共找到 20 条结果
Historical weather journals from across Canada, spanning 1768-1884, have been transcribed from handwritten records into machine readable formats. The NORTHERN (Nineteenth-century Overseas Records Transcribed for Historical Environmental Reconstruction in the North) project transcribed nearly 2 million weather observations from 46 locations. The original documents are in archives outside Canada. The two principal archives investigated for historical Canadian weather are the United States' National Administration and Records Archives (NARA) and the United Kingdom's Meteorological Office (UKMO) Library and Archives. Some observations were also located in the United States' National Centers for Environmental Information (NCEI) "Forts" dataset. Observers recorded from three to twenty weather variables, in most cases two or three times daily. Validation procedures are carried out with export files produced in both the original format and in modern units. Observations of pressure, temperature, precipitation, snow depth, cloud cover, cloud type, wind direction and wind force are transcribed along with detailed descriptions of events including fires, floods, ice formation and break up, storms and other weather phenomena. The value of these data lies in their detailed observations of sub-daily weather together with descriptive observation of disruptive or extreme weather events. These data will be used to expand knowledge of Canada's climate variability and extreme values for three centuries and to improve global reanalysis data products.
Retractions serve as a critical self-corrective mechanism within the scientific enterprise, yet their implementation remains inconsistent and unevenly consequential across disciplines and institutions. This paper examines the causes, patterns, processes, and impacts of retractions. This paper accentuates that the retractions arise from a complex interplay of individual misconduct, systemic publication pressures, inadequate peer review, and cultural conditions that discourage honest error correction. Retraction notices are found to be frequently incomplete, poorly disseminated, and inconsistently linked across bibliographic databases, allowing flawed findings to persist in citation networks long after formal withdrawal. The paper further demonstrates that retraction impacts fall disproportionately on early-career researchers, particularly those in developing countries. These findings carry relevance for journals such as Naunyn-Schmiedeberg's Archives of Pharmacology, which publishes translational research with direct clinical and regulatory implications. The paper also is aligned with the goals of United Nations Sustainable Development Goal 3 (good health and well-being) and SDG 16 (peace, justice, and strong institutions), as strengthening scientific integrity infrastructure supports both evidence-based healthcare and accountable research governance. Concrete reforms are proposed, including standardized retraction notices and open-access policies, to ensure that retractions fulfill their promise as instruments of scientific self-correction.
Follicular lymphoma (FL) is one of the most common subtypes of B-cell lymphoma. While often following an indolent course, a subset of patients develop histologic transformation to an aggressive B-cell or classic Hodgkin lymphoma (CHL). Transformation of other subtypes of low-grade B-cell lymphoma, such as chronic lymphocytic leukemia, are often associated with Epstein-Barr virus (EBV). However, EBV is only rarely associated with histologic transformation of FL. We encountered two rare occurrences of EBV-associated lymphoproliferative disease (EBV-LPD) arising in patients with FL and sought to further characterize the circumstances under which they arise. Pathology archives were searched for relevant cases following institutional review board approval. Seventeen patients with diagnoses of EBV-LPD and synchronous (11, same site in 8/11), metachronous (4) or suspected (2) diagnoses of FL were identified. The patients were predominantly elderly (median age 72, range 50-80 years) including 9 men and 8 women. At least four patients had known immune deficiency/dysfunction. The FL were predominantly low grade by 4th Revised Edition World Health Organization morphologic criteria (79%; 11/14 grade 1-2; 3/14 grade 3A) with detectable BCL2-rearrangements (90%, 9/10). All FL were negative for EBV where evaluable. The FL and associated EBV-LPD were often clonally related, including all evaluable cases with large B-cell morphology. Overall, EBV-associated histologic transformation is a rare complication of FL and is often synchronous and clonally related to the FL. Most patients were elderly, including a subset with additional known causes of immune dysfunction and other immunodeficiency-associated complications.
