Hasil untuk "Nutritional diseases. Deficiency diseases"

Junzhi Ning, Jiashi Lin, Yingying Fang et al.

Multimodal large language models (MLLMs) have advanced clinical tasks for common conditions, but their performance on rare diseases remains largely untested. In rare-disease scenarios, clinicians often lack prior clinical knowledge, forcing them to rely strictly on case-level evidence for clinical judgments. Existing benchmarks predominantly evaluate common-condition, single-image settings, leaving multimodal and multi-image evidence integration under rare-disease data scarcity systematically unevaluated. We introduce MMRareBench, to our knowledge the first rare-disease benchmark jointly evaluating multimodal and multi-image clinical capability across four workflow-aligned tracks: diagnosis, treatment planning, cross-image evidence alignment, and examination suggestion. The benchmark comprises 1,756 question-answer pairs with 7,958 associated medical images curated from PMC case reports, with Orphanet-anchored ontology alignment, track-specific leakage control, evidence-grounded annotations, and a two-level evaluation protocol. A systematic evaluation of 23 MLLMs reveals fragmented capability profiles and universally low treatment-planning performance, with medical-domain models trailing general-purpose MLLMs substantially on multi-image tracks despite competitive diagnostic scores. These patterns are consistent with a capacity dilution effect: medical fine-tuning can narrow the diagnostic gap but may erode the compositional multi-image capability that rare-disease evidence integration demands.

John Wu, Adam Cross, Jimeng Sun

Rare diseases affect 1 in 10 Americans, yet standard ICD coding systems fail to capture these conditions in electronic health records (EHR), leaving crucial information buried in clinical notes. Current approaches struggle with medical abbreviations, miss implicit disease mentions, raise privacy concerns with cloud processing, and lack clinical reasoning abilities. We present Rare Disease Mining Agents (RDMA), a framework that mirrors how medical experts identify rare disease patterns in EHR. RDMA connects scattered clinical observations that together suggest specific rare conditions. By handling clinical abbreviations, recognizing implicit disease patterns, and applying contextual reasoning locally on standard hardware, RDMA reduces privacy risks while improving F1 performance by upwards of 30\% and decreasing inferences costs 10-fold. This approach helps clinicians avoid the privacy risk of using cloud services while accessing key rare disease information from EHR systems, supporting earlier diagnosis for rare disease patients. Available at https://github.com/jhnwu3/RDMA.

Yiming Lei, Michael Nguyen, Tzu Chia Liu et al.

Chest X-rays play a pivotal role in diagnosing respiratory diseases such as pneumonia, tuberculosis, and COVID-19, which are prevalent and present unique diagnostic challenges due to overlapping visual features and variability in image quality. Severe class imbalance and the complexity of medical images hinder automated analysis. This study leverages deep learning techniques, including transfer learning on pre-trained models (AlexNet, ResNet, and InceptionNet), to enhance disease detection and classification. By fine-tuning these models and incorporating focal loss to address class imbalance, significant performance improvements were achieved. Grad-CAM visualizations further enhance model interpretability, providing insights into clinically relevant regions influencing predictions. The InceptionV3 model, for instance, achieved a 28% improvement in AUC and a 15% increase in F1-Score. These findings highlight the potential of deep learning to improve diagnostic workflows and support clinical decision-making.

Kimberly F. Greco, Zongxin Yang, Mengyan Li et al.

