Mohsen Sadeghi, Fatemeh Sadat Hosseini Khajouei, Nilsa Dourandish
et al.
Abstract Mounier Kuhn syndrome (MKS), also known as tracheobronchomegaly (TBM), is a very rare and chronic airway disease characterized by marked dilatation of the trachea and central bronchi. Currently, there are few epidemiological studies on MKS, and most data are derived from case reports, leading to a limited understanding of the disease by clinicians. In this article, we present a 45-year-old male patient with MKS who had been symptomatic for a long time and had undergone various treatments due to a delay in diagnosis. This case highlights the importance of considering Mounier-Kuhn syndrome in patients with recurrent or unexplained respiratory infections to ensure timely diagnosis and prevent unnecessary treatments.
Diseases of the respiratory system, Medical emergencies. Critical care. Intensive care. First aid
The scenario approach is an established data-driven design framework that comes equipped with a powerful theory linking design complexity to generalization properties. In this approach, data are simultaneously used both for design and for certifying the design's reliability, without resorting to a separate test dataset. This paper takes a step further by guaranteeing additional properties, useful in post-design usage but not considered during the design phase. To this end, we introduce a two-level framework of appropriateness: baseline appropriateness, which guides the design process, and post-design appropriateness, which serves as a criterion for a posteriori evaluation. We provide distribution-free upper bounds on the risk of failing to meet the post-design appropriateness; these bounds are computable without using any additional test data. Under additional assumptions, lower bounds are also derived. As part of an effort to demonstrate the usefulness of the proposed methodology, the paper presents two practical examples in H2 and pole-placement problems. Moreover, a method is provided to infer comprehensive distributional knowledge of relevant performance indexes from the available dataset.
Large language models deployed in supportive or advisory roles must balance helpfulness with preservation of user autonomy, yet standard alignment methods primarily optimize for helpfulness and harmlessness without explicitly modeling relational risks such as dependency reinforcement, overprotection, or coercive guidance. We introduce Care-Conditioned Neuromodulation (CCN), a state-dependent control framework in which a learned scalar signal derived from structured user state and dialogue context conditions response generation and candidate selection. We formalize this setting as an autonomy-preserving alignment problem and define a utility function that rewards autonomy support and helpfulness while penalizing dependency and coercion. We also construct a benchmark of relational failure modes in multi-turn dialogue, including reassurance dependence, manipulative care, overprotection, and boundary inconsistency. On this benchmark, care-conditioned candidate generation combined with utility-based reranking improves autonomy-preserving utility by +0.25 over supervised fine-tuning and +0.07 over preference optimization baselines while maintaining comparable supportiveness. Pilot human evaluation and zero-shot transfer to real emotional-support conversations show directional agreement with automated metrics. These results suggest that state-dependent control combined with utility-based selection is a practical approach to multi-objective alignment in autonomy-sensitive dialogue.
This report summarizes changes in federal research funding to the medical physics community between FY24 and FY25. By linking the AAPM membership database with NIH RePORTER records, we quantified the distribution of NIH funding for projects led by AAPM researchers. Although total NIH funding to AAPM members remained relatively stable across the two years, the composition of that funding shifted substantially. Competing (new and renewal) awards declined 50%, driven largely by an 80% collapse in new R01 grants from the National Cancer Institute (NCI). In contrast, noncompeting continuation awards increased by 10%, following a shift in how NIH funds multi-year projects. These changes occurred in the context of widespread disruptions to NIH review and grantmaking, including delayed study sections and more stringent administrative requirements. Federal funding is essential to sustaining innovation, supporting early-stage investigators, and ensuring that patients receive the best possible care. The trends identified here raise concerns about the long-term vitality and stability of the medical physics research pipeline.
Victor Rudondy, Pierre-Antoine Barral, Thibaut Markarian
et al.
