Informatics

50 pages, 1686 KB

Open AccessReview

Data Foundations for Medical AI: Provenance, Reliability and Limitations of Russian Clinical NLP Resources

by Arsenii Litvinov, Lev Malishevskii, Evgeny Karpulevich, Iaroslav Bespalov, Yaroslav Nedumov, Sergey Zhdanov, Ivan Oseledets, Evgeniy Shlyakhto and Arutyun Avetisyan

Informatics 2026, 13(3), 45; https://doi.org/10.3390/informatics13030045 - 20 Mar 2026

Viewed by 2411

Abstract

Russian-language resources for medical natural language processing (NLP) are expanding rapidly; however, their fragmentation, uneven curation, and limited clinical reliability hinder the development of safe machine learning systems for prognosis, prevention, and precision medicine. We provide the first systematic survey of Russian medical [...] Read more.

Russian-language resources for medical natural language processing (NLP) are expanding rapidly; however, their fragmentation, uneven curation, and limited clinical reliability hinder the development of safe machine learning systems for prognosis, prevention, and precision medicine. We provide the first systematic survey of Russian medical NLP datasets and analyze their suitability for clinically meaningful tasks as defined by the MedHELM taxonomy. We additionally perform expert clinical validation of three representative public corpora—RuMedPrimeData (real outpatient notes), MedSyn (synthetic clinical notes), and RuMedNLI (translated natural language inference)—assessing clinical plausibility, diagnosis accuracy, and logical consistency. Experts identified substantial reliability issues: across randomly sampled subsets of each corpus, only approximately 20% of RuMedPrimeData records, fewer than 15% of MedSyn records, and approximately 55% of RuMedNLI pairs met essential quality criteria, which can hinder downstream ML systems built on these data. To support robust applications—ranging from medical chatbots and triage assistants to predictive and preventive models—we outline practical requirements for high-quality datasets: coordinated, expert-validated, machine-readable corpora aligned with clinical guidelines and insurance logic, standardized de-identification, and transparent provenance. Strengthening these data foundations will enable the development of reliable, reproducible, and clinically relevant AI systems suitable for real-world healthcare applications. Full article

(This article belongs to the Special Issue From Data to Evidence: Transformative AI for Real-World Data)

► Show Figures

Figure 1

24 pages, 1251 KB

Open AccessArticle

Machine Learning and Generative AI in Administrative Processes in Peru: Administrative Efficiency in the National Public Sector

by Miluska Odely Rodriguez Saavedra, Juliana Mery Bautista Lopez, Wilian Quispe Nina, Antonio Víctor Morales Gonzales, Iván Cuentas Galindo, Luis Miguel Campos Ascuña, Anthony Stefano Saenz Colana, Robinson Bernardino Almanza Cabe, Paola Gabriela Lujan Tito and Sharon Veronika Liendo Teran

Informatics 2026, 13(3), 44; https://doi.org/10.3390/informatics13030044 - 19 Mar 2026

Cited by 2 | Viewed by 2894

Abstract

Public organizations in Peru have committed substantial resources to artificial intelligence over recent years, yet evidence on whether these investments produce measurable returns has remained scarce. This study evaluated the causal impact of AI adoption on administrative efficiency across 20 Peruvian national public [...] Read more.

