O artigo apresenta a contribuição de José Roberto Ferreira para as discussões, nos âmbitos nacional e internacional, a respeito da formação de profissionais da saúde e de suas contribuições para a construção de uma proposta crítica para a cooperação Sul-Sul em saúde. Argumenta-se que Ferreira defendia ser a educação em saúde, desenvolvida com base em realidades particulares de países menos desenvolvidos, um caminho para a emancipação. Inicialmente apresentamos a formação e os primeiros trabalhos de Ferreira no campo da educação (1959-1969); as suas experiências internacionais e os contatos com figuras importantes da medicina social latino-americana (1969-1996); e, finalmente, o seu trabalho na Fiocruz no âmbito da política externa para desenvolvimento da cooperação estruturante em saúde (1996-2019).
Amaç: Bu çalışma, Türkiye’de GD ile ilgili yapılan lisansüstü ebelik tezlerini incelemek amacıyla yapılmıştır.Yöntem: Bu araştırma, Yükseköğretim Kurulu Başkanlığı Ulusal Tez Merkezi Veri Tabanı kullanılarak yapılmıştır. Tarama ‘Gestasyonel Diyabet’, ‘Gestasyonel Diyabetes Mellitus’, ‘Ebelilk’ anahtar kelimeleri kullanılarak Aralık 2023- Ocak 2024 tarihleri arasında yapılmıştır. Verilerin değerlendirilmesinde, tezler tablolar halinde sunulmuş; yazar, yayın yılı, tez türü, amacı, çalışma tipi, örneklem özellikleri, veri toplama araçları ve sonuçlar özetlenmiştir. Verilerin analizleri bilgisayar ortamında yapılmıştır. Bulgular: Son on yıl içinde yapılmış toplam 10 tez incelenmiş olup, bu tezlerin çoğunluğunun tanımlayıcı tipte olduğu belirlenmiştir. Taranan yedi yüksek lisans tezinden beşi tanımlayıcı, biri metodolojik, biri deneysel; üç doktora tezinden ikisi deneysel, biri metadolojiktir. Tezlerin ikisinin GD riskini ölçme ve değerlendirmeye yarayan ölçek geliştirme çalışması olduğu, ikisinde eğitim ve danışmanlık yöntemlerinin etkinliğinin değerlendirildiği, ikisinin simülasyon ve öğrenci eğitimini içeren çalışma olduğu, ikisinde GD’li kadınların sağlık durumlarının incelendiği, birinde fiziksel aktivite ve yaşam kalitesinin incelendiği ve birinin GD tanısı ve ilişkili faktörlerle ilgili olduğu görülmüştür. Örneklem gruplarına bakıldığında, dört tezde GD’li gebeler, üç tezde genel gebe grubu, bir tezde ebelik, bir tezde ebelik ve hemşirelik bölümü öğrencileri, bir tezde ise hem gebeler hem de GD’li gebeler yer almaktadır.Sonuç: Dünyada ve ülkemizde prevelansı giderek artan GD’nin gebelerin yaşam kalitesini olumsuz etkilediği belirlenmiştir. Öğrencilere verilen simülasyon eğitiminin, gebelere verilen eğitim ve danışmanlığın etkili olduğu görülmüştür. Ebelerin GD ile ilgili daha fazla nitel ve deneysel çalışma yapması önerilmektedir.
History of medicine. Medical expeditions, Miscellaneous systems and treatments
Pedro M. Gordaliza, Nataliia Molchanova, Jaume Banus
et al.
Deep learning models for medical image segmentation suffer significant performance drops due to distribution shifts, but the causal mechanisms behind these drops remain poorly understood. We extend causal attribution frameworks to high-dimensional segmentation tasks, quantifying how acquisition protocols and annotation variability independently contribute to performance degradation. We model the data-generating process through a causal graph and employ Shapley values to fairly attribute performance changes to individual mechanisms. Our framework addresses unique challenges in medical imaging: high-dimensional outputs, limited samples, and complex mechanism interactions. Validation on multiple sclerosis (MS) lesion segmentation across 4 centers and 7 annotators reveals context-dependent failure modes: annotation protocol shifts dominate when crossing annotators (7.4% $\pm$ 8.9% DSC attribution), while acquisition shifts dominate when crossing imaging centers (6.5% $\pm$ 9.1%). This mechanism-specific quantification enables practitioners to prioritize targeted interventions based on deployment context.
