Hasil untuk "Medicine"

G. Buettner

W. Bolch, K. Eckerman, G. Sgouros et al.

P. Li, C. Szeto, B. Piraino et al.

Hasi Hays, William J. Richardson

Network topology excels at structural predictions but fails to capture functional semantics encoded in biomedical literature. We present a retrieval-augmented generation (RAG) embedding framework that integrates graph neural network representations with dynamically retrieved literature-derived knowledge through contrastive learning. Benchmarking against ten embedding methods reveals task-specific complementarity: topology-focused methods achieve near-perfect link prediction (GCN: 0.983 AUROC), while RAG-GNN is the only method achieving positive silhouette scores for functional clustering (0.001 vs. negative scores for all baselines). Information-theoretic decomposition shows network topology contributes 77.3% of predictive information, while retrieved documents provide 8.6% unique information. Applied to cancer signaling networks (379 proteins, 3,498 interactions), the framework identifies DDR1 as a therapeutic target based on retrieved evidence of synthetic lethality with KRAS mutations. These results establish that topology-only and retrieval-augmented approaches serve complementary purposes: structural prediction tasks are solved by network topology alone, while functional interpretation uniquely benefits from retrieved knowledge.

Kaiyuan Ji, Yijin Guo, Zicheng Zhang et al.

With the increasing use of large language models (LLMs) in medical decision-support, it is essential to evaluate not only their final answers but also the reliability of their reasoning. Two key risks are Chain-of-Thought (CoT) faithfulness -- whether reasoning aligns with responses and medical facts -- and sycophancy, where models follow misleading cues over correctness. Existing benchmarks often collapse such vulnerabilities into single accuracy scores. To address this, we introduce MedOmni-45 Degrees, a benchmark and workflow designed to quantify safety-performance trade-offs under manipulative hint conditions. It contains 1,804 reasoning-focused medical questions across six specialties and three task types, including 500 from MedMCQA. Each question is paired with seven manipulative hint types and a no-hint baseline, producing about 27K inputs. We evaluate seven LLMs spanning open- vs. closed-source, general-purpose vs. medical, and base vs. reasoning-enhanced models, totaling over 189K inferences. Three metrics -- Accuracy, CoT-Faithfulness, and Anti-Sycophancy -- are combined into a composite score visualized with a 45 Degrees plot. Results show a consistent safety-performance trade-off, with no model surpassing the diagonal. The open-source QwQ-32B performs closest (43.81 Degrees), balancing safety and accuracy but not leading in both. MedOmni-45 Degrees thus provides a focused benchmark for exposing reasoning vulnerabilities in medical LLMs and guiding safer model development.

Maximilian Schuessler, Erik Sverdrup, Robert Tibshirani

Robust estimation of heterogeneous treatment effects is a fundamental challenge for optimal decision-making in domains ranging from personalized medicine to educational policy. In recent years, predictive machine learning has emerged as a valuable toolbox for causal estimation, enabling more flexible effect estimation. However, accurately estimating conditional average treatment effects (CATE) remains a major challenge, particularly in the presence of many covariates. In this article, we propose pretraining strategies that leverage a phenomenon in real-world applications: factors that are prognostic of the outcome are frequently also predictive of treatment effect heterogeneity. In medicine, for example, components of the same biological signaling pathways frequently influence both baseline risk and treatment response. Specifically, we demonstrate our approach within the R-learner framework, which estimates the CATE by solving individual prediction problems based on a residualized loss. We use this structure to incorporate side information and develop models that can exploit synergies between risk prediction and causal effect estimation. In settings where these synergies are present, this cross-task learning enables more accurate signal detection, yields lower estimation error, reduced false discovery rates, and higher power for detecting heterogeneity.

Rawan Alyahya, Asrar Alruwayqi, Atheer Alqarni et al.

The presence of MGMT promoter methylation significantly affects how well chemotherapy works for patients with Glioblastoma Multiforme (GBM). Currently, confirmation of MGMT promoter methylation relies on invasive brain tumor tissue biopsies. In this study, we explore radiogenomics techniques, a promising approach in precision medicine, to identify genetic markers from medical images. Using MRI scans and deep learning models, we propose a new multi-view approach that considers spatial relationships between MRI views to detect MGMT methylation status. Importantly, our method extracts information from all three views without using a complicated 3D deep learning model, avoiding issues associated with high parameter count, slow convergence, and substantial memory demands. We also introduce a new technique for tumor slice extraction and show its superiority over existing methods based on multiple evaluation metrics. By comparing our approach to state-of-the-art models, we demonstrate the efficacy of our method. Furthermore, we share a reproducible pipeline of published models, encouraging transparency and the development of robust diagnostic tools. Our study highlights the potential of non-invasive methods for identifying MGMT promoter methylation and contributes to advancing precision medicine in GBM treatment.