While individual Jewish plastic surgeons have been recognized in the literature for their groundbreaking achievements, a comprehensive review of their collective contributions to plastic surgery is lacking. The goal of this review is to document and highlight the major advancements made by Jewish physicians whose work significantly impacted plastic and reconstructive surgery from its inception as a discipline to the present day. A comprehensive literature search was conducted to identify key developments in the history of plastic surgery, which were then narrowed to those made by Jewish practitioners. The Jewish identity of these practitioners, along with the significance of their work, was verified through historical records, newspaper archives, and peer-reviewed publications. Contributions were then organized chronologically to construct a cohesive historical narrative. Beginning in the 1800s and accelerating through the 20th century, Jewish plastic surgeons have been instrumental in shaping the field. Their innovations include advancements in the treatment of facial deformities, the development of the lumpectomy with adjunctive radiation therapy as a less invasive alternative to radical mastectomy, and the founding of key institutions such as the American Society of Plastic Surgeons and the International College of Surgeons. Additional breakthroughs include the use of albumin in burn management, foundational work in angiogenesis physiology, and numerous innovations in cosmetic surgery. The history of plastic and reconstructive surgery is a rich tapestry of discovery and evolution. Fully understanding its development is not possible without acknowledging the enduring and transformative contributions of Jewish surgeons and researchers.
To characterize the clinicopathological spectrum of ocular masses diagnosed over 12 years at a tertiary referral center in Northern China and identify demographic and regional patterns relevant to clinical practice. We retrospectively reviewed 1,345 ocular masses diagnosed at Jinan Second People's Hospital between February 2014 and July 2025. Cases were identified from pathology archives and included patients with surgical excision or biopsy and definitive histopathology. Data on age, sex, anatomic location, and mass type were analyzed. Classification followed AFIP atlas standards and ICD-O coding. Of the 1,345 masses, 1,146 (85.2%) were non-malignant and 199 (14.8%) malignant. Lesions were located in the eyelid (35.2%), ocular surface (35.7%), intraocular region (1.8%), and orbit (27.3%). For eyelid masses, nevi (20.8%) and cysts (11.2%) were the most common non-malignant lesions, and basal cell carcinoma (55.6%) and sebaceous gland carcinoma (24.2%) dominated malignancies. For ocular surface masses, the top non-malignant lesions included cysts (20.0%) and inflammatory lesions (18.1%), and malignant melanoma (34.6%) and squamous cell carcinoma (26.9%) were the leading malignancies. Most intraocular masses (20/24) were malignant, and the majority was melanoma (16/20). For orbital masses, hemangiomas (21.1%) were the most frequent benign lesions, and lymphoma (51.9%) was the leading malignancy. The majority of malignancies occurred in patients aged > 60 years (132/199, 66.3%). Compared to patients aged 18-60 years, those aged < 18 years had significantly lower odds of malignancy (OR = 0.21, 95% CI 0.08-0.59), whereas those > 60 years had significantly higher odds of malignancy (OR = 3.29, 95% CI 2.37-4.56). Most ocular masses were non-malignant. The overall risk of malignancy increased significantly after age 60, and this trend was primarily driven by extraocular lesions.
The automatic recognition of artistic styles is an important part of the management, analysis, and access to the huge digital cultural heritage archives. Nevertheless, the aesthetic complexity and minor distinctions between works of art become serious obstacles to this process. To overcome this, this study introduces a multi-stage, novel technique that, in comparison to current techniques, intelligently derives and combines local and global features to obtain a more comprehensive style representation. In the suggested approach, initially, the image is split into visual parts with the help of the K-means algorithm, and the local characteristics of each part are obtained with the help of specific CNNs. Simultaneously, a more detailed CNN obtains global features of the whole picture. The two sets of features are then intelligently fused with an attention mechanism. In the third step, an Autoencoder (AE) is employed to downsize the dimensions of the extracted features and eliminate noise and redundant data because of the large dimensionality of the extracted features. Lastly, a SoftMax classifier is used to perform classification. The proposed method was assessed using the Metropolitan Museum of Art (MET) Public Archives dataset in this study. Our approach on the MET dataset had an 89.07% accuracy and F-measure of 0.8867, which proves it to be superior.
Pigments form a chemotaxonomic system widely used to determine phytoplankton composition and reconstruct water body histories by using sedimentary pigments as biological archives. This chapter presents protocols for pigment extraction from water and sediment samples. It also describes a method to separate, identify, and quantify major phytoplankton pigments in a single run. The method is based on high performance liquid chromatography with diode array detection (HPLC-DAD).