Public organizations in Peru have committed substantial resources to artificial intelligence over recent years, yet evidence on whether these investments produce measurable returns has remained scarce. This study evaluated the causal impact of AI adoption on administrative efficiency across 20 Peruvian national public organizations, using a quasi-experimental design combining Difference-in-Differences with Propensity Score Matching, complemented by XGBoost version 1.7.6, Random Forest, GPT-4, and SHAP explainability analysis. The sample comprised 428 civil servants across treatment and control organizations. Results showed significant efficiency gains as perceived by civil servants through validated Likert instruments: work absenteeism decreased by 9.4%, processing times by 8.7%, and administrative costs by 18.2%, all at p < 0.001 with Cohen’s d ranging from 0.55 to 0.90. The convergence between DiD and PSM estimates supports a causal reading of these effects. Four of five hypotheses were supported. AI delivered comparable efficiency gains regardless of institutional complexity, so H2 was not confirmed. Digital infrastructure significantly moderated AI effectiveness (H3: r = 0.198, p = 0.004). Higher resistance to change was significantly associated with lower efficiency outcomes (H5: r = −0.256, p < 0.001), reinforcing the role of proactive change management as a positive moderator of AI effectiveness. SHAP analysis revealed that training investment, specialized IT personnel, and resistance management together explained 51% of predictive importance, outweighing structural variables such as budget size or geographic location. These findings provide the first systematic causal evidence on AI efficiency in Peruvian public administration and offer actionable benchmarks for comparable middle-income public sectors. Full article

► Show Figures

Figure 1

26 pages, 843 KB

Open AccessArticle

Artificial Intelligence in Literature Review Synthesis: A Step-by-Step Methodological Approach for Researchers and Academics

by Matolwandile M. Mtotywa, Jeri-Lee J. Mowers, Wavhudi Ndou, Thabang V. Q. Moleko and Matsobane J. Ledwaba

Informatics 2026, 13(3), 43; https://doi.org/10.3390/informatics13030043 - 13 Mar 2026

Cited by 6 | Viewed by 10351

Abstract

The integration of artificial intelligence (AI) in literature reviews aims to transform research by potentially automating processes, enhancing rigour, and improving quality. The study proposes a structured step-by-step approach to integrate AI tools into the literature review synthesis process. The developed methodological approach [...] Read more.

The integration of artificial intelligence (AI) in literature reviews aims to transform research by potentially automating processes, enhancing rigour, and improving quality. The study proposes a structured step-by-step approach to integrate AI tools into the literature review synthesis process. The developed methodological approach has five steps. The first step, planning and readiness, involves scoping, understanding practices, and defining boundaries of AI use. Next is selecting AI tools and aligning their capabilities with the literature needs through a matrix. The third step focuses on using AI to conduct the review, followed by validation and cross-referencing of AI-generated results. The final step is disclosing AI use in line with ethical and reporting standards. The approach is demonstrated through five scenarios: emerging or fragmented literature, large or saturated fields, interdisciplinary domains, methodologically diverse studies, and under-researched topics. This approach is designed to enhance transparency, potentially reduce bias, and support reproducibility by aligning AI functions with research goals. It also addresses ethical considerations and promotes human–AI collaboration. For researchers and academics, it aims to provide a practical roadmap for the responsible adoption of AI in literature reviews, supporting efficiency, ethical tool use, transparency, and the balance between machine assistance and academic judgment. Full article

► Show Figures

Figure 1

19 pages, 3031 KB

Open AccessArticle

Voice, Text, or Embodied AI Avatar? Effects of Generative AI Interface Modalities in VR Museums

by Pakinee Ariya, Perasuk Worragin, Songpon Khanchai, Darin Poollapalin and Phichete Julrode

Informatics 2026, 13(3), 42; https://doi.org/10.3390/informatics13030042 - 11 Mar 2026

Cited by 1 | Viewed by 2654

Abstract

Virtual museums delivered through immersive virtual reality (VR) function as information environments where users access interpretive content while navigating spatially. With the integration of generative artificial intelligence (AI), conversational assistants can dynamically mediate information interaction; however, evidence remains limited regarding how different AI [...] Read more.