This paper explores optimal data selection strategies for Reinforcement Learning with Verified Rewards (RLVR) training in the medical domain. While RLVR has shown exceptional potential for enhancing reasoning capabilities in large language models, most prior implementations have focused on mathematics and logical puzzles, with limited exploration of domain-specific applications like medicine. We investigate four distinct data sampling strategies from MedQA-USMLE: random sampling (baseline), and filtering using Phi-4, Gemma-3-27b-it, and Gemma-3-12b-it models. Using Gemma-3-12b-it as our base model and implementing Group Relative Policy Optimization (GRPO), we evaluate performance across multiple benchmarks including MMLU, GSM8K, MMLU-Pro, and CMMLU. Our findings demonstrate that models trained on filtered data generally outperform those trained on randomly selected samples. Notably, training on self-filtered samples (using Gemma-3-12b-it for filtering) achieved superior performance in medical domains but showed reduced robustness across different benchmarks, while filtering with larger models from the same series yielded better overall robustness. These results provide valuable insights into effective data organization strategies for RLVR in specialized domains and highlight the importance of thoughtful data selection in achieving optimal performance. You can access our repository (https://github.com/Qsingle/open-medical-r1) to get the codes.
Ariel Lubonja, Pedro R. A. S. Bassi, Wenxuan Li
et al.
Open challenges have become the de facto standard for comparative ranking of medical AI methods. Despite their importance, medical AI leaderboards exhibit three persistent limitations: (1) score gaps are rarely tested for statistical significance, so rank stability is unknown; (2) single averaged metrics are applied to every organ, hiding clinically important boundary errors; (3) performance across intersecting demographics is seldom reported, masking fairness and equity gaps. We introduce RankInsight, an open-source toolkit that seeks to address these limitations. RankInsight (1) computes pair-wise significance maps that show the nnU-Net family outperforms Vision-Language and MONAI submissions with high statistical certainty; (2) recomputes leaderboards with organ-appropriate metrics, reversing the order of the top four models when Dice is replaced by NSD for tubular structures; and (3) audits intersectional fairness, revealing that more than half of the MONAI-based entries have the largest gender-race discrepancy on our proprietary Johns Hopkins Hospital dataset. The RankInsight toolkit is publicly released and can be directly applied to past, ongoing, and future challenges. It enables organizers and participants to publish rankings that are statistically sound, clinically meaningful, and demographically fair.
Federated Learning (FL) is a collaborative learning method that enables decentralized model training while preserving data privacy. Despite its promise in medical imaging, recent FL methods are often sensitive to local factors such as optimizers and learning rates, limiting their robustness in practical deployments. In this work, we revisit vanilla FL to clarify the impact of edge device configurations, benchmarking recent FL methods on colorectal pathology and blood cell classification task. We numerically show that the choice of local optimizer and learning rate has a greater effect on performance than the specific FL method. Moreover, we find that increasing local training epochs can either enhance or impair convergence, depending on the FL method. These findings indicate that appropriate edge-specific configuration is more crucial than algorithmic complexity for achieving effective FL.
Recent advancements in Large Language Models (LLMs) have catalyzed a paradigm shift from static prediction systems to agentic AI agents capable of reasoning, interacting with tools, and adapting to complex tasks. While LLM-based agentic systems have shown promise across many domains, their application to medical imaging remains in its infancy. In this work, we introduce AURA, the first visual linguistic explainability agent designed specifically for comprehensive analysis, explanation, and evaluation of medical images. By enabling dynamic interactions, contextual explanations, and hypothesis testing, AURA represents a significant advancement toward more transparent, adaptable, and clinically aligned AI systems. We highlight the promise of agentic AI in transforming medical image analysis from static predictions to interactive decision support. Leveraging Qwen-32B, an LLM-based architecture, AURA integrates a modular toolbox comprising: (i) a segmentation suite with phase grounding, pathology segmentation, and anatomy segmentation to localize clinically meaningful regions; (ii) a counterfactual image-generation module that supports reasoning through image-level explanations; and (iii) a set of evaluation tools including pixel-wise difference-map analysis, classification, and advanced state-of-the-art components to assess diagnostic relevance and visual interpretability.
Data scarcity is a major limiting factor for applying modern machine learning techniques to clinical tasks. Although sufficient data exists for some well-studied medical tasks, there remains a long tail of clinically relevant tasks with poor data availability. Recently, numerous foundation models have demonstrated high suitability for few-shot learning (FSL) and zero-shot learning (ZSL), potentially making them more accessible to practitioners. However, it remains unclear which foundation model performs best on FSL medical image analysis tasks and what the optimal methods are for learning from limited data. We conducted a comprehensive benchmark study of ZSL and FSL using 16 pretrained foundation models on 19 diverse medical imaging datasets. Our results indicate that BiomedCLIP, a model pretrained exclusively on medical data, performs best on average for very small training set sizes, while very large CLIP models pretrained on LAION-2B perform best with slightly more training samples. However, simply fine-tuning a ResNet-18 pretrained on ImageNet performs similarly with more than five training examples per class. Our findings also highlight the need for further research on foundation models specifically tailored for medical applications and the collection of more datasets to train these models.