Underwater sediment density currents triggered by marine volcanic eruptions threaten island communities and infrastructure, while their deposits provide archives of past eruptions. Despite their significance, scarce real-time density current observations and concurrent deposit samples limit our understanding of their behaviour and relationship to varying volcanic mechanisms. Using data acquired following the explosive, VEI 6, shallow-submarine eruption of Hunga Volcano in 2022, we show that syn-eruptive delivery of pyroclastic material into the ocean via low-column collapses and fountaining triggered the multidirectional dispersal of highly-concentrated underwater density currents. Rapid supply of > 6.5 km3 of dense pyroclastic material onto the steep volcanic flanks over minutes-to-hours generated currents that maintain high density and velocity 10-100s of kilometres from the volcano. We outline diagnostic criteria to differentiate deposits of shallow-submarine generated underwater currents from other volcanic processes - enabling better reconstruction of the records of volcanic activity in marine sediments and enhancing hazard assessments in submerged volcanic settings worldwide.
Oral squamous cell carcinoma (OSCC) accounts for about 90% of oral malignancies, with a higher occurrence among males than females. Use of molecular biomarkers acts as a reliable predictor of tumour aggressiveness. CDK1 modulates the centrosome cycle and mitotic onset, promoting G2-M transition and regulation of G1-S transition associated with multiple interphase cyclins. Inhibiting CDK1 and its substrates could increase the efficacy of OSCC treatment by overcoming immune or apoptotic resistance. To compare the immunohistochemical (IHC) expression of CDK1 in normal mucosa (NOM), oral epithelial dysplasia (OED), OSCC, to assess the prognostic value of CDK1 in OSCC and to compare and correlate those three groups. This study includes 55 cases, which consist of 20 cases of OSCC, 20 cases of OED and 15 cases of NOM, which were retrieved from the archives of our department and were subjected to IHC staining. The Mann-Whitney test was utilised to find out the association among the categorical variables. The Wilcoxon test was used to compare the continuous parameters between three groups. On IHC analysis, CDK1 expression was found to be 80% positive in the OED group, 85% in the OSCC group, and negative in the NOM group. Negative CDK1 expression was seen in NOM. CDK1 can be used as a predictor in the prognosis of OSCC and determine the degree of malignancy at an early stage of OSCC. Targeting CDK1 interacting proteins and their signalling pathways produces less toxicity in treating oral carcinogenesis.
Pneumonia remains a leading cause of in-hospital mortality worldwide. Current prognostic tools such as the IDSA/ATS severity score have meaningful limitations, particularly in capturing dynamic disease progression or integrating heterogeneous biological signals. Artificial intelligence (AI) offers the opportunity to derive complementary prognostic information from routinely collected electronic health record data. This exploratory, proof-of-concept retrospective study enrolled adults (≥18 years) admitted with a primary diagnosis of acute pneumonia at Hospital Alma Máter de Antioquia (Medellín, Colombia) between January 1 and June 30, 2024. After applying pre-defined exclusion criteria, 121 patients (19 non-survivors, 15.7%) comprised the final analytic cohort. Three independent AI modules were applied: (i) a ResNet-18 deep learning model quantified lung consolidation from chest radiographs (CXRs) using Class Activation Mapping; (ii) a Spanish regular-expression natural language processing (NLP) pipeline extracted modified IDSA/ATS severity scores from clinical notes; and (iii) NeuroKit2-based quantitative heart rate variability (HRV) analysis processed electrocardiogram (ECG) signals digitised from PDF archives. Bivariate associations with all-cause in-hospital mortality were examined using logistic regression. Several features exhibited statistically significant associations with mortality under conventional thresholds: AI-quantified total lung compromise ratio (OR 8.32, 95% CI 1.23-56.29), NLP-derived IDSA/ATS severity score (OR 1.78, 95% CI 1.09-2.88), and a coherent ECG/HRV profile characterised by higher heart rate (120.1 vs. 84.4 bpm, p = 0.023), reduced RMSSD (4.1 vs. 23.5 ms, p = 0.041), reduced Poincaré SD1 (3.0 vs. 17.6 ms, p = 0.041), and T-wave amplitude reductions surviving FDR correction in multiple leads. Given the small sample size and low event count (n = 19; events-per-variable ≈ 4), all associations are preliminary and hypothesis-generating only. These proof-of-concept findings suggest that integrated multimodal AI biomarkers automatically derived from low-resource clinical data can capture a cardiopulmonary stress profile associated with pneumonia mortality, and support the design of larger prospective validation studies.