Rare diseases affect an estimated 300-400 million people worldwide, yet individual conditions remain underdiagnosed and poorly characterized due to their low prevalence and limited clinician familiarity. Computational phenotyping offers a scalable approach to improving rare disease detection, but algorithm development is hindered by the scarcity of high-quality labeled data for training. Expert-labeled datasets from chart reviews and registries are clinically accurate but limited in scope and availability, whereas labels derived from electronic health records (EHRs) provide broader coverage but are often noisy or incomplete. To address these challenges, we propose WEST (WEakly Supervised Transformer for rare disease phenotyping and subphenotyping from EHRs), a framework that combines routinely collected EHR data with a limited set of expert-validated cases and controls to enable large-scale phenotyping. At its core, WEST employs a weakly supervised transformer model trained on extensive probabilistic silver-standard labels - derived from both structured and unstructured EHR features - that are iteratively refined during training to improve model calibration. We evaluate WEST on two rare pulmonary diseases using EHR data from Boston Children's Hospital and show that it outperforms existing methods in phenotype classification, identification of clinically meaningful subphenotypes, and prediction of disease progression. By reducing reliance on manual annotation, WEST enables data-efficient rare disease phenotyping that improves cohort definition, supports earlier and more accurate diagnosis, and accelerates data-driven discovery for the rare disease community.

Anika Islam, Tasfia Tahsin, Zaarin Anjum et al.

Accurate and timely identification of plant leaf diseases is essential for resilient and sustainable agriculture, yet most deep learning approaches rely on large annotated datasets and computationally intensive models that are unsuitable for data-scarce and resource-constrained environments. To address these challenges we present a few-shot learning approach within a lightweight yet efficient framework that combines domain-adapted MobileNetV2 and MobileNetV3 models as feature extractors, along with a feature fusion technique to generate robust feature representation. For the classification task, the fused features are passed through a Bi-LSTM classifier enhanced with attention mechanisms to capture sequential dependencies and focus on the most relevant features, thereby achieving optimal classification performance even in complex, real-world environments with noisy or cluttered backgrounds. The proposed framework was evaluated across multiple experimental setups, including both laboratory-controlled and field-captured datasets. On tomato leaf diseases from the PlantVillage dataset, it consistently improved performance across 1 to 15 shot scenarios, reaching 98.23+-0.33% at 15 shot, closely approaching the 99.98% SOTA benchmark achieved by a Transductive LSTM with attention, while remaining lightweight and mobile-friendly. Under real-world conditions using field images from the Dhan Shomadhan dataset, it maintained robust performance, reaching 69.28+-1.49% at 15-shot and demonstrating strong resilience to complex backgrounds. Notably, it also outperformed the previous SOTA accuracy of 96.0% on six diseases from PlantVillage, achieving 99.72% with only 15-shot learning. With a compact model size of approximately 40 MB and inference complexity of approximately 1.12 GFLOPs, this work establishes a scalable, mobile-ready foundation for precise plant disease diagnostics in data-scarce regions.

Ian Miles, Mayumi Wakimoto, Wagner Meira et al.

This review explores the integration of Artificial Intelligence into Horizon Scanning, focusing on identifying and responding to emerging threats and opportunities linked to Infectious Diseases. We examine how AI tools can enhance signal detection, data monitoring, scenario analysis, and decision support. We also address the risks associated with AI adoption and propose strategies for effective implementation and governance. The findings contribute to the growing body of Foresight literature by demonstrating the potential and limitations of AI in Public Health preparedness.

Cesar Campos-Ramírez, Nicolas Camacho-Calderon, Maria Elena Villagran-Herrera et al.

It has been proposed that the consumption of foods high in sugar or fat may cause addictive behavior. The aim of this study was to adapt and validate a soft drink addiction scale that can be used in future studies and to strengthen the proposal of food addiction with the hypothesis that people with high consumption of soft drinks have similar characteristics to people who consume abuse drugs. A non-probabilistic convenience sample of 394 Mexican participants answered a soft drinks’ consumption frequency questionnaire, an addiction scale, and a self-efficacy scale for soft drinks’ consumption. Additionally, anthropometric measurements were taken. The addiction scale showed three factors with an adequate reliability (Cronbach’s alpha coefficient = 0.903), as well as construct validity and criterion validity with the self-efficacy scale and total body fat percentage on soft drinks, mainly those with substantial caloric content. Additionally, the results showed a predictive value for soft drink consumption, strengthening its validity. This scale is useful to identify and evaluate the characteristic patterns of a substance addiction. The total reliability indicates that the items as a whole are correlated with each other and that the scale is stable to be used over time. This is the first study that evaluates the addictive characteristics of soft drink consumption through a scale, and it represents an advance in the exploration of the behavioral sciences field and an important tool for the creation of public health policies, mainly in countries with a high consumption of these beverages.