Gurucharan Marthi Krishna Kumar, Aman Chadha, Janine Mendola
et al.
Large Language Models (LLMs), known for their versatility in textual data, are increasingly being explored for their potential to enhance medical image segmentation, a crucial task for accurate diagnostic imaging. This study explores enhancing Vision Transformers (ViTs) for medical image segmentation by integrating pre-trained LLM transformer blocks. Our approach, which incorporates a frozen LLM transformer block into the encoder of a ViT-based model, leads to substantial improvements in segmentation performance across various medical imaging modalities. We propose a Hybrid Attention Mechanism that combines global and local feature learning with a Multi-Scale Fusion Block for aggregating features across different scales. The enhanced model shows significant performance gains, including an average Dice score increase from 0.74 to 0.79 and improvements in accuracy, precision, and the Jaccard Index. These results demonstrate the effectiveness of LLM-based transformers in refining medical image segmentation, highlighting their potential to significantly boost model accuracy and robustness. The source code and our implementation are available at: https://github.com/AS-Lab/Marthi-et-al-2025-MedVisionLlama-Pre-Trained-LLM-Layers-to-Enhance-Medical-Image-Segmentation
Medical imaging cohorts are often confounded by factors such as acquisition devices, hospital sites, patient backgrounds, and many more. As a result, deep learning models tend to learn spurious correlations instead of causally related features, limiting their generalizability to new and unseen data. This problem can be addressed by minimizing dependence measures between intermediate representations of task-related and non-task-related variables. These measures include mutual information, distance correlation, and the performance of adversarial classifiers. Here, we benchmark such dependence measures for the task of preventing shortcut learning. We study a simplified setting using Morpho-MNIST and a medical imaging task with CheXpert chest radiographs. Our results provide insights into how to mitigate confounding factors in medical imaging.
Artificial Intelligence (AI) has the potential to revolutionize diagnosis and segmentation in medical imaging. However, development and clinical implementation face multiple challenges including limited data availability, lack of generalizability, and the necessity to incorporate multi-modal data effectively. A foundation model, which is a large-scale pre-trained AI model, offers a versatile base that can be adapted to a variety of specific tasks and contexts. Here, we present VIsualization and Segmentation Masked AutoEncoder (VIS-MAE), novel model weights specifically designed for medical imaging. Specifically, VIS-MAE is trained on a dataset of 2.5 million unlabeled images from various modalities (CT, MR, PET,X-rays, and ultrasound), using self-supervised learning techniques. It is then adapted to classification and segmentation tasks using explicit labels. VIS-MAE has high label efficiency, outperforming several benchmark models in both in-domain and out-of-domain applications. In addition, VIS-MAE has improved label efficiency as it can achieve similar performance to other models with a reduced amount of labeled training data (50% or 80%) compared to other pre-trained weights. VIS-MAE represents a significant advancement in medical imaging AI, offering a generalizable and robust solution for improving segmentation and classification tasks while reducing the data annotation workload. The source code of this work is available at https://github.com/lzl199704/VIS-MAE.
The Yalta Conference of 1945 brought together three of the most influential leaders of the 20th century: Franklin D. Roosevelt, Joseph Stalin and Winston Churchill. Surprisingly, all three leaders would go on to suffer strokes after the conference. This manuscript examines the health status of these leaders during and after the Yalta Conference, the factors that contributed to their strokes (including the role of hypertension), and other modifiable risk factors present in each one of them, and the impact of their declining health on their countries and the world. Roosevelt's demise, prior to the conclusion of the war, triggered a leadership transition during a critical moment in history, while Churchill and Stalin's passing shaped the early Cold War era. A veil of secrecy shrouded the health conditions of these pivotal leaders. “The Big Three” made considerable efforts to hide their health conditions from both the press and the public at large.Understanding the health of political leaders is crucial as it can affect their decision-making abilities and the course of history. The fates of Roosevelt, Stalin and Churchill serve as important reminders of the potential consequences of poor health in the highest echelons of political power.