Ahmed Morad Asaad, Sara A. Saied, Mohammad M. Torayah et al.

Abstract Background Recent advances in nanomedicine have derived novel prospects for development of various bioactive nanoparticles and nanocomposites with significant antibacterial and antifungal properties. This study aims to investigate some characteristics of the novel Se-NPs/Cu2O nanocomposite such as morphological, physicochemical, and optical properties, as well as to assess the antibacterial activity of this fabricated composite in different concentrations against some MDR Gram-positive and Gram-negative clinical bacterial isolates. Methods The Se-NPs/Cu2O nanocomposite was fabricated using the chemical deposition method. The fabricated nanocomposite was fully characterized by X-Ray diffraction analysis (XRD), fourier transforms infrared spectroscopy (FTIR), and transmission electron microscope (TEM). The antimicrobial activity of Se-NPs/Cu2O was investigated using the standard broth microdilution method. The fabricated Se-NPs/Cu2O nanocomposites were detected as stable and highly crystallized nanospheres with an average size of 98.6 nm. Results The Se-NPs/Cu2O nanocomposite showed a potent antimicrobial activity with MIC values ranged from 6.25 to 12.5 µg/ml for Gram-positive isolates, and 25 to 50 µg/ml for gram-negative isolates. The bactericidal activity was higher for gram-negative isolates with MBC/MIC ratios of 1–2 µg/ml for gram-negative, versus 8 µg/ml for gram positive pathogens. Conclusion These findings would support further research in development of a novel Se-NPs/Cu2O nanocomposite as a promising alternative therapeutic option for improving the quality of patients’ management.

Anastasia Krithara, Fotis Aisopos, Vassiliki Rentoumi et al.

The vision of IASIS project is to turn the wave of big biomedical data heading our way into actionable knowledge for decision makers. This is achieved by integrating data from disparate sources, including genomics, electronic health records and bibliography, and applying advanced analytics methods to discover useful patterns. The goal is to turn large amounts of available data into actionable information to authorities for planning public health activities and policies. The integration and analysis of these heterogeneous sources of information will enable the best decisions to be made, allowing for diagnosis and treatment to be personalised to each individual. The project offers a common representation schema for the heterogeneous data sources. The iASiS infrastructure is able to convert clinical notes into usable data, combine them with genomic data, related bibliography, image data and more, and create a global knowledge base. This facilitates the use of intelligent methods in order to discover useful patterns across different resources. Using semantic integration of data gives the opportunity to generate information that is rich, auditable and reliable. This information can be used to provide better care, reduce errors and create more confidence in sharing data, thus providing more insights and opportunities. Data resources for two different disease categories are explored within the iASiS use cases, dementia and lung cancer.

Mai M. Assi, Mohammed E. Grawish, Heba Mahmoud Elsabaa et al.

Abstract Conditioned media (CM) is derived from mesenchymal stem cells (MSC) culture and contains biologically active components. CM is easy to handle and reduces inflammation while repairing injured joints. Combination therapy of the CM with cross-linked hyaluronic acid (HA) could ameliorate the beneficial effect of HA in treating degenerative changes of articulating surfaces associated with arthritic rats’ temporomandibular joints (TMJs). This study aimed to evaluate the therapeutic potential of HA hydrogel combined with bone marrow stem cells-conditioned medium (BMSCs-CM) on the articulating surfaces of TMJs associated with complete Freund’s adjuvant (CFA)-induced arthritis. Fifty female Sprague-Dawley rats were divided randomly into five equal groups. Rats of group I served as the negative controls and received intra-articular (IA) injections of 50 µl saline solution, whereas rats of group II were subjected to twice IA injections of 50 µg CFA in 50 µl; on day 1 of the experiment to induce persistent inflammation and on day 14 to induce arthritis. Rats of group III and IV were handled as group II and instead, they received an IA injection of 50 µl HA hydrogel and 50 µl of BMSCs-CM, respectively. Rats of group V were given combined IA injections of 50 µl HA hydrogel and BMSCs-CM. All rats were euthanized after the 4th week of inducing arthritis. The joints were processed for sectioning and histological staining using hematoxylin and eosin, Masson’s trichrome and toluidine blue special staining, and immunohistochemical staining for nuclear factor-kappa B (NF-κB). SPSS software was used to analyze the data and one-way analysis of variance followed by post-hoc Tukey statistical tests were used to test the statistical significance at 0.05 for alpha and 0.2 for beta. In the pooled BMSC-CM, 197.14 pg/ml of platelet-derived growth factor and 112.22 pg/ml of interleukin-10 were detected. Compared to TMJs of groups III and IV, TMJs of group V showed significant improvements (P = 0.001) in all parameters tested as the disc thickness was decreased (331.79 ± 0.73), the fibrocartilaginous layer was broadened (0.96 ± 0.04), and the amount of the trabecular bone was distinctive (19.35 ± 1.07). The mean values for the collagen amount were increased (12.29 ± 1.38) whereas the mean values for the NF-κB expression were decreased (0.62 ± 0.15). Combination therapy of HA hydrogel and BMSCs-CM is better than using HA hydrogel or BMSCs-CM, separately to repair degenerative changes in rats’ TMJs associated with CFA-induced arthritis.

