Incremental Learning for Patient-Specific EEG-Based Seizure Detection.

Real-time monitoring of fracture critical steel bridges can potentially enhance inspection practices by tracking the behavior of the bridge. Significant advances have occurred in recent years on the development of robust hardware for field monitoring applications. These systems can monitor, process, and store data from a variety of sensors (e.g. strain gages, crack propagation gages etc.) to track changes in the behavior of the bridge. Thus, for a long-term monitoring system to be successful, the reliability of gages that are to be monitored for several years is very important. This paper focuses on the results of a research study focused on developing a wireless monitoring system with a useful life of more than 10 years. An important aspect of the study is to identify strain gages and installation procedures that result in long lives as well as characterizing the effect of temperature fluctuations and other environmental factors on the sensor drift and noise. In long-term monitoring applications, slight sensor drift and noise can build up over time to produce misleading results. Thus, a wide variety of gages that can be used to monitor bridges have been tested for over a year through environmental tests. The environmental tests were developed to determine the durability of the gages and their protective coatings (e.g. zinc-based spray, wax and silicon, etc.) against humidity, sun exposure and other environmental effects that are expected in long-term bridge monitoring applications. Moreover, fatigue tests were performed to determine the fatigue category of the weldable gages and to reveal any debonding issues of the bondable gages. This paper focuses on the results of laboratory tests on gage durability that were conducted as part of a research project sponsored by the National Institute of Standards and Technology (NIST).

Minimizing caging effects in murine lung microbiome studies.

With the advent of parameter-efficient fine-tuning techniques for pre-trained vision-language models, interest in adapting them for various incremental learning scenarios has grown, i.e., sequential increments on task, class, and domain. However, no high-performance incremental learning framework has integrated these three incremental scenarios to achieve Unified Incremental Learning (Uni-IL) in complex settings. In this work, we propose an incremental learning framework called Mixture of Attribute-Guided Experts (MAGE) to alleviate the long-term forgetting in vision-language model incremental learning. Our approach involves acquiring image attribute knowledge via LLMs to form an attribute pool. We match the most relevant attributes as inputs to the Mixture of Experts (MoE) to fine-tune the pre-trained CLIP. Then the expert routers learn to select specific expert combinations based on the data and attribute features, alleviating catastrophic forgetting. The attribute pool incorporates both domain and class knowledge, enabling our approach to adapt to the three types of incremental learning scenarios and thus facilitating unified incremental learning. Through extensive experiments on our newly proposed benchmark and existing incremental learning scenarios, the results demonstrate that our proposed method not only performs well on the new Uni-IL tasks but also consistently outperforms previous state-of-the-art methods. Source code is available at https://github.com/ElectricField/Uni-IL.

Strategies to Sustain Quality Interventions: Case Examples of Chlorhexidine Gluconate Bathing.

Chapter 6 - Genetic and environmental factors of gluten-related disorders

ABSTRACTIntroduction: There is considerable variability in the rates and extent of drug metabolism between patients due to physiological, genetic, pharmacologic, environmental and nutritional factors such as fasting. This variability in drug metabolism may result in treatment failure or, conversely, in increased side effects or toxicity. Preclinical studies have shown that fasting alters drug metabolism by modulating the activity of drug metabolizing enzymes involved. However, until recently little was known about the effects of fasting on drug metabolism in humans.Areas covered: This review describes the effects of fasting on drug metabolism based on both preclinical studies and studies performed in humans.Expert opinion: A better understanding of the effects of fasting may improve the efficacy and safety of pharmacotherapy for individual patients. Fasting contributes to variability in human drug metabolism by differentially affecting drug metabolizing enzymes. Although the effects of fasting on drug metabolism appear to be small (between 10-20%), fasting may be relevant for drugs with a small therapeutic range and/or in combination with other factors that contribute to variability in drug metabolism such as physiological, genetic or pharmacological factors. Therefore, additional research on this topic is warranted.