History of medicine. Medical expeditions, Medical philosophy. Medical ethics
El objetivo del presente trabajo es ofrecer una reconstrucción e interpretación de tres exposiciones de arte psicopatológico organizadas por el neuropsiquiatra Gonzalo Rodríguez Lafora en los meses previos a la Guerra Civil Española. Inicialmente, el proyecto fue ideado en el marco de la VII Reunión de la Asociación Española de Neuropsiquiatras y la VI Asamblea de la Liga Española de Higiene Mental, celebradas simultáneamente en el Instituto Santiago Ramón y Cajal de Madrid entre el 2 y el 4 de diciembre de 1935. Aunque, acabó siendo itinerante al trasladarse dos semanas después al Ateneo de Madrid (17 de diciembre), y, una vez más, a la Casa del Médico de Barcelona en mayo de 1936. El estudio de las tres exposiciones brinda a la investigación actual varias lecturas sobre el caso español con referencia al estudio del arte psicopatológico a principios del siglo XX, las cuales se desarrollan a lo largo del artículo.
History of scholarship and learning. The humanities, History of medicine. Medical expeditions
Purpose. The article considers the peculiarities of the images of the representatives of peoples of Siberia on the ethnographic maps of the First and Second Kamchatka expeditions, taking place in the first half of the 18th century.Results. Maps are an important historical source, moreover not only the cartographic information is studied, but also other materials located on the map. Three versions of the final map of the First Kamchatka Expedition, as well as the Ethnographic Map of Siberia of the Second Kamchatka Expedition are a valuable source for research. They contain images of representatives of different peoples and scenes from their lives. A comparative analysis of the drawings of Yakut, Tungus (Evenks), Koryak, Kuril, Chukchi, Kamchadal (Itelmen) was carried out, the features of images of clothing and objects were revealed, and ethnographic analogies were attracted. So, gradually, on different maps a fur coat on Yakut loses its fur trim, the bow becomes a crooked stick with a rope. On later maps, frames appear around the images, the number of representatives of the Siberian peoples increases.The maps are made by different mapmakers. The first of them was created in St. Petersburg, sketches made during the work of the First Kamchatka expedition were copied on it. Local mapmakers did not quite understand what they were depicting and, therefore, already at this stage there is a loss of part of the ethnographic meaning. The rest of the maps were already copied from the first one, so there is a further loss of ethnographic specificity, simplification of the pictures.The sequence of implementation of the maps was determined by increasing the number of images, the appearance of frames around them, and their gradual simplification. The earliest version of the final map of the First Kamchatka Expedition shows the summer camp of the Kamchadals under the cartouche. This is the most complete image, on the other sources only individual elements of this sketch were drawn. The ethnographic map of Siberia of the Second Kamchatka Expedition is the most complete in both cartographic and ethnographic contexts. But the greatest losses of the ethnographic specificity of the Siberian peoples are observed on it.Conclusion. As a result of the conducted research, the sequence of compiling versions of the final map of the First Kamchatka expedition (ethnographic version) was determined. The variant of the drawings of representatives of peoples when compiling the ethnographic map of Siberia of the Second Kamchatka expedition is revealed. A gradual partial loss of ethnographic information occurred when copying maps by different mapmakers who did not quite understand what they were depicting.
Caner Ozer, Arda Guler, Aysel Turkvatan Cansever
et al.
To maintain a standard in a medical imaging study, images should have necessary image quality for potential diagnostic use. Although CNN-based approaches are used to assess the image quality, their performance can still be improved in terms of accuracy. In this work, we approach this problem by using Swin Transformer, which improves the poor-quality image classification performance that causes the degradation in medical image quality. We test our approach on Foreign Object Classification problem on Chest X-Rays (Object-CXR) and Left Ventricular Outflow Tract Classification problem on Cardiac MRI with a four-chamber view (LVOT). While we obtain a classification accuracy of 87.1% and 95.48% on the Object-CXR and LVOT datasets, our experimental results suggest that the use of Swin Transformer improves the Object-CXR classification performance while obtaining a comparable performance for the LVOT dataset. To the best of our knowledge, our study is the first vision transformer application for medical image quality assessment.
We present a meta-learning framework for interactive medical image registration. Our proposed framework comprises three components: a learning-based medical image registration algorithm, a form of user interaction that refines registration at inference, and a meta-learning protocol that learns a rapidly adaptable network initialization. This paper describes a specific algorithm that implements the registration, interaction and meta-learning protocol for our exemplar clinical application: registration of magnetic resonance (MR) imaging to interactively acquired, sparsely-sampled transrectal ultrasound (TRUS) images. Our approach obtains comparable registration error (4.26 mm) to the best-performing non-interactive learning-based 3D-to-3D method (3.97 mm) while requiring only a fraction of the data, and occurring in real-time during acquisition. Applying sparsely sampled data to non-interactive methods yields higher registration errors (6.26 mm), demonstrating the effectiveness of interactive MR-TRUS registration, which may be applied intraoperatively given the real-time nature of the adaptation process.