Abstract Background Acute mesenteric ischaemia (AMI) is an emergency with a poor prognosis. In France, a structure dedicated to AMI has been created in Paris in 2016 (SURVI), with promising results. A similar organization has been created in Marseille in 2021 (SOS AMI). Our aim was to compare the results of SOS AMI with those of a previous cohort of AMI patients managed without any dedicated structure. Methods The first 100 patients with AMI, managed by the SOS AMI, between November 2021 and December 2023 were prospectively included. They were compared with 100 AMI patients from a previous retrospective cohort (from January 2017 to December 2020), managed without any dedicated structure in the same center. Results The first 100 AMI patients managed by SOS AMI have similar demographic characteristics to those previously managed without SOS. The vascular causes of AMI were also similar between groups: arterial occlusive (61 vs. 56%, p = 0.5), venous occlusive (17 vs. 13%, p = 0.5), or non occlusive (22 vs. 31%, p = 0.2). AMI patients managed by SOS AMI were more frequently transferred from another center (41 vs. 19%, p = 0.001), had a shorter median time between CT scan and intervention (4 [range, 1-129] vs. 5 [0-285] hours, p = 0.05), a higher revascularisation rate (61 vs. 28%, p = 0.02), and lower 30-day (32 vs. 58%, p < 0.001) and 90-day (45 vs. 62%, p = 0.02) mortality rates. Conclusion The creation of SOS AMI has significantly improved the management of AMI patients, by better organizing the role of the various specialties involved, particularly in terms of revascularisation and survival rates. These promising results support the further development and expansion of this dedicated structure.
Surgery, Medical emergencies. Critical care. Intensive care. First aid
Understanding model decisions is crucial in medical imaging, where interpretability directly impacts clinical trust and adoption. Vision Transformers (ViTs) have demonstrated state-of-the-art performance in diagnostic imaging; however, their complex attention mechanisms pose challenges to explainability. This study evaluates the explainability of different Vision Transformer architectures and pre-training strategies - ViT, DeiT, DINO, and Swin Transformer - using Gradient Attention Rollout and Grad-CAM. We conduct both quantitative and qualitative analyses on two medical imaging tasks: peripheral blood cell classification and breast ultrasound image classification. Our findings indicate that DINO combined with Grad-CAM offers the most faithful and localized explanations across datasets. Grad-CAM consistently produces class-discriminative and spatially precise heatmaps, while Gradient Attention Rollout yields more scattered activations. Even in misclassification cases, DINO with Grad-CAM highlights clinically relevant morphological features that appear to have misled the model. By improving model transparency, this research supports the reliable and explainable integration of ViTs into critical medical diagnostic workflows.
Medical image segmentation is a pivotal task within the realms of medical image analysis and computer vision. While current methods have shown promise in accurately segmenting major regions of interest, the precise segmentation of boundary areas remains challenging. In this study, we propose a novel network architecture named CTO, which combines Convolutional Neural Networks (CNNs), Vision Transformer (ViT) models, and explicit edge detection operators to tackle this challenge. CTO surpasses existing methods in terms of segmentation accuracy and strikes a better balance between accuracy and efficiency, without the need for additional data inputs or label injections. Specifically, CTO adheres to the canonical encoder-decoder network paradigm, with a dual-stream encoder network comprising a mainstream CNN stream for capturing local features and an auxiliary StitchViT stream for integrating long-range dependencies. Furthermore, to enhance the model's ability to learn boundary areas, we introduce a boundary-guided decoder network that employs binary boundary masks generated by dedicated edge detection operators to provide explicit guidance during the decoding process. We validate the performance of CTO through extensive experiments conducted on seven challenging medical image segmentation datasets, namely ISIC 2016, PH2, ISIC 2018, CoNIC, LiTS17, and BTCV. Our experimental results unequivocally demonstrate that CTO achieves state-of-the-art accuracy on these datasets while maintaining competitive model complexity. The codes have been released at: https://github.com/xiaofang007/CTO.