Md. Makshuder Rahman Zim, Nurnabi Ahmed, Mostak Ahmed et al.

Bovine anaplasmosis is an infectious, tick-borne disease caused by Anaplasma species, which is accountable for huge economic loss in dairy industry. This study was aimed to determine the seroprevalence of bovine anaplasmosis on randomly selected 61 commercial dairy farms in 3 intensive regions of Bangladesh. A total of 1472 sera were analysed using VMRD Anaplasma Antibody Test Kit cELISA v2 for the presence of Anaplasma-specific antibodies. The highest regional seroprevalence of Anaplasma was 45.93% in individual level and 74.4% in herd level recorded in the southeast region, whereas it was 48.8% in individual level and 83.3% in herd level in Khagrachari and Sherpur districts, indicating an emerging state of the disease. The herd size and type in herd level and regions, districts, sex, age and breed in individual level were significantly (P ≤ 0.05) associated with anaplasmosis. Multivariate logistic regression analysis showed that cattle aged >1 year had 1.86 times higher odds compared to cattle younger than 1 year. Dairy cows had the highest odds (2.25) of anaplasmosis, followed by dairy heifers (1.68), compared to bulls. Compared to herd sizes of <4, the odds of Anaplasma infection were 11.3 and 7.45 times greater in herd sizes of >28 and 4–28. Crossbred cattle had 2.4 times higher odds of anaplasmosis compared to indigenous cattle. This first seroprevalence study signifies the widespread presence and underscores the importance of monitoring and managing anaplasmosis to safeguard cattle health in Bangladesh. Study on the molecular epidemiology and genetic diversity of Anaplasma among cattle populations should be prioritized.

Chenlong Yang, Xiaohui Lou, Xiaohui Lou et al.

ObjectiveThis study aimed to develop an arbitrary-dimensional nerve root reconstruction magnetic resonance imaging (ANRR-MRI) technique for identifying the leakage orificium of sacral meningeal cysts (SMCs) without spinal nerve root fibres (SNRFs).MethodsThis prospective study enrolled 40 consecutive patients with SMCs without SNRFs between March 2021 and March 2022. Magnetic resonance neural reconstruction sequences were performed for preoperative evaluation. The cyst and the cyst-dura intersection planes were initially identified based on the original thin-slice axial T2-weighted images. Sagittal and coronal images were then reconstructed by setting each intersecting plane as the centre. Then, three-dimensional reconstruction was performed, focusing on the suspected leakage point of the cyst. Based on the identified leakage location and size of the SMC, individual surgical plans were formulated.ResultsThis cohort included 30 females and 10 males, with an average age of 42.6 ± 12.2 years (range, 17–66 years). The leakage orificium was located at the rostral pole of the cyst in 23 patients, at the body region of the cyst in 12 patients, and at the caudal pole in 5 patients. The maximum diameter of the cysts ranged from 2 cm to 11 cm (average, 5.2 ± 1.9 cm). The leakage orificium was clearly identified in all patients and was ligated microscopically through a 4 cm minimally invasive incision. Postoperative imaging showed that the cysts had disappeared.ConclusionANRR-MRI is an accurate and efficient approach for identifying leakage orificium, facilitating the precise diagnosis and surgical treatment of SMCs without SNRFs.

Pengcheng Chen, Ziyan Huang, Zhongying Deng et al.