Yanxue Lian, Pincheng Luo

Abstract Background Despite the potential superiority of the non-high-density lipoprotein cholesterol to high-density lipoprotein cholesterol ratio (NHHR) as a diagnostic and predictive marker, no study has investigated the link between different leisure-time physical activity (LTPA) patterns and the NHHR. This study aims to explore this relationship. Methods Data was extracted from the National Health and Nutrition Examination Survey (NHANES) cycles spanning from 2007 to 2008 to 2017–2018. Participants (N = 14,211) were classified into four groups based on their LTPA patterns: (1) inactive (LTPA = 0 min/week); (2) insufficiently active (LTPA < 150 min/week); (3) weekend warrior (LTPA ≥ 150 min/week within 1 or 2 sessions); and (4) regularly active (LTPA ≥ 150 min/week in more than 2 sessions). Weighted multiple linear regression analysis was employed twice, using inactive and regular active groups as reference groups, respectively. Weighted stratification analyses and interaction tests were performed by demographics. Results Compared to the inactive group, each additional unit of LTPA time was associated with a significant 0.23-unit greater decrease in the NHHR in the regularly active group [-0.23 (-0.29; -0.16)]. However, no significant decrease was observed in the “Weekend Warrior” [-0.11 (-0.22; 0.008)] or insufficiently active groups [-0.03 (-0.11; 0.04)]. Moreover, compared to the regularly active group, the insufficiently active [0.21 (0.13; 0.29)], “Weekend Warrior” [0.13 (0.004; 0.25)], and inactive [0.26 (0.20; 0.32)] groups had significantly higher NHHR. The associations between the NHHR and various LTPA patterns did not significantly differ by demographic factors, except for race. Conclusion The regularly active pattern is significantly associated with a lower NHHR, but no significant difference in the NHHR was detected between the insufficiently active and “Weekend Warriors” patterns. The study suggests that frequency and regularity of PA are crucial for optimal lipid management, supporting clinical recommendations to meet or exceed 150 min of PA in more than two sessions per week.

Abdullah Al Mamun, Rafif Naufi Waskitha Hapsari

Not Mandatory

Ann E. Caldwell, Kimberly R. More, Tsz Kiu Chui et al.

Abstract Background Identifying as someone who engages in health promoting behaviors like healthy eating and exercising may be associated with sustained engagement in those behaviors, but reliable and valid instruments are needed to improve the rigor of this research. Two studies were conducted to (1) examine the psychometric properties of a four-item exerciser identity measure (4-EI) and an adapted healthy-eater identity measure (4-HEI) and (2) examine differences in identity strengths across categories of weight loss success. Methods Data from 1,709 community dwelling adults in the International Weight Control Registry (IWCR) were used. A random half of the sample was used to assess the proposed unidimensional factor structure of the 4-EI and 4-HEI and examine convergent and discriminant validity using Spearman rank-order correlations. One-way ANOVA was used in the other random half of the sample to compare 4-EI and 4-HEI scores (-3 to + 3) across three self-defined weight loss categories (‘Successful’, ‘Regain’, and ‘Unsuccessful’) and those maintaining ≥ 5% weight loss for > 1 year vs. not. Results Results support the unidimensional factor structure with all four items (eigenvalue scores > 2.89) as well as convergent and discriminant validity for both measures. Exercise identity was strongly correlated with self-reported physical activity (r (735) = 0.52, p <.001) and measures of autonomous motivation. Healthy eating identity was moderately correlated with cognitive restraint in eating (r (744) = 0.42, p <.001) and other measures predictive of eating behavior. 4-EI and 4-HEI are stronger in Successful (4-EI: M = 0.90, SD = 1.77; 4-HEI: M = 1.56 SD = 1.37) vs. Regain (4-EI: M=-0.18, SD = 1.68; 4-HEI: M =.57, SD = 1.48) and Unsuccessful (4-EI:M=-0.28, SD = 1.62; 4-HEI: M = 0.51, SD = 1.33) and those maintaining ≥ 5% weight loss (4-EI:M = 0.47, SD = 1.78; 4-HEI: M = 1.13, SD = 1.49) vs. not (4-EI:M=-0.27, SD = 1.66; 4-HEI: M = 0.53, SD = 1.47), p’s < 0.001. Conclusions The 4-EI and 4-HEI have acceptable psychometric properties and can advance understanding of the role of identity in exercise and dietary behaviors and weight loss maintenance. Trial registration The parent observational study, International Weight Control Registry (IWCR), for these sub-studies is registered in ClinicalTrials.gov (NCT04907396).