Artificial Intelligence (AI) is impacting several aspects of modern life with its ability to enhance decision-making, automate complex tasks, and generate human-like content. It is now an indispensable tool in both everyday life and academic inquiry. In particular, the rapid evolution of AI technologies, especially machine learning, deep learning, and natural language processing (NLP), has given rise to large language models (LLMs), which have transformed how we analyze, interpret, and generate text-based, structured data and unstructured data. Among these, Generative AI (GenAI) has become increasingly popular due to its capacity to create content ranging from text and code to protein sequences and molecular structures, all based on patterns found in large training datasets. GenAI tools can assist with literature reviews, writing support, data processing, hypothesis generation, and code or visualization tasks, although outputs require critical oversight to ensure accuracy and relevance. More advanced GenAI applications include the generation of synthetic data and even the design of biological molecules and materials. Within this broader context, the fields of immunology, vaccinology, and infectious diseases research are witnessing a wave of innovation driven by AI. In this review, we explore how these recent advances in GenAI, especially those based on LLMs, are being applied to immunological research, antibody design, vaccine development, infectious diseases research and pandemic preparedness. This review is structured as a scoping review, aiming to map the rapidly evolving applications of GenAI and LLMs in immunology, vaccine development, infectious disease research, and adjacent biomedical fields. Relevant studies were identified through searching PubMed, Google Scholar and preprint archives and included if they introduced, demonstrated, or benchmarked AI-based approaches with clear relevance to immunology and infectious disease, while older preprints without subsequent peer-reviewed publication were excluded. We aim to provide a comprehensive overview of current contributions, emerging tools and models, and future perspectives of GenAI in transforming how we understand and manipulate immune responses and infectious diseases. Therefore, the reported capabilities should be interpreted as indicative of potential rather than definitive performance.
Feline ulcerative dermatitis is an uncommon disease of domestic and wild cats associated with felid alphaherpesvirus 1 (FeAHV1) infection. Here we report dermal inflammatory vascular changes with or without thrombosis in six of eight cats with feline ulcerative dermatitis. Submissions were searched retrospectively from the Athens Veterinary Diagnostic Laboratory biopsy archives between 2010 and 2025. FeAHV1 infection was confirmed by immunohistochemistry in all cases. The affected cats were five castrated males and three spayed females. Domestic Shorthair cats were affected more often (seven cases), with one affected Domestic Mediumhair cat. The age of patients ranged from 6 months to 10 years (median age = 8.5 years). Clinical signs consisted of ulcerative and crusting lesions that affected sites including the nose (six cases), right flank (one case) and right elbow, right shoulder and tarsal area (one case). Histological changes were typical of FeAHV1 infection in all cases and included epidermal necrosis that extended to the hair follicle, ulceration and serocellular crusts, with intranuclear eosinophilic viral inclusions within epithelial cells of the epidermis, hair follicle and sebocytes. Intranuclear viral inclusions within macrophages were observed in two cases. The dermis was infiltrated by eosinophils admixed with neutrophils, macrophages, lymphocytes and plasma cells. Transmural vascular changes consisting of mixed inflammatory cell infiltrates, fibrinoid change and thrombosis were present in small- to medium-sized arterioles and venules in the dermis in six cases. In two cases, these vascular changes were associated with perivascular dermal necrosis and haemorrhage. Although not previously reported in detail, inflammatory -vascular changes occur in cases of FeAHV1 dermatitis and may be secondary to extensive epidermal necrosis and ulceration.
PurposeThis descriptive study explores how a trusted information source within the Black community-the Black News Media-promoted COVID-19 vaccination during a time when this population experienced high rates of vaccine hesitancy and low COVID-19 vaccine uptake.ApproachTo explore the media's role in promoting COVID-19 vaccination, a multi-level sampling method was used to identify newspaper articles for review.SettingNewspaper articles discussing COVID-19 vaccination published by newspapers primarily serving Black communities between March 2021 and October 2022.ParticipantsNewspaper publishers were identified through the Black Media Initiative -a resource that catalogs media outlets in the U.S that primarily serve Black communities. Publishers were selected based on their publishing frequency and online accessibility.MethodOne hundred fifty newspaper articles were randomly sampled from a sample of 906 articles. Using an inductive thematic analysis, this study examined the key themes used to encourage vaccination among Africans Americans.ResultsThe study found five key themes the Black News Media used to promote vaccination: Protecting others, particularly the vulnerable, the unvaccinated, deadly decisions, responsibility, and return to normalcy.LimitationsSampling constraints, including uneven outlet representation across regions and the exclusion of difficult-to-search archives, may have limited the diversity of perspectives and themes identified. Moreover, the review may have excluded some viable articles.ConclusionSeveral of the themes used by media aligned with frames recognized as effective in promoting vaccination, making them well-suited to address high levels of vaccine hesitancy and skepticism toward the COVID-19 vaccine within Black communities.