OpenAI's latest large vision-language model (LVLM), GPT-4V(ision), has piqued considerable interest for its potential in medical applications. Despite its promise, recent studies and internal reviews highlight its underperformance in specialized medical tasks. This paper explores the boundary of GPT-4V's capabilities in medicine, particularly in processing complex imaging data from endoscopies, CT scans, and MRIs etc. Leveraging open-source datasets, we assessed its foundational competencies, identifying substantial areas for enhancement. Our research emphasizes prompt engineering, an often-underutilized strategy for improving AI responsiveness. Through iterative testing, we refined the model's prompts, significantly improving its interpretative accuracy and relevance in medical imaging. From our comprehensive evaluations, we distilled 10 effective prompt engineering techniques, each fortifying GPT-4V's medical acumen. These methodical enhancements facilitate more reliable, precise, and clinically valuable insights from GPT-4V, advancing its operability in critical healthcare environments. Our findings are pivotal for those employing AI in medicine, providing clear, actionable guidance on harnessing GPT-4V's full diagnostic potential.

Rodrigo Agerri, Iñigo Alonso, Aitziber Atutxa et al.

Providing high quality explanations for AI predictions based on machine learning is a challenging and complex task. To work well it requires, among other factors: selecting a proper level of generality/specificity of the explanation; considering assumptions about the familiarity of the explanation beneficiary with the AI task under consideration; referring to specific elements that have contributed to the decision; making use of additional knowledge (e.g. expert evidence) which might not be part of the prediction process; and providing evidence supporting negative hypothesis. Finally, the system needs to formulate the explanation in a clearly interpretable, and possibly convincing, way. Given these considerations, ANTIDOTE fosters an integrated vision of explainable AI, where low-level characteristics of the deep learning process are combined with higher level schemes proper of the human argumentation capacity. ANTIDOTE will exploit cross-disciplinary competences in deep learning and argumentation to support a broader and innovative view of explainable AI, where the need for high-quality explanations for clinical cases deliberation is critical. As a first result of the project, we publish the Antidote CasiMedicos dataset to facilitate research on explainable AI in general, and argumentation in the medical domain in particular.

Piotr Kica, Magdalena Otta, Krzysztof Czechowicz et al.

Digital twins are virtual representations of physical objects or systems used for the purpose of analysis, most often via computer simulations, in many engineering and scientific disciplines. Recently, this approach has been introduced to computational medicine, within the concept of Digital Twin in Healthcare (DTH). Such research requires verification and validation of its models, as well as the corresponding sensitivity analysis and uncertainty quantification (VVUQ). From the computing perspective, VVUQ is a computationally intensive process, as it requires numerous runs with variations of input parameters. Researchers often use high-performance computing (HPC) solutions to run VVUQ studies where the number of parameter combinations can easily reach tens of thousands. However, there is a viable alternative to HPC for a substantial subset of computational models - serverless computing. In this paper we hypothesize that using the serverless computing model can be a practical and efficient approach to selected cases of running VVUQ calculations. We show this on the example of the EasyVVUQ library, which we extend by providing support for many serverless services. The resulting library - CloudVVUQ - is evaluated using two real-world applications from the computational medicine domain adapted for serverless execution. Our experiments demonstrate the scalability of the proposed approach.

Nicoletta Prentzas, Antonis Kakas, Constantinos S. Pattichis

Artificial Intelligence in Medicine has made significant progress with emerging applications in medical imaging, patient care, and other areas. While these applications have proven successful in retrospective studies, very few of them were applied in practice.The field of Medical AI faces various challenges, in terms of building user trust, complying with regulations, using data ethically.Explainable AI (XAI) aims to enable humans understand AI and trust its results. This paper presents a literature review on the recent developments of XAI solutions for medical decision support, based on a representative sample of 198 articles published in recent years. The systematic synthesis of the relevant articles resulted in several findings. (1) model-agnostic XAI techniques were mostly employed in these solutions, (2) deep learning models are utilized more than other types of machine learning models, (3) explainability was applied to promote trust, but very few works reported the physicians participation in the loop, (4) visual and interactive user interface is more useful in understanding the explanation and the recommendation of the system. More research is needed in collaboration between medical and AI experts, that could guide the development of suitable frameworks for the design, implementation, and evaluation of XAI solutions in medicine.

Demi Brownlie, Andreas von Kries, Giampiero Valenzano et al.