Virtual museums delivered through immersive virtual reality (VR) function as information environments where users access interpretive content while navigating spatially. With the integration of generative artificial intelligence (AI), conversational assistants can dynamically mediate information interaction; however, evidence remains limited regarding how different AI interface representations affect user experience. This study compares three generative AI interface modalities in a VR virtual museum: voice only, voice with synchronized text, and voice with an embodied AI avatar. A controlled experiment with 75 participants examined their effects on user engagement, perceived information quality, and subjective cognitive workload while holding informational content constant. The results indicate that the voice-and-text modality produced the highest perceived information quality, whereas the embodied AI avatar modality yielded the highest user engagement. No significant differences were observed in cognitive workload across modalities. These findings suggest that AI interface modalities play complementary roles in VR-based information interaction and provide design guidance for selecting appropriate AI representations in immersive information systems. Full article

(This article belongs to the Special Issue Real-World Applications and Prototyping of Information Systems for Extended Reality (VR, AR, and MR))

► Show Figures

Figure 1

16 pages, 1344 KB

Open AccessReview

Dr. Google vs. Dr. ChatGPT in Online Health Self-Consultation: A Scoping Review of Accuracy, Bias, and Actionability (2023–2025)

by Magdalena Trillo-Domínguez, Juan Ignacio Martin-Neira and María Dolores Olvera-Lobo

Informatics 2026, 13(3), 41; https://doi.org/10.3390/informatics13030041 - 5 Mar 2026

Viewed by 2185

Abstract

The rapid adoption of generative artificial intelligence (AI) systems has transformed health information seeking, raising questions about their role as intermediaries in non-professional health self-consultation. This study compares Google Search and ChatGPT as paradigmatic models of algorithmic mediation of health information, focusing on [...] Read more.

The rapid adoption of generative artificial intelligence (AI) systems has transformed health information seeking, raising questions about their role as intermediaries in non-professional health self-consultation. This study compares Google Search and ChatGPT as paradigmatic models of algorithmic mediation of health information, focusing on accuracy, biases, information quality and potential harms. A scoping review was conducted following the PRISMA-ScR framework. Empirical studies published between 2023 and 2025 were retrieved from PubMed/MEDLINE, Web of Science (WoS) and Scopus. After screening and eligibility assessment, 63 original empirical studies were included. The results indicate that ChatGPT consistently outperforms Google Search in terms of factual accuracy and information quality, achieving moderate to high DISCERN scores (4–5 out of 5) and showing moderate to strong correlations with expert clinical evaluations. Users also tend to value ChatGPT responses positively due to their clarity, coherence and perceived empathy. However, these advantages coexist with significant structural limitations. Hallucinations are reported in an estimated 31–45% of references, source provenance remains opaque, linguistic complexity is high, and actionability is limited, with only around 40% of responses providing clearly actionable guidance. In contrast, Google Search offers greater source traceability and verifiability, but at the cost of fragmented information and higher exposure to commercial content. The review identifies critical research gaps related to behavioural impacts, critical health literacy, equity of access, professional integration and vulnerable contexts. Overall, the findings highlight the need for hybrid human–AI models, professional mediation and critical AI literacy to ensure safe, equitable and trustworthy use of generative AI in public health communication. Full article

► Show Figures

Figure 1

18 pages, 729 KB

Open AccessArticle

Organizational Characteristics Associated with Health Information Systems Adoption in Local Health Departments During the COVID-19 Pandemic

by Nardeen Shafik, Gulzar H. Shah, Timothy C. McCall, Bettye A. Apenteng, Mansoor Abro and William A. Mase

Informatics 2026, 13(3), 40; https://doi.org/10.3390/informatics13030040 - 4 Mar 2026

Viewed by 1255

Abstract

Background: The COVID-19 pandemic revealed persistent gaps in local health department (LHD) health informatics capacity. This study examines organizational characteristics of LHDs associated with the adoption of six health information systems: electronic case reporting (eCR), electronic disease reporting systems (EDRS), electronic health records [...] Read more.