Cleft lip and palate is an anomaly that affects both women and men. It is considered to be among the most frequent congenital abnormalities and is related to modifications in chromosomal DNA and multiple genetic alterations. This anomaly can also be associated with various environmental factors, such as tobacco and alcohol consumption, medication use, and exposure to different environmental and industrial toxic substances. The objective of this study was to document the frequency of risk factors related to cleft lip and palate through a systematic review of Mexican studies. In this systematic review, a bibliographic search was conducted following PRISMA guidelines in the databases Scielo, ScienceDirect, PubMed, and EBSCO. Keywords related to cleft lip and palate, epidemiology, and risk factors were used. In all, 3 independent reviewers (J.A.S.L., S.L.V., and N.M.F.) selected and evaluated a total of 17 articles included in this analysis, achieving a coefficient of κ = 0.84. The analysis revealed that the highest frequency of conducted studies was in the State of Mexico. The most common risk factors identified were environmental, pharmacological, consumption habits, and gynecological factors. Identifying the main risk factors for cleft lip and palate in the Mexican population will enable the implementation of preventive measures aimed at reducing exposure to these factors. Additionally, early intervention can improve the quality of life for individuals affected by this condition.

This paper presents an adaptive ensemble framework for real-time anomaly detection in large-scale data streams, addressing the challenges of concept drift, high-velocity data processing, and computational efficiency in modern distributed systems. We propose a Hybrid Statistical-Machine Learning Anomaly Detection (HSML-AD) algorithm that combines sliding window-based statistical analysis with incremental machine learning techniques. The framework employs a three-tier architecture: (1) lightweight statistical pre-filtering using modified Z-score and interquartile range methods, (2) adaptive feature extraction through exponential moving averages, and (3) ensemble classification using online random forest with dynamic weight adjustment based on recent prediction accuracy. Experimental evaluation on five benchmark datasets (KDD Cup 99, NSL-KDD, CICIDS2017, Yahoo S5, and Numenta Anomaly Benchmark) demonstrates that HSML-AD achieves an average F1-score of 94.3%, precision of 93.8%, and recall of 94.7%, outperforming baseline methods including Isolation Forest (F1: 87.2%), LSTM-Autoencoder (F1: 89.6%), and SPOT (F1: 86.4%). The algorithm maintains processing throughput of 127,000 records per second with average latency of 7.8 milliseconds on commodity hardware. The novelty lies in the adaptive weight mechanism that dynamically adjusts ensemble components based on data stream characteristics and recent performance, coupled with a memory-efficient incremental learning strategy that limits model size to 45 MB while maintaining detection accuracy. The proposed framework is applicable to network intrusion detection, IoT sensor monitoring, financial fraud detection, and industrial system health monitoring, particularly in resource-constrained environments requiring real-time processing.

Contrast-enhanced computed tomography (CE-CT) is the gold standard for diagnosing aortic dissection (AD). However, contrast agents can cause allergic reactions or renal failure in some patients. Moreover, AD diagnosis by radiologists using non-contrast-enhanced CT (NCE-CT) images has poor sensitivity. To address this issue, we propose a novel cascaded multi-task generative framework for AD detection using NCE-CT volumes. The framework includes a 3D nnU-Net and a 3D multi-task generative architecture (3D MTGA). Specifically, the 3D nnU-Net was employed to segment aortas from NCE-CT volumes. The 3D MTGA was then employed to simultaneously synthesize CE-CT volumes, segment true & false lumen, and classify the patient as AD or non-AD. A theoretical formulation demonstrated that the 3D MTGA could increase the Jensen-Shannon Divergence (JSD) between AD and non-AD for each NCE-CT volume, thus indirectly improving the AD detection performance. Experiments also showed that the proposed framework could achieve an average accuracy of 0.831, a sensitivity of 0.938, and an F1-score of 0.847 in comparison with seven state-of-the-art classification models used by three radiologists with junior, intermediate, and senior experiences, respectively. The experimental results indicate that the proposed framework obtains superior performance to state-of-the-art models in AD detection. Thus, it has great potential to reduce the misdiagnosis of AD using NCE-CT in clinical practice. The source codes and supplementary materials for our framework are available at https://github.com/yXiangXiong/CMTGF.