Schwannomas, also known as neurilemmomas, are benign nerve sheath tumours derived from Schwann cells. Schwannomas are truly encapsulated, slow-growing tumours with a very wide anatomical distribution but with a predilection for the extremities and head and neck regions. The study aimed to analyse the histomorphological spectrum and diagnostic utility of immunohistochemistry in schwannomas. The frequency distribution and demographic characteristics of these tumours were also determined. This was a 10-year hospital-based retrospective study of all cases of schwannomas that strictly met the inclusion criteria. Patients' records and tissue blocks were retrieved from archives. Formalin-fixed paraffin-embedded tissue blocks were subjected to histochemical and immunohistochemical analysis following standard protocols, and the slides produced were reviewed and interpreted. Collated data were analysed and depicted using statistical charts and tables. Twenty-six cases were seen which occurred more in females with a male-to-female ratio of 1:1.2. The most frequent decade of occurrence was the fourth decade of life, and the upper limb was the most frequent anatomical site of occurrence with few cases seen in unusual sites. The most frequent histological variant seen was the conventional type, constituting 14 cases (53.8%). SOX10 showed 96% immunopositivity, calretinin showed 96% immunopositivity, and S100 showed 81% immunopositivity with varying staining intensities. CD34 antibody showed 73% immunopositivity which was mainly seen in the Antoni B areas. Ki67 antibody was not expressed in any of the cases subjected to it. Histomorphology and immunohistochemistry are valuable tools in the definitive diagnosis of schwannomas.
Artefacts are tissue alterations on a prepared microscopic slide due to external factors. Artefacts cause morphological and cytological alterations, which lead to misdiagnosis and result during biopsy procedures, particularly fixation artefacts due to inappropriate use/concentration of fixative agents, followed by artefacts during tissue processing, embedding, staining, sectioning, and poor handling of tissues. The present study aims to assess various histological artefacts encountered in our laboratory over four years. In this cross-sectional study, oral and maxillofacial histological slides prepared by haematoxylin and eosin staining were collected from the archives of our department from 2021 to 2024. A total of 3793 histopathological slides were evaluated. Two oral and maxillofacial pathologists evaluated the histopathological sections for the presence or absence of various artefacts. The results were tabulated in a Microsoft Excel sheet and were analysed descriptively. The most common defect was folds in the sections, accounting for 34% of all samples (1,317 occurrences), with peaks in the years 2022 (363 cases) and 2023 (396 cases). Crush defects followed, noted in 14% of the samples (554 cases), with a notable rise in 2024 (198 cases). The study found that a large percentage of artefacts in histopathological sections were caused by technical or human faults. Folding artefact, which can cause tissues to become deformed and unclear, was reported to occur most frequently.
Tumor budding (TB) and poorly differentiated clusters (PDCs) are histopathological parameters that have been associated with poor prognosis in various malignancies, particularly colorectal carcinoma. Although numerous studies have demonstrated an association between the microcystic, elongated, and fragmented (MELF) pattern in endometrial carcinomas and lymphovascular invasion as well as lymph node metastasis, the literature regarding TB and PDCs in this context remains limited.This study aimed to investigate the potential associations of the MELF pattern, TB, and PDCs with overall survival, progression-free survival, and clinicopathological parameters in cases of endometrioid endometrial carcinoma (EEC). A total of 190 cases diagnosed with EEC through hysterectomy specimens between 2010 and 2023, with complete hospital records, were selected from the archives of the Department of Pathology at Manisa Celal Bayar University Faculty of Medicine. The presence of TB and PDCs was significantly associated with high histological grade (p < 0.001), deep myometrial invasion (p < 0.001), lymphovascular invasion (p < 0.001), lymph node metastasis (p < 0.001), cervical stromal involvement (p < 0.001), serosal involvement (p < 0.001), advanced stage (p < 0.001), and larger tumor size (p = 0.042 and p = 0.027, respectively). The presence of TB and PDCs was found to be associated with reduced overall survival (p = 0.001 and p < 0.001, respectively) and reduced progression-free survival (p = 0.043 and p = 0.004, respectively). The MELF pattern was not significantly associated with overall survival (p = 0.772). These findings suggest that the presence of TB and PDCs may be valuable in stratifying prognostic risk in EEC and support the inclusion of these parameters in routine pathology reporting.
