Search Results (448)

Background and Objectives: Public health needs collaborative, privacy-preserving analytics, but centralized AI is constrained by data sharing and governance. Federated learning (FL) enables training without moving sensitive data. This review assessed how FL is used for disease prevention in population and public health, and mapped benefits, challenges, and policy implications. Methods: Following PRISMA 2020, we searched PubMed, Scopus, Web of Science, IEEE Xplore, and Google Scholar for peer reviewed English-language studies from January 2020–30 June 2025, applying FL to surveillance, outbreak detection, risk prediction, or policy support. Two reviewers screened and extracted data with third-reviewer arbitration. Quality was appraised with a tool adapted from MMAT and AI reporting frameworks. No meta-analysis was performed. Results: Of 5230 records identified (4720 after deduplication), 200 full texts were assessed and 19 were included. Most used horizontal FL across multiple institutions for communicable diseases, COVID-19, tuberculosis and some chronic conditions. Reported gains included privacy preservation across sites, better generalizability from diverse data, near real-time intelligence, localized risk stratification, and support for resource planning. Common barriers were non-IID data, interoperability gaps, compute and network limits in low-resource settings, unclear legal pathways, and concerns about fairness and transparency. Few studies linked directly to formal public-health policy or low-resource deployments. Conclusions: FL is promising for equitable, secure, and scalable disease-prevention analytics that respect data sovereignty. Priorities include robust methods for heterogeneity, interoperable standards, secure aggregation, routine fairness auditing, clearer legal and regulatory guidance, and capacity building in underrepresented regions. Full article

This paper presents a systematic evaluation of large language models (LLMs) and retrieval-augmented generation (RAG) approaches for question answering (QA) in the low-resource Kazakh language. We assess the performance of existing proprietary (GPT-4o, Gemini 2.5-flash) and open-source Kazakh-oriented models (KazLLM-8B, Sherkala-8B, Irbis-7B) across closed-book and RAG settings. Within a three-stage evaluation framework we benchmark retriever quality, examine LLM abilities such as knowledge-gap detection, external truth integration and context grounding, and measures gains from realistic end-to-end RAG pipelines. Our results show a clear pattern: proprietary models lead in closed-book QA, but RAG narrows the gap substantially. Under the Ideal RAG setting, KazLLM-8B improves from its closed-book baseline of 0.427 to reach answer correctness of 0.867, closely matching GPT-4o’s score of 0.869. In the end-to-end RAG setup, KazLLM-8B paired with Snowflake retriever achieved answer correctness up to 0.754, surpassing GPT-4o’s best score of 0.632. Despite improvements, RAG outcomes show an inconsistency: high retrieval metrics do not guarantee high QA system accuracy. The findings highlight the importance of retrievers and context grounding strategies in enabling open-source Kazakh models to deliver competitive QA performance in a low-resource setting. Full article

Metabolic dysfunction-associated steatotic liver disease (MASLD) is defined as steatotic liver disease with at least one cardiometabolic risk factor, in the absence of harmful alcohol intake, and includes a spectrum of conditions. These range from isolated liver steatosis to metabolic dysfunction-associated steatohepatitis (MASH), fibrosis, cirrhosis, and MASH-related hepatocellular carcinoma. Patients with MASLD and type 2 diabetes are at increased risk of developing MASH and significant/advanced fibrosis. The severity of fibrosis is a key determinant of long-term prognosis in MASLD. The most recent AASLD and EASL-EASD-EASO Guidelines on the Management of MASLD recommend a step-by-step approach to identify patients at higher risk of fibrotic progression. Recent epidemiological trends highlight the socioeconomic impact of MASLD and MASH, particularly in middle- and low-income countries. Given the high cost of new targeted therapies, implementing effective treatment strategies in low-resource settings is essential in managing MASLD and MASH patients. Pioglitazone is an oral antidiabetic agent of the thiazolidinedione class that targets peroxisome proliferator-activated receptors activated by fatty acids and derivatives or pharmacological agonists and involved in lipid metabolism, cell differentiation, and inflammation. Pioglitazone treatment is a potential cost-effective option, particularly for low-resource settings. This review examines recent epidemiological trends in MASLD and MASH, outlines the mechanisms of action of pioglitazone with an emphasis on its role in improving liver fibrosis, and summarizes clinical studies on fibrosis evaluation during pioglitazone treatment. The literature search focused on English-language studies from the past two years in the PubMed database. Full article