Abstract Background To explore a novel ultrasound (US) modality for lumbar transforaminal epidural injections (TFEIs) in patients with low back pain (LBP) and L5 radicular pain combined with high iliac crest (HIC). Methods One-hundred and forty-one patients were retrospectively stratified into two groups based on the treatment they received: novel group, receiving US-guided and fluoroscopy (FL)-controlled TFEIs using a sagittal oblique approach between the superior articular process of L5 and S1, and control group, receiving US-guided TFEIs with conventional transverse approach combined with FL confirmation. Accuracy of contrast dispersing into lumbar epidural space was set as the primary endpoint. Radiation dosages, procedure time, numeric rating scale (NRS) scores, Modified Oswestry Disability Questionnaire (MODQ) scores, adverse events, and rescue analgesic requirement were also recorded. The generalized liner mixed model (GLMMs) was employed to compare the repeatedly measured variables between groups, taking individual confounding factors as covariance. Results The accuracy of TFEIs was 92.8% and 65.2% in novel and control group, with a significant difference of 26.7% (95% CI: 15.4%, 39.8%) between two modalities (p < 0.001). Significant pain relief was observed in novel group as opposed to control group after one injection. Procedure time in novel group (8.4 ± 1.6 min) was shorter than control group (15.8 ± 3.5 min) (p < 0.001) with less radiation dosage (3047 ± 5670 vs. 8808 ± 1039 μGy/m2, p < 0.001). Significantly, lower incidence of L5 paresthesia occurred in novel group. Statistical differences of NRS scores between the novel and control group were reached at 1 week after procedure (1 (IQR: − 1–3) vs. 3 (IQR: − 1–7), p = 0.006), while not reached at both 1- (1 (IQR: 0–2) vs. 1 (IQR: − 1–3), p = 0.086) or 3-month follow-up (0 (IQR: − 1–1) vs. 1 (IQR: 0–2), p = 0.094). Both groups showed similar functional improvement (F = 0.103, p = 0.749) during follow-up. Conclusions The novel sonographic technique provided superior accuracy needle placement and better pain-relieving effect through one injection as compared to the routine transverse approach. Consequently, in situations where the HIC imposed limitations for TFEIs performance on L5, the novel technique should be recommended to consider increasing accurate puncture, minimizing radiation exposure, consuming procedure time, and reducing the risk of neuraxial injury.
Anesthesiology, Medical emergencies. Critical care. Intensive care. First aid
Hospital-to-home transitions are a critical component of effective healthcare delivery, especially for patients aged 75 and older. This study evaluates the cost-effectiveness of the 'Your Care Needs You' (YCNY) intervention, a patient-centred approach designed to empower older adults during discharge, compared to standard care. The analysis adopts the perspective of the National Health Service (NHS) and Personal Social Services. Data were drawn from a cluster randomised controlled trial (cRCT) conducted within the UK NHS over a 90-day post-discharge follow-up period. Adjusted differences in costs and Quality-Adjusted Life Years (QALYs) were estimated using Multilevel Mixed-Effects Generalised Linear Models (MME-GLM) to account for the hierarchical structure of the trial design. Alternatively, Seemingly Unrelated Regression (SUR) models were employed to address potential correlations between costs and QALYs. Scenario analyses and probabilistic sensitivity analyses were conducted to assess the robustness of the results. The YCNY intervention reduced costs by £269 and achieved a QALY gain of 0.0057, resulting in a net health benefit (NHB) of 0.0246 at a £15,000/QALY threshold. It demonstrated an 89% probability of cost-effectiveness compared to standard care within the trial's time horizon. Findings remained robust across alternative scenarios and sensitivity analyses. These results suggest that YCNY is a promising and potentially cost-effective strategy for improving hospital-to-home transitions for older adults. By enhancing outcomes and reducing costs, the study supports integrating patient-centred interventions like YCNY into routine NHS practice, with the potential to improve both efficiency and quality of healthcare delivery.
Recently, the research community of computerized medical imaging has started to discuss and address potential fairness issues that may emerge when developing and deploying AI systems for medical image analysis. This chapter covers some of the pressing challenges encountered when doing research in this area, and it is intended to raise questions and provide food for thought for those aiming to enter this research field. The chapter first discusses various sources of bias, including data collection, model training, and clinical deployment, and their impact on the fairness of machine learning algorithms in medical image computing. We then turn to discussing open challenges that we believe require attention from researchers and practitioners, as well as potential pitfalls of naive application of common methods in the field. We cover a variety of topics including the impact of biased metrics when auditing for fairness, the leveling down effect, task difficulty variations among subgroups, discovering biases in unseen populations, and explaining biases beyond standard demographic attributes.