Background: The COVID-19 pandemic revealed persistent gaps in local health department (LHD) health informatics capacity. This study examines organizational characteristics of LHDs associated with the adoption of six health information systems: electronic case reporting (eCR), electronic disease reporting systems (EDRS), electronic health records (EHR), electronic lab reporting (ELR), health information exchange (HIE), and immunization registries (IR). Methods: We used a mixed-methods design, including multinomial or binary logistic regression analyses of quantitative data from the 2022 NACCHO National Profile of Local Health Departments (n = 441) and thematic analysis of semi-structured interviews with five LHD staff members. Results: About half (49.9%) of LHDs had implemented eCR, while higher proportions had implemented EDRS (78.0%), EHR (62.4%), ELR (57.2%), HIE (92.6%), and IR (92.6%). Workforce size was associated with the implementation of eCR, EHR, and IR. The number of vacant staff positions was associated with a lower odds of IR implementation; compared with medium-sized LHDs, both small and large LHDs had higher odds of IR implementation. Shared-governance LHDs had higher odds of adopting ELR and HIE than state-governed LHDs. Qualitative themes highlighted challenges, including staff burnout, high turnover, pay inequities, role ambiguity, political pressures, rapid changes in informatics, and interoperability problems. Conclusions: Findings underscore the need to improve LHD workforce capacity and governance structures to support a resilient public health informatics infrastructure. Full article

► Show Figures

Figure 1

19 pages, 1100 KB

Open AccessArticle

Integrating Agentic Artificial Intelligence to Automate International Classification of Diseases, Tenth Revision, Medical Coding

by Kitti Akkhawatthanakun, Lalita Narupiyakul, Konlakorn Wongpatikaseree, Narit Hnoohom, Chakkrit Termritthikun and Paisarn Muneesawang

Informatics 2026, 13(3), 39; https://doi.org/10.3390/informatics13030039 - 4 Mar 2026

Viewed by 3082

Abstract

Automating ICD-10 coding from discharge summaries remains demanding because coders analyze clinical narratives while justifying decisions. This study compares three automation patterns: PLM-ICD as a standalone deep learning system emitting 15 codes per case, LLM-only generation with full autonomy, and a hybrid approach [...] Read more.

Automating ICD-10 coding from discharge summaries remains demanding because coders analyze clinical narratives while justifying decisions. This study compares three automation patterns: PLM-ICD as a standalone deep learning system emitting 15 codes per case, LLM-only generation with full autonomy, and a hybrid approach where PLM-ICD drafts candidates for an agentic LLM audit to accept or reject. All strategies were evaluated on 19,801 MIMIC-IV summaries using four LLMs spanning compact (Qwen2.5-3B-Instruct, Llama-3.2-3B-Instruct, Phi-4-mini-instruct) to large-scale (Sonnet-4.5). Precision guided evaluation because coders still supply any missing diagnoses. PLM-ICD alone reached 55.8% precision while always surfacing 15 suggestions. LLM-only generation lagged severely (1.5–34.6% precision) and produced inconsistent output sizes. The agentic audit delivered the best trade-off: compact LLMs reviewed the 15 candidates, discarded weak evidence, and returned 2–8 high-confidence codes. Llama-3.2-3B-Instruct, for example, improved from 1.5% as a generator to 55.1% as a verifier while trimming false positives by 73%. These results show that positioning LLMs as quality controllers, rather than primary generators, yields reliable support for clinical coding teams, while formal recall/F1 reporting remains future work for fully autonomous implementations. Full article

(This article belongs to the Special Issue Health Data Management in the Age of AI)

► Show Figures

Figure 1

3 pages, 144 KB

Open AccessEditorial

Learning to Live with Gen-AI

by Antony Bryant

Informatics 2026, 13(3), 38; https://doi.org/10.3390/informatics13030038 - 4 Mar 2026

Viewed by 682

Abstract

In 2023, in the wake of the launch of ChatGPT, based on GPT-3, we invited contributions on the Topic AI chatbots: threat or opportunity [...] Full article

27 pages, 1280 KB

Open AccessArticle

Enhancing Causal Text Detection Using Uncertainty-Weighted Machine Learning Ensembles

by Sivachandra K B, Neethu Mohan, Mithun Kumar Kar, Sikha O K and Sachin Kumar S

Informatics 2026, 13(3), 37; https://doi.org/10.3390/informatics13030037 - 2 Mar 2026

Viewed by 1748

Abstract

Causal inference in text data has been a demanding objective in the field of natural language processing, mainly due to the intrinsic ambiguity and context sensitivity inherent in data, inducing uncertainty. Diminishing this uncertainty is essential in identifying reliable causal connections and advancing [...] Read more.