Artificial Intelligence (AI) is a key technological enabler for the transition of agricultural production and management from experience-driven to data-driven, continuously advancing modern agriculture toward smart agriculture. This evolution ultimately aims to achieve a precise agricultural production model characterized by low resource consumption, high safety, high quality, high yield, and stable, sustainable development. Although machine learning, deep learning, computer vision, Internet of Things, and other AI technologies have made significant progress in numerous agricultural production applications, most studies focus on singular agricultural scenarios or specific AI algorithm research, such as object detection, navigation, agricultural machinery maintenance, and food safety, resulting in relatively limited coverage. To comprehensively elucidate the applications of AI in agriculture and provide a valuable reference for practitioners and policymakers, this paper reviews relevant research by investigating the entire agricultural production process—including planting, management, and harvesting—covering application scenarios such as seed selection during the cultivation phase, pest and disease identification and intelligent management during the growth phase, and agricultural product grading during the harvest phase, as well as agricultural machinery and devices like fault diagnosis and predictive maintenance of agricultural equipment, agricultural robots, and the agricultural Internet of Things. It first analyzes the fundamental principles and potential advantages of typical AI technologies, followed by a systematic and in-depth review of the latest progress in applying these core technologies to smart agriculture. The challenges faced by existing technologies are also explored, such as the inherent limitations of AI models—including poor generalization capability, low interpretability, and insufficient real-time performance—as well as the complex agricultural operating environments that result in multi-source, heterogeneous, and low-quality, unevenly annotated data. Furthermore, future research directions are discussed, such as lightweight network models, transfer learning, embodied intelligent agricultural robots, multimodal perception technologies, and large language models for agriculture. The aim is to provide meaningful insights for both theoretical research and practical applications of AI technologies in agriculture. Full article

Speech recognition, as a key driver of artificial intelligence and global communication, has advanced rapidly in major languages, while studies on low-resource languages remain limited. Tongan, a representative Polynesian language, carries significant cultural value. However, Tongan speech recognition faces three main challenges: data scarcity, limited adaptability of transfer learning, and weak dictionary modeling. This study proposes improvements in adaptive transfer learning and NBPE-based dictionary modeling to address these issues. An adaptive transfer learning strategy with layer-wise unfreezing and dynamic learning rate adjustment is introduced, enabling effective adaptation of pretrained models to the target language while improving accuracy and efficiency. In addition, the MEA-AGA is developed by combining the Mind Evolutionary Algorithm (MEA) with the Adaptive Genetic Algorithm (AGA) to optimize the number of byte-pair encoding (NBPE) parameters, thereby enhancing recognition accuracy and speed. The collected Tongan speech data were expanded and preprocessed, after which the experiments were conducted on an NVIDIA RTX 4070 GPU (16 GB) using CUDA 11.8 under the Ubuntu 18.04 operating system. Experimental results show that the proposed method achieved a word error rate (WER) of 26.18% and a word-per-second (WPS) rate of 68, demonstrating clear advantages over baseline methods and confirming its effectiveness for low-resource language applications. Although the proposed approach demonstrates promising performance, this study is still limited by the relatively small corpus size and the early stage of research exploration. Future work will focus on expanding the dataset, refining adaptive transfer strategies, and enhancing cross-lingual generalization to further improve the robustness and scalability of the model. Full article

Whisper is a transformer-based multilingual model that has illustrated state-of-the-art behavior in numerous languages. However, the efficiency remains persistent with the limited computational resources. To address this issue, an experiment was performed on librispeech-train-clean-100 for training purposes. The test-clean set was utilized to evaluate its performance. To enhance efficiency and to cater the computational needs, a parameter-efficient fine-tuning technique, i.e., Low-Rank Adaptation, was employed to add a limited number of trainable parameters into the frozen layers of the model. The results showed that Low-Rank Adaptation attained excellent Automatic Speech Recognition results while using fewer computational resources, showing its effectiveness for resource-saving adaptation. The research work emphasizes the promise of Low-Rank Adaptation as a lightweight and scalable fine-tuning strategy for large speech models using a transformer architecture. The baseline Whisper Small model achieved a word error rate of 16.7% without any parameter-efficient adaptation. In contrast, the Low-Rank Adaptation enhanced fine-tuned model achieved a lower word error rate of 6.08%, demonstrating the adaptability of the proposed parameter-efficient approach. Full article