Long-term monitoring data on the evolution of soil organic carbon (SOC) content and soil stocks are crucial considering climate change and carbon sequestration. Due to the slow processes involved, it takes longer time spans until changes in SOC stocks become detectable. Thus, long-term monitoring is essential.Beside measurements, models are often used to report changes in SOC stocks for greenhouse gas (GHG) inventories. Consistent field data over extended periods are essential for validating such models whereby data from long-term monitoring is of great value. We present SOC measurements from the topsoil (uppermost 0.2 m) of 24 grassland sites of the Swiss Soil Monitoring Network (NABO) for six sampling campaigns from 1985 to 2014 and compare their evolution with predictions by the RothC model, which, for example, is used for the Swiss GHG Inventory. While there was no general temporal trend of the measured data over all sites, from a single site perspective, SOC stocks decreased/increased on 8% and 25% of the sites respectively. When comparing the measured data with the predictions from modelled values from RothC, a good fit was achieved in two thirds of the cases. However, at 5 out of the 24 sites (21%), the modelled SOC changes over time were lower than the measured ones. For 3 out of the 24 sites (12%) the model predicted higher SOC stock increases than were measured. Overall the modelled data fit well together with the measured ones. Most of the sites revealed stable values over time or increasing trends between 1985 and 2014, which might indicate that the SOC stocks potentially are in equilibrium and that the farming management used in Switzerland is suited to maintain these stocks in topsoils. There might not be a big potential for C sequestration, however, it is also very important to maintain SOC stocks over time. For future climate change scenarios accurate long-term SOC data and modelling are important for gathering the required information.

A deep learning approach for objective evaluation of microscopic neuro-drilling craniotomy skills.

Class incremental learning with analytic learning for hyperspectral image classification

A computational scaling analysis of multiobjective evolutionary algorithms in long-term groundwater monitoring applications

Current software defect prediction and code quality assessment methods treat these inherently related tasks independently, failing to leverage their complementary information. Existing graph-based approaches lack the ability to jointly model structural dependencies and quality characteristics, limiting their effectiveness in capturing the complex relationships between defect patterns and code quality indicators. This paper proposes a novel integrated model that simultaneously tackles both objectives using graph neural networks to leverage the inherent graph structure of software systems. Our novelty lies in the first-of-its-kind integration of multi-level graph representations (AST, CFG, DFG) with a dual-branch attention-based GNN architecture for simultaneous defect prediction and quality assessment. Our approach constructs multi-level graph representations by integrating abstract syntax trees, control flow graphs, and data flow graphs, capturing both syntactic and semantic relationships in source code. The proposed dual-branch GNN architecture employs shared representation learning with attention mechanisms and multi-task optimization to exploit complementary information between defect prediction and quality assessment tasks. Comprehensive experiments on six real-world software projects demonstrate significant improvements over traditional methods, achieving F1-scores of 0.811 and AUC values of 0.896 for defect prediction, while showing 9.3% average improvement in code quality assessment accuracy across multiple quality dimensions. The integration strategy proves effective in capturing complex structural dependencies and provides actionable insights for software development teams, establishing a foundation for intelligent software engineering tools that deliver comprehensive code analysis capabilities.

Residence time, site fidelity and movements for red snapper Lutjanus campechanus (Lutjanidae) were estimated from long-term telemetry monitoring (December 2005 to June 2010) at 6 sites (1 natural and 5 artificial reefs) in the northeastern Gulf of Mexico. Each site consisted of an array of 5 receivers, with 1 at the center reef site and 4 receivers placed 1100 m north, south, east and west of center (2 km radius detection area). Throughout the study, a stationary control transmitter was located 400 m south of the center receiver at each site to estimate changes in detection rates from environmental factors. These receiver arrays enabled fishery independent estimations of fishing mortality, natural mortality and emigration. Event analysis was used to esti- mate site fidelity and residence, based on right-censoring fishing and natural mortality. Median residence time was 542 d, ranging from 1 to 1099 d, with 72% of fish staying at least 1 yr at the site. Some fish (n = 12) showed seasonal and directed movements to other sites (up to 8 km away) and returned to original sites up to 7 mo later. Diel movements away from the structure tended to occur at night. Site fidelity and residence times of red snapper in the present study were greater than in any previous study and show the importance of artificial reefs for this species.