Yingzhou Lu, Yaojun Hu, Chenhao Li

Bringing a novel drug from the original idea to market typically requires more than ten years and billions of dollars. To alleviate the heavy burden, a natural idea is to reuse the approved drug to treat new diseases. The process is also known as drug repurposing or drug repositioning. Machine learning methods exhibited huge potential in automating drug repurposing. However, it still encounter some challenges, such as lack of labels and multimodal feature representation. To address these issues, we design DrugCLIP, a cutting-edge contrastive learning method, to learn drug and disease's interaction without negative labels. Additionally, we have curated a drug repurposing dataset based on real-world clinical trial records. Thorough empirical studies are conducted to validate the effectiveness of the proposed DrugCLIP method.

Thomas Thebaud, Anna Favaro, Casey Chen et al.

Motor dysfunction is a common sign of neurodegenerative diseases (NDs) such as Parkinson's disease (PD) and Alzheimer's disease (AD), but may be difficult to detect, especially in the early stages. In this work, we examine the behavior of a wide array of interpretable features extracted from the handwriting signals of 113 subjects performing multiple tasks on a digital tablet, as part of the Neurological Signals dataset. The aim is to measure their effectiveness in characterizing NDs, including AD and PD. To this end, task-agnostic and task-specific features are extracted from 14 distinct tasks. Subsequently, through statistical analysis and a series of classification experiments, we investigate which features provide greater discriminative power between NDs and healthy controls and amongst different NDs. Preliminary results indicate that the tasks at hand can all be effectively leveraged to distinguish between the considered set of NDs, specifically by measuring the stability, the speed of writing, the time spent not writing, and the pressure variations between groups from our handcrafted interpretable features, which shows a statistically significant difference between groups, across multiple tasks. Using various binary classification algorithms on the computed features, we obtain up to 87% accuracy for the discrimination between AD and healthy controls (CTL), and up to 69% for the discrimination between PD and CTL.

Tanzina Taher Ifty, Saleh Ahmed Shafin, Shoeb Mohammad Shahriar et al.

Lung diseases remain a critical global health concern, and it's crucial to have accurate and quick ways to diagnose them. This work focuses on classifying different lung diseases into five groups: viral pneumonia, bacterial pneumonia, COVID, tuberculosis, and normal lungs. Employing advanced deep learning techniques, we explore a diverse range of models including CNN, hybrid models, ensembles, transformers, and Big Transfer. The research encompasses comprehensive methodologies such as hyperparameter tuning, stratified k-fold cross-validation, and transfer learning with fine-tuning.Remarkably, our findings reveal that the Xception model, fine-tuned through 5-fold cross-validation, achieves the highest accuracy of 96.21\%. This success shows that our methods work well in accurately identifying different lung diseases. The exploration of explainable artificial intelligence (XAI) methodologies further enhances our understanding of the decision-making processes employed by these models, contributing to increased trust in their clinical applications.