Chronic wounds, including diabetic ulcers, pressure ulcers, and ulcers secondary to venous hypertension, affects more than 6.5 million patients and a yearly cost of more than $25 billion in the United States alone. Chronic wound treatment is currently a manual process, and we envision a future where robotics and automation will aid in this treatment to reduce cost and improve patient care. In this work, we present the development of the first robotic system for wound dressing removal which is reported to be the worst aspect of living with chronic wounds. Our method leverages differentiable physics-based simulation to perform gradient-based Model Predictive Control (MPC) for optimized trajectory planning. By integrating fracture mechanics of adhesion, we are able to model the peeling effect inherent to dressing adhesion. The system is further guided by carefully designed objective functions that promote both efficient and safe control, reducing the risk of tissue damage. We validated the efficacy of our approach through a series of experiments conducted on both synthetic skin phantoms and real human subjects. Our results demonstrate the system's ability to achieve precise and safe dressing removal trajectories, offering a promising solution for automating this essential healthcare procedure.
Pratik Gajane, Sean Newman, Mykola Pechenizkiy
et al.
Chronic pain significantly diminishes the quality of life for millions worldwide. While psychoeducation and therapy can improve pain outcomes, many individuals experiencing pain lack access to evidence-based treatments or fail to complete the necessary number of sessions to achieve benefit. Reinforcement learning (RL) shows potential in tailoring personalized pain management interventions according to patients' individual needs while ensuring the efficient use of scarce clinical resources. However, clinicians, patients, and healthcare decision-makers are concerned that RL solutions could exacerbate disparities associated with patient characteristics like race or gender. In this article, we study gender fairness in personalized pain care recommendations using a real-world application of reinforcement learning (Piette et al., 2022a). Here, adhering to gender fairness translates to minimal or no disparity in the utility received by subpopulations as defined by gender. We investigate whether the selection of relevant patient information (referred to as features) used to assist decision-making affects gender fairness. Our experiments, conducted using real-world data Piette, 2022), indicate that included features can impact gender fairness. Moreover, we propose an RL solution, NestedRecommendation, that demonstrates the ability: i) to adaptively learn to select the features that optimize for utility and fairness, and ii) to accelerate feature selection and in turn, improve pain care recommendations from early on, by leveraging clinicians' domain expertise.
Ezzat Hajmolarezaei, Farzaneh Khoroushi, Mahboubeh Haddad Nameghi
et al.
Cervical ectopic pregnancies (CEPs) are rare, comprising less than 1% of ectopic pregnancies. On one hand, the abundant blood supply of the cervix and its incompatibility to keep the pregnancy in progress increases the potential for bleeding following CEP, mortality, complications, and infertility in affected women. CEPs are more difficult to diagnose and get identified at later gestational ages. CEP is one of the rarest forms of ectopic pregnancies and most commonly are a consequence of assisted reproductive technology (ART); while definitive risk factors are not fully known. Possible risk factors include cervical and uterine anomalies, previous curettages or cesarean sections, smoking, tubal factor infertility, or in vitro fertilization (IVF) treatment. Our analysis of literature in 200 patient restricted to retrospective case series , showed that a history of previous C-section, uterine curettage or D&C procedures, and a history of using assisted reproductive technology might be some of the potential risk factors. The increasing application of Hegar dilators was hypothesized as the cause of the rise in CEPs. Dilation and curettage (D&C) might also make the subject vulnerable to CEP development in the future. Previous D&C history could be a potential predisposing factor that is common among CEP patients. In this review, we critically reviewed these potential risk factors. In conclusion, the risk factors of CEP and their effect on fertility are also not studied properly. The rarity of these cases makes it difficult to predict as well if the risk of their recurrence is elevated.
Medical emergencies. Critical care. Intensive care. First aid
Ya-Hui Chang, Yu-Wen Chien, Chiung-Hsin Chang
et al.
Abstract Background Understanding demographic profiles is essential to the assessment of health burden imposed by motor vehicle crashes (MVCs) on pregnant women. However, Asian studies that have examined it are lacking. The study aimed to describe the demographic characteristics and prevalence of MVCs involving pregnant women in Taiwan. Methods A cross-sectional study conducted by the Taiwan Birth Notification dataset from 2008 to 2017 was linked with the police-reported traffic collision registry to identify pregnant women involved in MVCs. The pregnant women were categorized according to their gestational age, age at delivery, the role of road user (driver, passenger, or pedestrian), and vehicle types (car, two-wheeled motor vehicle, or others). A chi-square test was performed for the significance test. Results A total of 22,134 (1.13%) pregnant women were involved in MVCs in the study period. Two-wheeled motor vehicle (47.9%) and driver (81.4%) were the mainly reported vehicle type and road user at the crash scenes, respectively. The majority of MVCs occurred in pregnant women aged 28–34 years. The number of MVCs rapidly declined after 37 weeks of gestation, especially two-wheeled motor vehicle or car crashes. However, the number of pedestrian victims climbed up during the third trimester. Conclusion Pregnant women are susceptible to MVCs regardless of their gestational age, role of a road user, or type of vehicle. The findings of this study emphasize the need for increased awareness of traffic collision prevention among pregnant women aged 28–34. In addition, improving pedestrian safety is essential for the reduction of pregnant victims.