Breast cancer is a leading cause of mortality among women, disproportionately affecting Hispanic populations in the U.S., particularly those with limited health literacy and language access. To address these disparities, we present a bilingual, web-based educational platform tailored to low-literacy Hispanic users. The platform supports full navigation in English and Spanish, with seamless language switching and both written and spoken input options. It incorporates automatic speech recognition (ASR) capable of handling code-switching, enhancing accessibility for bilingual users. Educational content is delivered through culturally sensitive videos organized into four categories: prevention, detection, diagnosis, and treatment. Each video includes embedded and post-video assessment questions aligned with Bloom’s Taxonomy to foster active learning. Users can monitor their progress and quiz performance via a personalized dashboard. An integrated chatbot, powered by large language models (LLMs), allows users to ask foundational breast cancer questions in natural language. The platform also recommends relevant resources, including nearby treatment centers, and support groups. LLMs are further used for ASR, question generation, and semantic response evaluation. Combining language technologies and LLMs reduces disparities in cancer education and supports informed decision-making among underserved populations, playing a pivotal role in reducing information gaps and promoting informed healthcare decisions. Full article

Reference-less evaluation of machine translation, or Quality Estimation (QE), is vital for low-resource language pairs where high-quality references are often unavailable. In this study, we investigate segment-level QE methods comparing encoder-based models such as MonoTransQuest, CometKiwi, and xCOMET with various decoder-based methods (Tower+, ALOPE, and other instruction-fine-tuned language models). Our work primarily focused on utilizing eight low-resource language pairs, involving both English on the source side and the target side of the translation. Results indicate that while fine-tuned encoder-based models remain strong performers across most low-resource language pairs, decoder-based Large Language Models (LLMs) show clear improvements when adapted through instruction tuning. Importantly, the ALOPE framework further enhances LLM performance beyond standard fine-tuning, demonstrating its effectiveness in narrowing the gap with encoder-based approaches and highlighting its potential as a viable strategy for low-resource QE. In addition, our experiments demonstrates that with adaptation techniques such as LoRA (Low Rank Adapters) and quantization, decoder-based QE models can be trained with competitive GPU memory efficiency, though they generally require substantially more disk space than encoder-based models. Our findings highlight the effectiveness of encoder-based models for low-resource QE and suggest that advances in cross-lingual modeling will be key to improving LLM-based QE in the future. Full article

Objectives: To guide language model (LM) selection by comparing finetuning vs. zero-shot use, generic pretraining vs. domain-adjacent vs. further domain-specific pretraining, and bidirectional language models (BiLMs) such as BERT vs. unidirectional LMs (LLMs) for clinical classification. Materials and Methods: We evaluated BiLMs (RoBERTa, PathologyBERT, Gatortron) and LLM (Mistral nemo instruct 12B) on three British Columbia Cancer Registry (BCCR) pathology classification tasks varying in difficulty/data size. We assessed zero-shot vs. finetuned BiLMs, zero-shot LLM, and further BCCR-specific pretraining using macro-average F1 scores. Results: Finetuned BiLMs outperformed zero-shot BiLMs and zero-shot LLM. The zero-shot LLM outperformed zero-shot BiLMs but was consistently outperformed by finetuned BiLMs. Domain-adjacent BiLMs generally outperformed generic BiLMs after finetuning. Further domain-specific pretraining boosted complex/low-data task performance, with otherwise modest gains. Conclusions: For specialized classification, finetuning BiLMs is crucial, often surpassing zero-shot LLMs. Domain-adjacent pretrained models are recommended. Further domain-specific pretraining provides significant performance boosts, especially for complex/low-data scenarios. BiLMs remain relevant, offering strong performance/resource balance for targeted clinical tasks. Full article