Sushil Pathak, G. Shirisha, K. Venkata Ratnam

When an infectious disease propagates throughout society, the incidence function may rise at first due to an increase in pathogenicity and then decrease due to inhibitory effects until it reaches saturation. Effective vaccination and treatment are very helpful for controlling the effects of such infectious diseases. To analyze the impacts of these diseases, we proposed a new compartmental model with a generalized non-linear incidence function, vaccination function, and treatment function, along with time delays in the respective functions, which show how its monotonic features influence the stability of the model. Fundamental properties of a model, such as positivity, boundedness, and the existence of equilibria, are examined in this work. The basic reproduction number has been computed, and correlative studies for local stability in view of the basic reproduction number have been examined at the disease-free and endemic equilibrium points. A delay-independent global stability result has been established, and to be more precise, we explicitly derived the result on global stability by restricting delay parameters within a very specific range. Furthermore, numerical simulations and some examples based on COVID-19 real-time data are pointed out to emphasize the significance of how the disease's dynamical behavior is characterized by various functions for controlling the spread of disease in a population and to justify the mathematical conclusions.

AKM Enzam-Ul Haque, Golam Rabbany, Md. Siam

One of the most serious corneal disorders, keratoconus is difficult to diagnose in its early stages and can result in blindness. This illness, which often appears in the second decade of life, affects people of all sexes and races. Convolutional neural networks (CNNs), one of the deep learning approaches, have recently come to light as particularly promising tools for the accurate and timely diagnosis of keratoconus. The purpose of this study was to evaluate how well different D-CNN models identified keratoconus-related diseases. To be more precise, we compared five different CNN-based deep learning architectures (DenseNet201, InceptionV3, MobileNetV2, VGG19, Xception). In our comprehensive experimental analysis, the DenseNet201-based model performed very well in keratoconus disease identification in our extensive experimental research. This model outperformed its D-CNN equivalents, with an astounding accuracy rate of 89.14% in three crucial classes: Keratoconus, Normal, and Suspect. The results demonstrate not only the stability and robustness of the model but also its practical usefulness in real-world applications for accurate and dependable keratoconus identification. In addition, D-CNN DenseNet201 performs extraordinarily well in terms of precision, recall rates, and F1 scores in addition to accuracy. These measures validate the model's usefulness as an effective diagnostic tool by highlighting its capacity to reliably detect instances of keratoconus and to reduce false positives and negatives.

Patricia Thomas, Catherine Arden, Jenna Corcoran et al.

Abstract Background Rodent and human β-cells are differentially susceptible to the “lipotoxic” effects of long-chain saturated fatty acids (LC-SFA) but the factors accounting for this are unclear. Here, we have studied the intracellular disposition of the LC-SFA palmitate in human vs rodent β–cells and present data that reveal new insights into the factors regulating β-cell lipotoxicity. Methods The subcellular distribution of the LC-SFA palmitate was studied in rodent (INS-1E and INS-1 823/13 cells) and human (EndoC-βH1) β-cells using confocal fluorescence and electron microscopy (EM). Protein expression was assessed by Western blotting and cell viability, by vital dye staining. Results Exposure of INS-1 cells to palmitate for 24 h led to loss of viability, whereas EndoC-βH1 cells remained viable even after 72 h of treatment with a high concentration (1 mM) of palmitate. Use of the fluorescent palmitate analogue BODIPY FL C16 revealed an early localisation of the LC-SFA to the Golgi apparatus in INS-1 cells and this correlated with distention of intracellular membranes, visualised under the EM. Despite this, the PERK-dependent ER stress pathway was not activated under these conditions. By contrast, BODIPY FL C16 did not accumulate in the Golgi apparatus in EndoC-βH1 cells but, rather, co-localised with the lipid droplet-associated protein, PLIN2, suggesting preferential routing into lipid droplets. When INS-1 cells were treated with a combination of palmitate plus oleate, the toxic effects of palmitate were attenuated and BODIPY FL C16 localised primarily with PLIN2 but not with a Golgi marker. Conclusion In rodent β-cells, palmitate accumulates in the Golgi apparatus at early time points whereas, in EndoC- βH1 cells, it is routed preferentially into lipid droplets. This may account for the differential sensitivity of rodent vs human β-cells to “lipotoxicity” since manoeuvres leading to the incorporation of palmitate into lipid droplets is associated with the maintenance of cell viability in both cell types.