Causal inference in text data has been a demanding objective in the field of natural language processing, mainly due to the intrinsic ambiguity and context sensitivity inherent in data, inducing uncertainty. Diminishing this uncertainty is essential in identifying reliable causal connections and advancing predictive consistency. In this research, we introduce an uncertainty-aware ensemble architecture that combines multiple text embedding schemes with both linear and nonlinear classifiers to boost causal text detection. Both sparse and neural-level embeddings were employed, and then combined it with an ensemble weighting approach based on two uncertainty estimation techniques, namely entropy-based and KL divergence-based. Unlike conventional ensemble methods with uniform or fixed voting strategies, our approach assigns weights inversely proportional to classifier uncertainty, ensuring that confident models exert greater influence on the final decisions. Our results show that TF-IDF, through its effective word frequency weighting scheme, consistently outperforms other embedding techniques, achieving better performance across both linear and nonlinear classifiers on both datasets (News Corpus and CausalLM–Adjective group). The experimental results show that our uncertainty-aware ensemble approach enhances both calibration and confidence predictions. Entropy-based weighting improves confidence in the case of linear classifiers with accuracy, F1-score, entropy and prediction confidence values of 94.3%, 94.0%, 0.382 and 0.774, respectively, while in the case of nonlinear classifiers the KL divergence-based weighting acquires a better performance with an accuracy of 97.6%, F1-score of 97.2%, KL Mean value of around 0.055 and LogLoss of 0.221. Full article

(This article belongs to the Section Machine Learning)

► Show Figures

Figure 1

24 pages, 3918 KB

Open AccessArticle

Mapping 3D Digital Heritage at Scale: A ChatGPT-Assisted Analysis of Sketchfab’s “Cultural Heritage & History” Models

by Massimiliano Pepe, Andrei Crisan, Emmanuel Maravelakis, Donato Palumbo, Ahmed Kamal Hamed Dewedar and Przemysław Klapa

Informatics 2026, 13(3), 36; https://doi.org/10.3390/informatics13030036 - 2 Mar 2026

Cited by 1 | Viewed by 2115

Abstract

This paper evaluates the platform-mediated importance and impact of 3D cultural heritage models stored on Sketchfab by analyzing user engagement and retention metrics (views, likes, and comments), and provides a comparative assessment across other major 3D platforms. Our primary goal is to understand [...] Read more.

This paper evaluates the platform-mediated importance and impact of 3D cultural heritage models stored on Sketchfab by analyzing user engagement and retention metrics (views, likes, and comments), and provides a comparative assessment across other major 3D platforms. Our primary goal is to understand how cultural heritage content performs in terms of reach, engagement, and reuse conditions, and how platform design and taxonomies shape what becomes visible and measurable. We map Sketchfab’s Cultural Heritage & History ecosystem through a reproducible, API-driven workflow built on public metadata for over 1.37 million models (views, likes, comments, tags, and licences). The results depict a domain in rapid expansion between 2018 and 2025, while also revealing a strongly unequal attention economy: most models receive limited interaction, whereas a small minority concentrates visibility and engagement. The category Cultural Heritage & History shows high endorsement relative to reach, consistent with “high-value” engagement once content is discovered. Methodologically, large-scale harvesting required automation to manage cursor pagination, intermittent failures, and rate limits (e.g., HTTP 429). In this context, ChatGPT provided essential support by assisting the design and refinement of the extraction and counting algorithm, replacing what would otherwise have required extensive manual counting and verification at a scale that could plausibly take months. Full article