The emergence of large language models (LLMs) has redefined the potential of artificial intelligence in clinical domains. In this context, retrieval-augmented generation (RAG) systems provide a promising approach to enhance traceability, timeliness, and accuracy in tasks such as biomedical question answering (QA). This article presents a narrative and thematic review of the evolution of these technologies in maternal health, structured across five axes: technical foundations of RAG, advancements in biomedical LLMs, conversational agents in healthcare, clinical validation frameworks, and specific applications in obstetric telehealth. Through a systematic search in scientific databases covering the period from 2022 to 2025, 148 relevant studies were identified. Notable developments include architectures such as BiomedRAG and MedGraphRAG, which integrate semantic retrieval with controlled generation, achieving up to 18% improvement in accuracy compared to pure generative models. The review also highlights domain-specific models like PMC-LLaMA and Med-PaLM 2, while addressing persistent challenges in bias mitigation, hallucination reduction, and clinical validation. In the maternal care context, the review outlines applications in prenatal monitoring, the automatic generation of clinically validated QA pairs, and low-resource deployment using techniques such as QLoRA. The article concludes with a proposed research agenda emphasizing federated evaluation, participatory co-design with patients and healthcare professionals, and the ethical design of adaptable systems for diverse clinical settings. Full article

This paper presents a novel visual approach to teaching multiplication to elementary school pupils using stick intersections. Within the South African context, where students consistently demonstrate low mathematics achievement, particularly in foundational arithmetic operations, this research explores an alternative pedagogical strategy that transforms abstract multiplication concepts into visual, concrete, countable representations. Building on theories of embodied cognition and visual mathematics, this study implemented and evaluated the stick intersection method with 45 Grade 4 students in Polokwane, Limpopo Province. Using a mixed-methods approach combining quantitative assessments with qualitative observations, the results revealed statistically significant improvements in multiplication performance across all complexity levels, with particularly substantial gains among previously low-performing students (61.3% improvement, d = 1.87). Qualitative findings demonstrated enhanced student engagement, deeper conceptual understanding of place value, and overwhelmingly positive learner perceptions of the method. The visual approach proved especially valuable in the multilingual South African classroom context, where it transcended language barriers by providing direct visual access to mathematical concepts. High retention rates (94.9%) one-month post-intervention suggest the method facilitated lasting conceptual understanding rather than temporary procedural knowledge. This research contributes to mathematics education by demonstrating how visually oriented, culturally responsive pedagogical approaches can address persistent challenges in developing mathematics proficiency, particularly in resource-constrained educational environments. Full article

This paper presents a rule-based negation and litotes detection system for Modern Standard Arabic. Unlike purely statistical approaches, the proposed pipeline leverages linguistic structures, lexical resources, and dependency parsing to identify negated expressions, exception clauses, and instances of litotic inversion, where rhetorical negation conveys an implicit positive meaning. The system was applied to a large-scale subset of the Arabic OSCAR corpus, filtered by sentence length and syntactic structure. The results show the successful detection of 5193 negated expressions and 1953 litotic expressions through antonym matching. Additionally, 200 instances involving exception prepositions were identified, reflecting their syntactic specificity and rarity in Arabic. The system is fully interpretable, reproducible, and well-suited to low-resource environments where machine learning approaches may not be viable. Its ability to scale across heterogeneous data while preserving linguistic sensitivity demonstrates the relevance of rule-based systems for morphologically rich and structurally complex languages. This work contributes a practical framework for analyzing negation phenomena and offers insight into rhetorical inversion in Arabic discourse. Although coverage of rarer structures is limited, the pipeline provides a solid foundation for future refinement and domain-specific applications in figurative language processing. Full article