Mehreen Sirshar, Taimur Hassan, Muhammad Usman Akram et al.

Pulmonary diseases can cause severe respiratory problems, leading to sudden death if not treated timely. Many researchers have utilized deep learning systems to diagnose pulmonary disorders using chest X-rays (CXRs). However, such systems require exhaustive training efforts on large-scale data to effectively diagnose chest abnormalities. Furthermore, procuring such large-scale data is often infeasible and impractical, especially for rare diseases. With the recent advances in incremental learning, researchers have periodically tuned deep neural networks to learn different classification tasks with few training examples. Although, such systems can resist catastrophic forgetting, they treat the knowledge representations independently of each other, and this limits their classification performance. Also, to the best of our knowledge, there is no incremental learning-driven image diagnostic framework that is specifically designed to screen pulmonary disorders from the CXRs. To address this, we present a novel framework that can learn to screen different chest abnormalities incrementally. In addition to this, the proposed framework is penalized through an incremental learning loss function that infers Bayesian theory to recognize structural and semantic inter-dependencies between incrementally learned knowledge representations to diagnose the pulmonary diseases effectively, regardless of the scanner specifications. We tested the proposed framework on five public CXR datasets containing different chest abnormalities, where it outperformed various state-of-the-art system through various metrics.

Mohammad Mahdi Derakhshani, Ivona Najdenkoska, Tom van Sonsbeek et al.

Deep learning models have shown a great effectiveness in recognition of findings in medical images. However, they cannot handle the ever-changing clinical environment, bringing newly annotated medical data from different sources. To exploit the incoming streams of data, these models would benefit largely from sequentially learning from new samples, without forgetting the previously obtained knowledge. In this paper we introduce LifeLonger, a benchmark for continual disease classification on the MedMNIST collection, by applying existing state-of-the-art continual learning methods. In particular, we consider three continual learning scenarios, namely, task and class incremental learning and the newly defined cross-domain incremental learning. Task and class incremental learning of diseases address the issue of classifying new samples without re-training the models from scratch, while cross-domain incremental learning addresses the issue of dealing with datasets originating from different institutions while retaining the previously obtained knowledge. We perform a thorough analysis of the performance and examine how the well-known challenges of continual learning, such as the catastrophic forgetting exhibit themselves in this setting. The encouraging results demonstrate that continual learning has a major potential to advance disease classification and to produce a more robust and efficient learning framework for clinical settings. The code repository, data partitions and baseline results for the complete benchmark will be made publicly available.

Dipkamal Bhusal, Sanjeeb Prasad Panday

Traditional methods of identifying pathologies in X-ray images rely heavily on skilled human interpretation and are often time-consuming. The advent of deep learning techniques has enabled the development of automated disease diagnosis systems. Still, the performance of such systems is opaque to end-users and limited to detecting a single pathology. In this paper, we propose a multi-label disease prediction model that allows the detection of more than one pathology at a given test time. We use a dense convolutional neural network (DenseNet) for disease diagnosis. Our proposed model achieved the highest AUC score of 0.896 for the condition Cardiomegaly with an accuracy of 0.826, while the lowest AUC score was obtained for Nodule, at 0.655 with an accuracy of 0.66. To build trust in decision-making, we generated heatmaps on X-rays to visualize the regions where the model paid attention to make certain predictions. Our proposed automated disease prediction model obtained highly confident high-performance metrics in multi-label disease prediction tasks.

Samereh Abdoli, Monica S. V. M. Silveira, Mehri Doosti-Irani et al.