(This article belongs to the Section Social Informatics and Digital Humanities)

► Show Figures

Figure 1

22 pages, 1673 KB

Open AccessArticle

Ontology-Based Digital Preservation Framework for Phum Riang Silk Heritage

by A-Phorn Molee, Thana Charuphanthuset, Wittawat Kunnu and Supaporn Chairungsee

Informatics 2026, 13(3), 35; https://doi.org/10.3390/informatics13030035 - 27 Feb 2026

Viewed by 1566

Abstract

Traditional textile crafts face significant challenges in preserving and transferring knowledge due to the aging of expert artisans and declining community engagement. The Phum Riang silk-weaving tradition in Suratthani Province is a critical example of indigenous knowledge systems that require systematic documentation and [...] Read more.

Traditional textile crafts face significant challenges in preserving and transferring knowledge due to the aging of expert artisans and declining community engagement. The Phum Riang silk-weaving tradition in Suratthani Province is a critical example of indigenous knowledge systems that require systematic documentation and digital conservation strategies. This research aims to develop a comprehensive ontological framework to support the capture, organization, and preservation of traditional knowledge related to Phum Riang silk production processes, establishing practical methodologies applicable to broader cultural heritage craft digitization and knowledge management systems. The research methodology employs ontology engineering principles, using the Web Ontology Language to create structured knowledge representation systems. Data collection was conducted through ethnographic fieldwork, in-depth interviews with expert craftspeople, and systematic documentation covering production processes, materials, tools, and cultural practices. The developed ontology encompasses five primary knowledge domains: production processes, raw materials, traditional tools, geographical context, and cultural significance. The framework comprises 23 distinct classes organized in hierarchical structures, 15 object properties, and 12 data properties, complemented by business rules ensuring authenticity and quality control mechanisms. This framework has significant implications for cultural heritage digitization, indigenous intellectual property protection, systematic knowledge transfer across generations, cultural authenticity preservation, and traditional craft community economic sustainability. Full article

► Show Figures

Figure 1

33 pages, 3658 KB

Open AccessArticle

Personalized Canine Diet Generation Using Machine Learning and Constraint Optimization

by Aliya Kalykulova, Kuanysh Bakirov, Aruzhan Shoman, Kadyrzhan Makangali and Gulzhan Tokysheva

Informatics 2026, 13(3), 34; https://doi.org/10.3390/informatics13030034 - 25 Feb 2026

Viewed by 2043

Abstract

The growing demand for customized pet diets highlights the shortcomings of commercial dog foods designed for all breeds, especially when it comes to addressing breed-specific diseases, metabolic disorders, and health risks. This research presents the development and evaluation of a hybrid system for [...] Read more.

The growing demand for customized pet diets highlights the shortcomings of commercial dog foods designed for all breeds, especially when it comes to addressing breed-specific diseases, metabolic disorders, and health risks. This research presents the development and evaluation of a hybrid system for formulating wet canine food recipes. The system combines data on ingredients, veterinary feeds, and breed-related diseases; the architecture includes a recommendation module for ingredient selection and a linear programming block for recipe optimization, considering veterinary nutrient restrictions. The evaluation of the system included automatic classification of foods by specialization, visual analysis of recipe clustering, and comparison of formulas obtained by different models. The average precision of label recovery was 85.4% for TF-IDF and 88.2% for the E5 model. A comparison of ingredient extraction methods showed that machine learning produces more stable recipes, while the statistical approach provides greater variability. The developed system demonstrates potential for automating recipe creation, filling in missing data, and developing veterinary decision support platforms aimed at personalized diet selection based on the physiological needs of animals. Full article

(This article belongs to the Topic Decision Science Applications and Models (DSAM))

► Show Figures

Figure 1

Journal Menu

Journal Browser

Informatics, Volume 13, Issue 3 (March 2026) – 12 articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI