyst enabling simultaneous optimization of the d-band electronic structure and spin state of Ni through regulation of the NiPt and Ni bridge sites. Combining operando spectroscopy, X-ray absorption spectroscopy, density functional theory, and ab initio molecular dynamics simulations, we establish that the coordination environment and spin states of Ni at the bridge sites were effectively modulated by altering the Pt content, achieving a transition of Ni centers from the low-spin to high-spin state, and optimized intermediate adsorption/desorption behaviors.

Prostate MR image segmentation using a multi-stage network approach.

Prostate cancer (PCa) remains one of the most prevalent cancers among men, with over 1.4 million new cases and 375,304 deaths reported globally in 2020. Current diagnostic approaches, such as prostate-specific antigen (PSA) testing and trans-rectal ultrasound (TRUS)-guided biopsies, are often Limited by low specificity and accuracy.

Genetic and Cellular Architecture of Breast Cancer Risk in Multi-Ancestry Studies of 159,297 Cases and 212,102 Controls.

Breast cancer genome-wide association studies (GWAS) have identified over 200 independent genome-wide significant susceptibility markers. However, most studies have focused on one or two ancestral groups. We examined breast cancer genetic architecture using GWAS summary statistics from African (AFR), East Asian (EAS), European (EUR) and Hispanic/Latina (H/L) samples, totaling 159,297 cases and 212,102 controls, comprising the largest multi-ancestry study of breast cancer to date.

Change point detection to analyze air pollution and its economic effects: an exponentially weighted moving average perspective.

Air pollution has a direct impact on every society, leading to consequential effects on the economy of a nation. Poor air quality adversely affects human health, resulting in various economic outcomes such as rising healthcare costs, diminished labor productivity, negative impacts on tourism and living standards, increased regulatory expenses for businesses, and heightened economic disparities. Effective control methods are essential to monitor factors influencing the economy, including air quality.

Superficial and deep white matter abnormalities in temporal lobe epilepsy.

Non-invasive neuroimaging is important in epilepsy to help identify cerebral abnormalities. Abnormally reduced fractional anisotropy (FA) in deep white matter (WM) from diffusion-weighted imaging (DWI) is widely reported in large multi-cohort studies across all types of epilepsies. However, abnormalities in FA for superficial WM are rarely investigated in epilepsy.

Estimating resting energy expenditure in severely obese patients with laparoscopic sleeve gastrectomy: Normalized lean weight appears to be more accurate-A prospective observational study.

Background: The accurate estimation of energy requirements in severely obese patients poses a significant challenge. This study compared the measured resting energy expenditure (mREE) obtained using indirect calorimetry with the estimated energy target in severely obese patients who underwent laparoscopic sleeve gastrectomy.

Methods: This study enrolled patients who underwent elective bariatric surgery for metabolic syndrome and were admitted to the ICU between September 2023 and October 2024 because of severe clinical complications.

Privacy-preserving dementia classification from EEG via hybrid-fusion EEGNetv4 and federated learning.

As global life expectancy rises, a growing proportion of the population is affected by dementia, particularly Alzheimer's disease (AD) and Frontotemporal dementia (FTD). Electroencephalography (EEG) based diagnosis presents a non-invasive, cost effective alternative for early detection, yet existing methods are challenged by data scarcity, inter-subject variability, and privacy concerns. This study proposes lightweight and privacy-preserving EEG classification framework combining deep learning and Federated Learning (FL).