Hasil untuk "Medical philosophy. Medical ethics"

Ping Chen, Zicheng Huang, Xiangming Wang et al.

We propose VL-DUN, a principled framework for joint All-in-One Medical Image Restoration and Segmentation (AiOMIRS) that bridges the gap between low-level signal recovery and high-level semantic understanding. While standard pipelines treat these tasks in isolation, our core insight is that they are fundamentally synergistic: restoration provides clean anatomical structures to improve segmentation, while semantic priors regularize the restoration process. VL-DUN resolves the sub-optimality of sequential processing through two primary innovations. (1) We formulate AiOMIRS as a unified optimization problem, deriving an interpretable joint unfolding mechanism where restoration and segmentation are mathematically coupled for mutual refinement. (2) We introduce a frequency-aware Mamba mechanism to capture long-range dependencies for global segmentation while preserving the high-frequency textures necessary for restoration. This allows for efficient global context modeling with linear complexity, effectively mitigating the spectral bias of standard architectures. As a pioneering work in the AiOMIRS task, VL-DUN establishes a new state-of-the-art across multi-modal benchmarks, improving PSNR by 0.92 dB and the Dice coefficient by 9.76\%. Our results demonstrate that joint collaborative learning offers a superior, more robust solution for complex clinical workflows compared to isolated task processing. The codes are provided in https://github.com/cipi666/VLDUN.

Gabriele Lozupone, Alessandro Bria, Francesco Fontanella et al.

This study presents Latent Diffusion Autoencoder (LDAE), a novel encoder-decoder diffusion-based framework for efficient and meaningful unsupervised learning in medical imaging, focusing on Alzheimer disease (AD) using brain MR from the ADNI database as a case study. Unlike conventional diffusion autoencoders operating in image space, LDAE applies the diffusion process in a compressed latent representation, improving computational efficiency and making 3D medical imaging representation learning tractable. To validate the proposed approach, we explore two key hypotheses: (i) LDAE effectively captures meaningful semantic representations on 3D brain MR associated with AD and ageing, and (ii) LDAE achieves high-quality image generation and reconstruction while being computationally efficient. Experimental results support both hypotheses: (i) linear-probe evaluations demonstrate promising diagnostic performance for AD (ROC-AUC: 90%, ACC: 84%) and age prediction (MAE: 4.1 years, RMSE: 5.2 years); (ii) the learned semantic representations enable attribute manipulation, yielding anatomically plausible modifications; (iii) semantic interpolation experiments show strong reconstruction of missing scans, with SSIM of 0.969 (MSE: 0.0019) for a 6-month gap. Even for longer gaps (24 months), the model maintains robust performance (SSIM > 0.93, MSE < 0.004), indicating an ability to capture temporal progression trends; (iv) compared to conventional diffusion autoencoders, LDAE significantly increases inference throughput (20x faster) while also enhancing reconstruction quality. These findings position LDAE as a promising framework for scalable medical imaging applications, with the potential to serve as a foundation model for medical image analysis. Code available at https://github.com/GabrieleLozupone/LDAE

Siyi Xun, Yue Sun, Jingkun Chen et al.

Rapid advances in medical imaging technology underscore the critical need for precise and automated image quality assessment (IQA) to ensure diagnostic accuracy. Existing medical IQA methods, however, struggle to generalize across diverse modalities and clinical scenarios. In response, we introduce MedIQA, the first comprehensive foundation model for medical IQA, designed to handle variability in image dimensions, modalities, anatomical regions, and types. We developed a large-scale multi-modality dataset with plentiful manually annotated quality scores to support this. Our model integrates a salient slice assessment module to focus on diagnostically relevant regions feature retrieval and employs an automatic prompt strategy that aligns upstream physical parameter pre-training with downstream expert annotation fine-tuning. Extensive experiments demonstrate that MedIQA significantly outperforms baselines in multiple downstream tasks, establishing a scalable framework for medical IQA and advancing diagnostic workflows and clinical decision-making.

Johannes Kiechle, Stefan M. Fischer, Daniel M. Lang et al.

The sharp rise in medical tomography examinations has created a demand for automated systems that can reliably extract informative features for downstream tasks such as tumor characterization. Although 3D volumes contain richer information than individual slices, effective 3D classification remains difficult: volumetric data encode complex spatial dependencies, and the scarcity of large-scale 3D datasets has constrained progress toward 3D foundation models. As a result, many recent approaches rely on 2D vision foundation models trained on natural images, repurposing them as feature extractors for medical scans with surprisingly strong performance. Despite their practical success, current methods that apply 2D foundation models to 3D scans via slice-based decomposition remain fundamentally limited. Standard slicing along axial, sagittal, and coronal planes often fails to capture the true spatial extent of a structure when its orientation does not align with these canonical views. More critically, most approaches aggregate slice features independently, ignoring the underlying 3D geometry and losing spatial coherence across slices. To overcome these limitations, we propose TomoGraphView, a novel framework that integrates omnidirectional volume slicing with spherical graph-based feature aggregation. Instead of restricting the model to axial, sagittal, or coronal planes, our method samples both canonical and non-canonical cross-sections generated from uniformly distributed points on a sphere enclosing the volume. We publicly share our accessible code base at http://github.com/compai-lab/2025-MedIA-kiechle and provide a user-friendly library for omnidirectional volume slicing at https://pypi.org/project/OmniSlicer.

Chang Hong, Minghao Wu, Qingying Xiao et al.

The integration of large language models into healthcare necessitates a rigorous evaluation of their ethical reasoning, an area current benchmarks often overlook. We introduce PrinciplismQA, a comprehensive benchmark with 3,648 questions designed to systematically assess LLMs' alignment with core medical ethics. Grounded in Principlism, our benchmark features a high-quality dataset. This includes multiple-choice questions curated from authoritative textbooks and open-ended questions sourced from authoritative medical ethics case study literature, all validated by medical experts. Our experiments reveal a significant gap between models' ethical knowledge and their practical application, especially in dynamically applying ethical principles to real-world scenarios. Most LLMs struggle with dilemmas concerning Beneficence, often over-emphasizing other principles. Frontier closed-source models, driven by strong general capabilities, currently lead the benchmark. Notably, medical domain fine-tuning can enhance models' overall ethical competence, but further progress requires better alignment with medical ethical knowledge. PrinciplismQA offers a scalable framework to diagnose these specific ethical weaknesses, paving the way for more balanced and responsible medical AI.

Ifrat Ikhtear Uddin, Longwei Wang, KC Santosh

Medical image analysis often faces significant challenges due to limited expert-annotated data, hindering both model generalization and clinical adoption. We propose an expert-guided explainable few-shot learning framework that integrates radiologist-provided regions of interest (ROIs) into model training to simultaneously enhance classification performance and interpretability. Leveraging Grad-CAM for spatial attention supervision, we introduce an explanation loss based on Dice similarity to align model attention with diagnostically relevant regions during training. This explanation loss is jointly optimized with a standard prototypical network objective, encouraging the model to focus on clinically meaningful features even under limited data conditions. We evaluate our framework on two distinct datasets: BraTS (MRI) and VinDr-CXR (Chest X-ray), achieving significant accuracy improvements from 77.09% to 83.61% on BraTS and from 54.33% to 73.29% on VinDr-CXR compared to non-guided models. Grad-CAM visualizations further confirm that expert-guided training consistently aligns attention with diagnostic regions, improving both predictive reliability and clinical trustworthiness. Our findings demonstrate the effectiveness of incorporating expert-guided attention supervision to bridge the gap between performance and interpretability in few-shot medical image diagnosis.

Sushant Gautam, Michael A. Riegler, Pål Halvorsen

We investigate fine-tuning Vision-Language Models (VLMs) for multi-task medical image understanding, focusing on detection, localization, and counting of findings in medical images. Our objective is to evaluate whether instruction-tuned VLMs can simultaneously improve these tasks, with the goal of enhancing diagnostic accuracy and efficiency. Using MedMultiPoints, a multimodal dataset with annotations from endoscopy (polyps and instruments) and microscopy (sperm cells), we reformulate each task into instruction-based prompts suitable for vision-language reasoning. We fine-tune Qwen2.5-VL-7B-Instruct using Low-Rank Adaptation (LoRA) across multiple task combinations. Results show that multi-task training improves robustness and accuracy. For example, it reduces the Count Mean Absolute Error (MAE) and increases Matching Accuracy in the Counting + Pointing task. However, trade-offs emerge, such as more zero-case point predictions, indicating reduced reliability in edge cases despite overall performance gains. Our study highlights the potential of adapting general-purpose VLMs to specialized medical tasks via prompt-driven fine-tuning. This approach mirrors clinical workflows, where radiologists simultaneously localize, count, and describe findings - demonstrating how VLMs can learn composite diagnostic reasoning patterns. The model produces interpretable, structured outputs, offering a promising step toward explainable and versatile medical AI. Code, model weights, and scripts will be released for reproducibility at https://github.com/simula/PointDetectCount.

Abdul Hanan bin Abdul Salam, Nur Syamim Syahirah Mat Hussin

Ismail Raji al-Faruqi’s vision of the Islamization of Knowledge (IoK) has profoundly influenced Malaysia’s education system, shaping both policies and pedagogical approaches. His engagement with Malaysian scholars and political figures, including Syed Naquib al-Attas, Anwar Ibrahim, and Mahathir Mohamad, contributed to integrating Islamic principles into national education. This influence is reflected in the Malaysian National Education Philosophy, which emphasizes holistic development grounded in Islamic values. At the school level, the integration of Islamic studies into general education has been expanded through policies that blend religious and secular knowledge. The establishment of Islamic secondary schools and tahfiz institutions underscores the government’s efforts to develop an education system aligned with Islamic teachings. Additionally, tertiary institutions have introduced Islamic perspectives in various disciplines, including science, law, and economics, aiming to produce professionals guided by ethical and religious principles. Beyond traditional education, al-Faruqi’s influence extends to professional fields such as public health, where Islamic values have been incorporated into medical ethics and healthcare training. However, the implementation of IoK faces challenges, particularly criticisms that non-Western frameworks may not align with global academic standards or local funding priorities. Furthermore, the increasing influence of Western educational models, market-driven policies, and accreditation demands has led to debates over whether the Islamization agenda is being diluted. This paper explores the extent to which al-Faruqi’s IoK principles continue to shape Malaysia’s education system amidst these evolving challenges. It argues that while efforts to integrate Islamic values persist, there is a need for continuous dialogue to balance Islamic epistemology with modern educational demands, ensuring that knowledge remains both relevant and rooted in ethical and spiritual foundations.

Sylvain René, André Souza dos Santos, Cristina Santos Duarte et al.

Resumo Esta revisão sistemática analisa o progresso e os desafios para atingir o Objetivo de Desenvolvimento Sustentável 3 nos países em desenvolvimento, com foco na saúde sexual e reprodutiva e nas tendências demográficas. O estudo examina diversos relatórios e dados para mostrar avanços no acesso a serviços de saúde reprodutiva, na redução das taxas de fertilidade e na saúde materna. No entanto, persistem obstáculos como lacunas na infraestrutura de saúde, desigualdades econômicas e barreiras culturais. A análise destaca a correlação entre o nível de desenvolvimento e o progresso em direção aos Objetivos de Desenvolvimento Sustentável das Organizações das Nações Unidas para 2030, com ênfase em vários países africanos. O artigo sublinha a importância de fortalecer os esforços atuais e identifica áreas que necessitam de melhorias para atingir plenamente as metas do Objetivo de Desenvolvimento Sustentável 3.

Hongwen LI

Amid the accelerating trend of global aging, intergenerational equity has emerged as a central issue in resource allocation and ethical governance. While contractarianism emphasizes institutional rationality and intergenerational responsibility-sharing, whereas Confucian ethics is rooted in family-based emotional obligations. Each approach holds distinct strengths and limitations: the former often overlooks cultural context and emotional support, whereas the latter lacks institutional guarantees. Within the Chinese context, a middle path integrating contractual principles with filial piety ethics is necessary. This involves reconstructing a tripartite responsibility system among family, state, and individual, establishing a three-tiered elderly care structure encompassing familial feedback, community coordination, and state-level support, and enhancing institutional legitimacy through ethical justification. Such a model aims to dynamically harmonize intergenerational relationships and achieve justice in eldercare.

Caroline Felisberto, Gabriela Yori Monteiro Shoji, Bruna Borges de Souza et al.

Considerando que o crescente uso das redes sociais está associado com o desrespeito de preceitos éticos estabelecidos pelo Código de Ética Odontológica, este artigo tem como objetivo analisar e identificar infrações éticas cometidas por perfis de Odontopediatria no Instagram. Foi realizado um estudo exploratório transversal, no qual buscou-se, durante aproximadamente 3 meses, por perfis de Odontopediatria nesta rede social. Posteriormente, as contas foram analisadas individualmente, os dados foram coletados e avaliados de forma descritiva, considerando postagens publicadas durante um período de dois anos. Para avaliar se houve ou não infração ética, foi explorada a ocorrência dos itens listados no Art.44 do Código de Ética Odontológica aprovado pela resolução CFO-118/2012 (CEO/2012). Dos 129 perfis analisados, a maioria cometeu ao menos uma infração ética. Das 24.563 publicações relacionadas à Odontologia, em 6.551 (26,7%) foram encontradas características que poderiam ser interpretadas como inadequações éticas aos incisos em análise, sendo a maioria (25,3%) considerada infração ética. Concluiu-se que a maioria dos perfis de Odontopediatria encontrados no Instagram estão em desacordo com os preceitos éticos estabelecidos pelo Código de Ética Odontológica, sendo que as infrações éticas mais cometidas estão relacionadas a divulgação da imagem de pacientes.

Mairon Wesley Galvik Mendes

El personalismo ontológico y la bioética personalista colocan a la persona humana en el centro de las consideraciones éticas, abogando por un respeto que refleje la dignidad y el valor intrínsecos de la persona. Este trabajo busca evidenciar cuantitativamente cuál es la influencia académica de la BPOF en el mundo actual y cuál es la visión que los actuales expertos en bioética, considerados personalistas, tienende la misma BPOF, para así establecer una comparación con la propuesta original. Esto nos ayuda a entender mejor sus críticas y actualizaciones para favorecer una mejor comprensión de la propuesta de la BPOF y su diálogo con el mundo actual. El número de publicaciones muestra el influjo de la BPOF en el mundo académico. El análisis de estas revela que muchas de las publicaciones bioéticas catalogadas como personalistas no son específicas de la BPOF, pues carecen del fundamento, metodología o argumentación propiamente de la misma. Tal carencia puede llevar a que estas propuestas singulares no sean realmente efectivas en la defensa y promoción de la dignidad de todo ser humano, además de generar confusión en la comprensión de la BPOF por su falta de unidad.

Sicheng Wang, Che Liu, Rossella Arcucci

Recent advancements in medical vision-language pre-training (MedVLP) have significantly enhanced zero-shot medical vision tasks such as image classification by leveraging large-scale medical image-text pair pre-training. However, the performance of these tasks can be heavily influenced by the variability in textual prompts describing the categories, necessitating robustness in MedVLP models to diverse prompt styles. Yet, this sensitivity remains underexplored. In this work, we are the first to systematically assess the sensitivity of three widely-used MedVLP methods to a variety of prompts across 15 different diseases. To achieve this, we designed six unique prompt styles to mirror real clinical scenarios, which were subsequently ranked by interpretability. Our findings indicate that all MedVLP models evaluated show unstable performance across different prompt styles, suggesting a lack of robustness. Additionally, the models' performance varied with increasing prompt interpretability, revealing difficulties in comprehending complex medical concepts. This study underscores the need for further development in MedVLP methodologies to enhance their robustness to diverse zero-shot prompts.

Miguel Rios

Large Language Models (LLMs) have shown promising results on machine translation for high resource language pairs and domains. However, in specialised domains (e.g. medical) LLMs have shown lower performance compared to standard neural machine translation models. The consistency in the machine translation of terminology is crucial for users, researchers, and translators in specialised domains. In this study, we compare the performance between baseline LLMs and instruction-tuned LLMs in the medical domain. In addition, we introduce terminology from specialised medical dictionaries into the instruction formatted datasets for fine-tuning LLMs. The instruction-tuned LLMs significantly outperform the baseline models with automatic metrics.

Qiang Qiao, Wenyu Wang, Meixia Qu et al.

The field of medical image segmentation is challenged by domain generalization (DG) due to domain shifts in clinical datasets. The DG challenge is exacerbated by the scarcity of medical data and privacy concerns. Traditional single-source domain generalization (SSDG) methods primarily rely on stacking data augmentation techniques to minimize domain discrepancies. In this paper, we propose Random Amplitude Spectrum Synthesis (RASS) as a training augmentation for medical images. RASS enhances model generalization by simulating distribution changes from a frequency perspective. This strategy introduces variability by applying amplitude-dependent perturbations to ensure broad coverage of potential domain variations. Furthermore, we propose random mask shuffle and reconstruction components, which can enhance the ability of the backbone to process structural information and increase resilience intra- and cross-domain changes. The proposed Random Amplitude Spectrum Synthesis for Single-Source Domain Generalization (RAS^4DG) is validated on 3D fetal brain images and 2D fundus photography, and achieves an improved DG segmentation performance compared to other SSDG models.

AR. F. AlSamhori, F. Alnaimat

Artificial Intelligence (AI) is a field that utilizes computer technology to imitate, improve, and expand human intelligence. The concept of AI was originally proposed in the mid-twentieth century, and it has evolved into a technology that serves different purposes, ranging from simple automation to complex decision-making processes. AI encompasses Artificial Narrow Intelligence, General Intelligence, and Super Intelligence. AI is transforming data analysis, language checks, and literature reviews in research. In many fields of AI applications, ethical considerations, including plagiarism, bias, privacy, responsibility, and transparency, need precise norms and human oversight. By promoting understanding and adherence to ethical principles, the research community may successfully utilize the advantages of AI while upholding academic accountability and integrity. It takes teamwork from all stakeholders to improve human knowledge and creativity, and ethical AI use in research is essential.

Nicolás Gaggion, Rodrigo Echeveste, Lucas Mansilla et al.

It has recently been shown that deep learning models for anatomical segmentation in medical images can exhibit biases against certain sub-populations defined in terms of protected attributes like sex or ethnicity. In this context, auditing fairness of deep segmentation models becomes crucial. However, such audit process generally requires access to ground-truth segmentation masks for the target population, which may not always be available, especially when going from development to deployment. Here we propose a new method to anticipate model biases in biomedical image segmentation in the absence of ground-truth annotations. Our unsupervised bias discovery method leverages the reverse classification accuracy framework to estimate segmentation quality. Through numerical experiments in synthetic and realistic scenarios we show how our method is able to successfully anticipate fairness issues in the absence of ground-truth labels, constituting a novel and valuable tool in this field.

Stephen R. Milford

Abstract Illness has a way of disorientating us, as if we are cast adrift in a foreign land. Like strangers in a dessert we seek oasis to recollect ourselves, find refuge and learn to build our own shelters. Using the philosophy of Levinas and Derrida, we can interpret health care providers (HCP), and the sites from which they act (e.g. hospitals), as dwelling hosts that offer hospitality to strangers in this foreign land. While often the dwellings are physical (e.g. hospitals), this is not always the case. Language represents a mobile home of refuge to the sick. Using language the HCP has built a shelter so as to dwell in the land of illness. However, while hospitality is an inviting concept, it also implies hostility. The door that opens may also be slammed shut. This article explores the paradox of the linguistic mobile home offered to patients. It highlights the power of language to construct a safe place in a strange land, but also explores the inherent violence. It ends with an exploration of the ways language can be used by HCP to assist patients to construct their own mobile shelters.

Mamak Tahmasebi

-----

Rosa Ruiz Aragoneses, María de las Mercedes López Mateo

El presente estudio tiene por objeto comprobar si la narratividad y el método hermenéutico-fenomenológico son una vía de acceso al conocimiento de la experiencia ajena del sufrimiento. Para ello, se empleará como caso concreto de estudio el Centro de Humanización de la Salud San Camilo en Madrid (de aquí en adelante CEHS) a través de cinco entrevistas semidirigidas a un grupo de residentes y su posterior análisis descriptivo. En consecuencia, la pregunta de hipótesis a la que queremos responder es la siguiente: ¿podemos acceder a la esfera privada del sufrimiento del otro a través de su narratividad? Los resultados muestran que la escucha y el despliegue de la narratividad no solo sirven para aproximarnos más al sufrimiento ajeno, sino que existe ya en esta puesta en común una experiencia de sanación para nuestros mayores.