Neda Yavari, Fariba Asghari, Zahra Shahvari
et al.
It appears that up until now, no comprehensive tool has been developed to assess medical students’ attitudes toward the different dimensions of professionalism. The present study aimed to develop a comprehensive quantitative tool to evaluate medical students’ attitudes toward professionalism. This study consisted of two phases: The first phase was item generation and questionnaire design based on literature review and a qualitative survey. The qualitative data were extracted from 49 semi-structured individual interviews and one focus group discussion. In the second phase, the questionnaire was developed and its face, content, and structure validity and reliability were evaluated. To measure the construct validity of the questionnaire, a cross-sectional study was conducted on 354 medical students at different academic levels at Isfahan University of Medical Sciences. The final questionnaire was loaded on five factors. The factors accounted for 43.5% of the total variance. Moreover, Cronbach's alpha was 0.84 for the total scale, and the interclass correlation coefficient was 0.77 for the test-retest reliability. The 17-item questionnaire measuring medical students’ professional attitude had acceptable validity and reliability and can be adopted in other studies on physicians’ and medical students’ professional attitudes.
History of medicine. Medical expeditions, Medical philosophy. Medical ethics
The article formulates the
theoretical foundations of the study of dramatic text, distinguishes
between the concepts of “dramatic text” (affects the reader) and
“dramatic work” (verbalizes the theatrical action for the viewer),
because these linguistic realities are in the relationship of inclusion. A
dramatic text is a unity organized according to certain laws, which has
a clear structure: a certain number of lines, designed with the help of
the author's remark, forms a dialogical unity; the combination of
dialogic units forms a scene; the set of scenes constitute an act; several
acts create a complete work.
Obligatory factors of expressiveness of the dramatic text that
influence the development of dialogic parts of the characters are
pragmatic components of speech which are considered as certain rules
of successful communication. In the communicative-pragmatic
paradigm dramatic text is defined as a specific type of artistic text that
has its own structural and speech features due to a combination of
informational, pragmatic, stylistic and cognitive aspects, where the
pragmatic aspect is found in speech acts, syntactic organization of cues.
The scientific novelty of the research is DT of Bukovinian’s
writers of the late nineteenth – early twentieth century that have not yet
been the subject of analysis.
The relevance of scientific research requires a holistic analysis
of the dramatic text and difference between the concepts of “dramatic
text” and “dramatic composition”, the study of mandatory factors of
expressiveness of dramatic text, influencing the development of
dialogic parts.
The following methods and techniques of linguistic analysis are
used in the article: system-functional analysis, method of discussion
analysis, contextual-interpretive method.
Conclusions. Dramatic text is a complex phenomenon with its
own peculiarities of functioning. It is possible to penetrate into the
structure of a dramatic text, to reveal the meaning of the author's
intentions due to the volume-pragmatic division of the text. The
communicative-pragmatic organization of DT influences the processes
of active aesthetic influence of a work of art on the consciousness of
the addressee. We see the prospect of the represented research in the
further deepening of knowledge about the categorical features of the
text in the communicative-pragmatic aspect.
History of medicine. Medical expeditions, Social Sciences
Xiaoyuan Guo, Judy Wawira Gichoya, Saptarshi Purkayastha
et al.
Traditional anomaly detection methods focus on detecting inter-class variations while medical image novelty identification is inherently an intra-class detection problem. For example, a machine learning model trained with normal chest X-ray and common lung abnormalities, is expected to discover and flag idiopathic pulmonary fibrosis which a rare lung disease and unseen by the model during training. The nuances from intra-class variations and lack of relevant training data in medical image analysis pose great challenges for existing anomaly detection methods. To tackle the challenges, we propose a hybrid model - Transformation-based Embedding learning for Novelty Detection (TEND) which without any out-of-distribution training data, performs novelty identification by combining both autoencoder-based and classifier-based method. With a pre-trained autoencoder as image feature extractor, TEND learns to discriminate the feature embeddings of in-distribution data from the transformed counterparts as fake out-of-distribution inputs. To enhance the separation, a distance objective is optimized to enforce a margin between the two classes. Extensive experimental results on both natural image datasets and medical image datasets are presented and our method out-performs state-of-the-art approaches.