Lassa fever, a neglected zoonotic hemorrhagic disease caused by Lassa virus (LASV) and endemic in West Africa, remains a significant public health concern associated with rodent exposure. Senegal lies at the western fringe of the LASV area, but only one 1988 serosurvey reported low antibody prevalence. Given recent ecological shifts, including the expansion of invasive Rattus rattus and Mus musculus, we reassessed LASV exposure in rodents from Senegal. We retrospectively analyzed 618 archived rodent sera collected in 2012-2013 from domestic and peri-domestic environments in central and eastern Senegal. Samples were screened for LASV-specific IgG by ELISA, and IgM was assessed descriptively in IgG-positive or equivocal specimens, with cautious interpretation because rodent validation for IgM remains limited. Spatial mapping and ecological analysis identified seropositivity clusters and potential environmental correlates. Eleven rodents were IgG-seroreactive (1.8%; 95% CI: 0.9-3.2%), and no IgM reactivity was detected, although IgM results were interpreted cautiously because rodent validation remains limited. All seroreactive animals belonged to the commensal species Rattus rattus (9/180; 5%) and Mus musculus (2/174; 1.1%) and were clustered in five villages along major transport corridors. At the locality level, overall positivity ranged from 3.3% to 28.6%, with the highest values observed in Didé Gassama (26.3%) and Kounkane (28.6%), indicating a focal rather than diffuse pattern of LASV exposure. This study provides the first update in over three decades on LASV exposure in Senegalese rodents. IgG seroreactivity confined to invasive commensal species suggests localized exposure patterns that warrant further investigation, while the weak, non-significant association between rodent diversity and seropositivity does not support a clear diversity effect. Spatial clustering of seropositive rodents along major transport routes points to low-level but persistent circulation in settings favoring human-rodent contact. These findings provide a retrospective baseline of LASV seroreactivity in Senegalese rodents and highlight the need to integrate rodent surveillance into One Health frameworks to strengthen early warning and regional preparedness. Not applicable.
Theorizing the failures of computer vision algorithms requires shifting from detecting and fixing biases towards understanding how algorithms are shaped by social, historical, and political real-world precursors. To better understand the socially embedded and historically rooted representational harms of these algorithms, we analyze how AI image captioning depicts archival images of living ethnological exibitions  (so-called 'human zoos'), mass stereotype-producing public exhibitions of colonized people common in Europe and the US from the 1870s to the 1930s, which were meant to symbolize the imagined superiority of Western societies and justify their colonial violence. We collected and analyzed more than 3800 captions from 100 archival images using MidJourney--a modern, state-of-the-art generative AI platform. Combining quantification with close reading of the captions, we found evidence of a 'colonial gaze,' an epistemological viewpoint from the perspective of colonizers characterized by significant representational harms representing five main themes: essentialism (41.6% of captions), cultural erasure (54.5%), dehumanization (11.1%), othering (28.4%), and infantilization (26.8%), with striking parallels between AI-generated captions and the original framings of human zoos informed by a broader colonial epistemology. Based on this analysis, we propose to conceptualize the colonial gaze in generative AI as an automated process of object identification and relational interpretation that draws on historical visual tropes and hierarchical logics rooted in colonial epistemologies. Trigger warning: This article contains extremely racialized text and images produced by both colonizers and the machines. The online version contains supplementary material available at 10.1007/s00146-025-02685-0.