Background/Objectives: Artificial intelligence (AI) has emerged as a transformative force in healthcare, with nutrition and dietetics becoming key areas of application. AI technologies are being employed to enhance dietary assessment, personalize nutrition plans, manage chronic diseases, deliver virtual coaching, and support public health nutrition. This review aims to critically synthesize the current literature on AI applications in nutrition, identify research gaps, and outline directions for future development. Methods: A systematic literature search was conducted across PubMed, Scopus, Web of Science, and Google Scholar for peer-reviewed publications from January 2020 to July 2025. The search included studies involving AI applications in nutrition, dietetics, or public health nutrition. Articles were screened based on predefined inclusion and exclusion criteria. Thematic analysis grouped findings into six categories: dietary assessment, personalized nutrition and chronic disease management, generative AI and conversational agents, global/public health nutrition, sensory science and food innovation, and ethical and professional considerations. Results: AI-driven systems show strong potential for improving dietary tracking accuracy, generating personalized diet recommendations, and supporting disease-specific nutrition management. Chatbots and large language models (LLMs) are increasingly used for education and support. Despite this progress, challenges remain regarding model transparency, ethical use of health data, limited generalizability across diverse populations, and underrepresentation of low-resource settings. Conclusions: AI offers promising solutions to modern nutritional challenges. However, responsible development, ethical oversight, and inclusive validation across populations are essential to ensure equitable and safe integration into clinical and public health practice. Full article

With the rapid development of artificial intelligence and machine learning technologies, automatic speech recognition (ASR) and text-to-speech (TTS) have become key components of the digital transformation of society. The Kazakh language, as a representative of the Turkic language family, remains a low-resource language with limited audio corpora, language models, and high-quality speech synthesis systems. This study provides a comprehensive analysis of existing speech recognition and synthesis models, emphasizing their applicability and adaptation to the Kazakh language. Special attention is given to linguistic and technical barriers, including the agglutinative structure, rich vowel system, and phonemic variability. Both open-source and commercial solutions were evaluated, including Whisper, GPT-4 Transcribe, ElevenLabs, OpenAI TTS, Voiser, KazakhTTS2, and TurkicTTS. Speech recognition systems were assessed using BLEU, WER, TER, chrF, and COMET, while speech synthesis was evaluated with MCD, PESQ, STOI, and DNSMOS, thus covering both lexical–semantic and acoustic–perceptual characteristics. The results demonstrate that, for speech-to-text (STT), the strongest performance was achieved by Soyle on domain-specific data (BLEU 74.93, WER 18.61), while Voiser showed balanced accuracy (WER 40.65–37.11, chrF 80.88–84.51) and GPT-4 Transcribe achieved robust semantic preservation (COMET up to 1.02). In contrast, Whisper performed weakest (WER 77.10, BLEU 13.22), requiring further adaptation for Kazakh. For text-to-speech (TTS), KazakhTTS2 delivered the most natural perceptual quality (DNSMOS 8.79–8.96), while OpenAI TTS achieved the best spectral accuracy (MCD 123.44–117.11, PESQ 1.14). TurkicTTS offered reliable intelligibility (STOI 0.15, PESQ 1.16), and ElevenLabs produced natural but less spectrally accurate speech. Full article

Recent advances in vision-language models such as BLIP-2 have made AI-generated image descriptions increasingly fluent and difficult to distinguish from human-authored texts. This paper investigates whether such differences can still be reliably detected by introducing a novel bilingual dataset of English and Romanian captions. The English subset was derived from the T4SA dataset, while AI-generated captions were produced with BLIP-2 and translated into Romanian using MarianMT; human-written Romanian captions were collected via manual annotation. We analyze the problem from two perspectives: (i) semantic alignment, using CLIP similarity, and (ii) supervised classification with both traditional and transformer-based models. Our results show that BERT achieves over 95% cross-validation accuracy (F1 = 0.95, ROC AUC = 0.99) in distinguishing AI from human texts, while simpler classifiers such as Logistic Regression also reach competitive scores (F1 ≈ 0.88). Beyond classification, semantic and linguistic analyses reveal systematic cross-lingual differences: English captions are significantly longer and more verbose, whereas Romanian texts—often more concise—exhibit higher alignment with visual content. Romanian was chosen as a representative low-resource language, where studying such differences provides insights into multilingual AI detection and challenges in vision-language modeling. These findings emphasize the novelty of our contribution: a publicly available bilingual dataset and the first systematic comparison of human vs. AI-generated captions in both high- and low-resource languages. Full article