Lung cancer is a leading cause of cancer-related death worldwide. Despite recent advances in tissue immunology, little is known about the spatial distribution of tissue-resident lymphocyte subsets in lung tumors. Using high-parameter flow cytometry, we identified an accumulation of tissue-resident lymphocytes including tissue-resident NK (trNK) cells and CD8+ tissue-resident memory T (TRM) cells toward the center of human non-small cell lung carcinomas (NSCLC). Chemokine receptor expression patterns indicated different modes of tumor-infiltration and/or residency between trNK cells and CD8+ TRM cells. In contrast to CD8+ TRM cells, trNK cells and ILCs generally expressed low levels of immune checkpoint receptors independent of location in the tumor. Additionally, granzyme expression in trNK cells and CD8+ TRM cells was highest in the tumor center, and intratumoral CD49a+CD16− NK cells were functional and responded stronger to target cell stimulation than their CD49a− counterparts, indicating functional relevance of trNK cells in lung tumors.In summary, the present spatial mapping of lymphocyte subsets in human NSCLC provides novel insights into the composition and functionality of tissue-resident immune cells, suggesting a role for trNK cells and CD8+ TRM cells in lung tumors and their potential relevance for future therapeutic approaches.

Juha Gogulski, Jessica M. Ross, Austin Talbot et al.

Personalized treatments are gaining momentum across all fields of medicine. Precision medicine can be applied to neuromodulatory techniques, where focused brain stimulation treatments such as repetitive transcranial magnetic stimulation (rTMS) are used to modulate brain circuits and alleviate clinical symptoms. rTMS is well-tolerated and clinically effective for treatment-resistant depression (TRD) and other neuropsychiatric disorders. However, despite its wide stimulation parameter space (location, angle, pattern, frequency, and intensity can be adjusted), rTMS is currently applied in a one-size-fits-all manner, potentially contributing to its suboptimal clinical response (~50%). In this review, we examine components of rTMS that can be optimized to account for inter-individual variability in neural function and anatomy. We discuss current treatment options for TRD, the neural mechanisms thought to underlie treatment, differences in FDA-cleared devices, targeting strategies, stimulation parameter selection, and adaptive closed-loop rTMS to improve treatment outcomes. We suggest that better understanding of the wide and modifiable parameter space of rTMS will greatly improve clinical outcome.

Fenglin Liu, Bang Yang, Chenyu You et al.

Language models (LMs), including large language models (such as ChatGPT), have the potential to assist clinicians in generating various clinical notes. However, LMs are prone to produce ``hallucinations'', i.e., generated content that is not aligned with facts and knowledge. In this paper, we propose the Re$^3$Writer method with retrieval-augmented generation and knowledge-grounded reasoning to enable LMs to generate faithful clinical texts. We demonstrate the effectiveness of our method in generating patient discharge instructions. It requires the LMs not to only understand the patients' long clinical documents, i.e., the health records during hospitalization, but also to generate critical instructional information provided both to carers and to the patient at the time of discharge. The proposed Re$^3$Writer imitates the working patterns of physicians to first \textbf{re}trieve related working experience from historical instructions written by physicians, then \textbf{re}ason related medical knowledge. Finally, it \textbf{re}fines the retrieved working experience and reasoned medical knowledge to extract useful information, which is used to generate the discharge instructions for previously-unseen patients. Our experiments show that, using our method, the performance of five representative LMs can be substantially boosted across all metrics. Meanwhile, we show results from human evaluations to measure the effectiveness in terms of fluency, faithfulness, and comprehensiveness.

Peter S Johnstone, Maite Ogueta, Olga Akay et al.

Circadian clocks are highly conserved transcriptional regulators that control ~24 hr oscillations in gene expression, physiological function, and behavior. Circadian clocks exist in almost every tissue and are thought to control tissue-specific gene expression and function, synchronized by the brain clock. Many disease states are associated with loss of circadian regulation. How and when circadian clocks fail during pathogenesis remains largely unknown because it is currently difficult to monitor tissue-specific clock function in intact organisms. Here, we developed a method to directly measure the transcriptional oscillation of distinct neuronal and peripheral clocks in live, intact Drosophila, which we term Locally Activatable BioLuminescence, or LABL. Using this method, we observed that specific neuronal and peripheral clocks exhibit distinct transcriptional properties. Loss of the receptor for PDF, a circadian neurotransmitter critical for the function of the brain clock, disrupts circadian locomotor activity but not all tissue-specific circadian clocks. We found that, while peripheral clocks in non-neuronal tissues were less stable after the loss of PDF signaling, they continued to oscillate. We also demonstrate that distinct clocks exhibit differences in their loss of oscillatory amplitude or their change in period, depending on their anatomical location, mutation, or fly age. Our results demonstrate that LABL is an effective tool that allows rapid, affordable, and direct real-time monitoring of individual clocks in vivo.