Medication errors remain a leading source of preventable harm in hospitalized patients, contributing to adverse drug events (ADEs), prolonged hospital stay, and avoidable healthcare costs. Although clinical decision support systems (CDSS) integrated with electronic health records (EHRs) have demonstrated potential to reduce prescribing errors, rigorous multicenter randomized evidence from low- and middle-income countries (LMICs), as classified by the World Bank income criteria, remains scarce. To evaluate the effect of an AI-enhanced, EHR-integrated CDSS on medication error rates and ADE incidence in hospitalized adults, with additional assessment of alert performance, clinician adoption, and cost-effectiveness. We conducted a prospective, parallel-arm, randomized controlled trial (RCT) across four tertiary-care teaching hospitals in Tashkent, Uzbekistan (January 2022 - August 2023). Adult inpatients were randomized 1:1 to CDSS-assisted care (MedGuard-UZ v1.3) or standard care. Primary outcomes were medication error rate per 1,000 patient-days and ADE incidence per 100 admissions. Analyses followed the intention-to-treat (ITT) principle. Among 2,384 randomized patients, the CDSS group demonstrated a 49.2% reduction in medication error rates (3.47 vs. 6.83 per 1,000 patient-days; p < 0.001) and a 47.2% reduction in ADE incidence (4.7 vs. 8.9 per 100 admissions; p < 0.001). Overall alert acceptance was 73.6%, with allergy/contraindication alerts achieving 95.5%. Clinician adoption rose from 42.1% to 88.7% daily active users over 12 months. Length of stay was significantly shorter in the CDSS group (7.1 vs. 7.8 days; p = 0.002), as were 30-day readmissions (11.2% vs. 13.7%; p = 0.041). Estimated return on investment was 489% over 12 months. AI-enhanced CDSS integration was associated with substantially improved medication safety and selected hospital outcomes in a multicenter LMIC tertiary-care setting. The MedGuard-UZ AI project materials are publicly available at https://github.com/Shakarbayev/MedGuard-UZ. For peer-review reproducibility, the repository state corresponding to this revision has been archived under the tagged release v1.0.0-ijmedi-rct (tag commit: f796c3d). MedGuard-UZ v1.3 denotes the internal trial system version, whereas v1.0.0-ijmedi-rct (tag commit: f796c3d) identifies the public revision archive supporting evaluation of the AI project component in resource-constrained health systems.
Benign breast disease (BBD) is common and confers heterogeneous increases in breast cancer risk; however, risk prediction relies mainly on histopathology and clinical factors. Sclerosing adenosis (SA) is a proliferative BBD lesion associated with an approximately two-fold increase in risk, yet most women with SA never develop breast cancer. We hypothesize that the immune-stromal microenvironment of SA and its surrounding lobular field relates to subsequent invasive breast cancer. In a nested case-control study within a BBD cohort, we profiled 24 sclerosing adenosis (SA) biopsies (9 developing invasive breast cancer within 15 years, cases; 15 cancer-free at ≥ 15 years, controls). We integrated whole-tissue NanoString gene-expression profiling with multiplex immunofluorescence (MxIF) imaging of SA lesions and surrounding morphologically normal lobules. We measured immune and stromal biomarkers in SA lesions and adjacent lobules, with image analysis masked to case-control status, integrated these data with whole-tissue gene expression, and summarized both microenvironment patterns and the proximity of immune cells to proliferating epithelium. SA biopsies from women who later developed cancer showed a low-immune, high-stromal gene-expression program, whereas controls were enriched for immune signatures. Stromal densities of CD8⁺, CD68⁺ and RUNX3⁺ cells in both lobular stroma and SA lesions mirrored this axis and were markedly lower in cases than controls. Unsupervised clustering identified immune-cold and immune-hot lobule types and four SA lesion field archetypes; immune-hot lobules and immune-hot/epithelium-proliferative lesion fields were enriched in controls. Spatial analyses further showed that immune-hot lobules have stromal immune cells positioned closer to proliferating epithelium and enriched CD27-CD8 microclusters, whereas SA lesions from cases exhibit greater immune-to-Ki67 distances, fewer boundary-proximal CD8⁺ sentinels, and depletion of CD27-RUNX3 and RUNX3-CD8 microclusters. These findings support an association of an immune-cold SA lesion embedded within an immune-cold lobular field phenotype with subsequent invasive breast cancer risk in women with SA, and suggest that spatially organized, RUNX3-rich immune microenvironments may contribute to epithelial surveillance. Validation in larger cohorts will be needed to confirm generalizability and clarify lesion-specific versus field-wide contributions.