Medical emergencies. Critical care. Intensive care. First aid, Public aspects of medicine
Amartya Chaudhuri, Shital mahendra Kuttarmare, Pradnya Milind Bhalerao
et al.
Background: Brachial plexus surgery requires neural repair with the use of intraoperative peripheral nerve stimulation without muscle relaxants.
Methods: Twelve cases were conducted under total intravenous anaesthesia, receiving intravenous propofol, fentanyl and dexmedetomidine infusion. Intraoperative hemodynamic conditions and postoperative functional recovery were assessed.
Results: 9 out of 12 cases were stable while one was in a lighter plane requiring 20 mg propofol and increased dexmedetomidine, two had bradycardia requiring reduced dexmedetomidine infusion. At three months, five cases showed improvement.
Conclusion: Satisfactory conditions were achieved including hemodynamic stability, and muscle-sparing improving prognoses of brachial plexus surgeries.
Anesthesiology, Medical emergencies. Critical care. Intensive care. First aid
Hussein Mozannar, Yuria Utsumi, Irene Y. Chen
et al.
A high-risk pregnancy is a pregnancy complicated by factors that can adversely affect the outcomes of the mother or the infant. Health insurers use algorithms to identify members who would benefit from additional clinical support. This work presents the implementation of a real-world ML-based system to assist care managers in identifying pregnant patients at risk of complications. In this retrospective evaluation study, we developed a novel hybrid-ML classifier to predict whether patients are pregnant and trained a standard classifier using claims data from a health insurance company in the US to predict whether a patient will develop pregnancy complications. These models were developed in cooperation with the care management team and integrated into a user interface with explanations for the nurses. The proposed models outperformed commonly used claim codes for the identification of pregnant patients at the expense of a manageable false positive rate. Our risk complication classifier shows that we can accurately triage patients by risk of complication. Our approach and evaluation are guided by human-centric design. In user studies with the nurses, they preferred the proposed models over existing approaches.
Background The high transmission and fatality rates of coronavirus disease 2019 (COVID-19) strain intensive care resources and affect the treatment and prognosis of critically ill patients without COVID-19. Therefore, this study evaluated the differences in characteristics, clinical course, and prognosis of critically ill medical patients without COVID-19 before and during the COVID-19 pandemic. Methods This retrospective cohort study included patients from three university-affiliated tertiary hospitals. Demographic data and data on the severity, clinical course, and prognosis of medical patients without COVID-19 admitted to the intensive care unit (ICU) via the emergency room (ER) before (from January 1 to May 31, 2019) and during (from January 1 to May 31, 2021) the COVID-19 pandemic were obtained from electronic medical records. Propensity score matching was performed to compare hospital mortality between patients before and during the pandemic. Results This study enrolled 1161 patients (619 before and 542 during the pandemic). During the COVID-19 pandemic, the Simplified Acute Physiology Score (SAPS) 3 and Sequential Organ Failure Assessment (SOFA) scores, assessed upon ER and ICU admission, were significantly higher than those before the pandemic ( p < 0.05). The lengths of stay in the ER, ICU, and hospital were also longer ( p < 0.05). Finally, the hospital mortality rates were higher during the pandemic than before (215 [39.7%] vs. 176 [28.4%], p < 0.001). However, in the propensity score-matched patients, hospital mortality did not differ between the groups ( p = 0.138). The COVID-19 pandemic did not increase the risk of hospital mortality (odds ratio [OR] 1.405, 95% confidence interval [CI], 0.937–2.107, p = 0.100). SAPS 3, SOFA score, and do-not-resuscitate orders increased the risk of in-hospital mortality in the multivariate logistic regression model. Conclusions In propensity score-matched patients with similarly severe conditions, hospital mortality before and during the COVID-19 pandemic did not differ significantly. However, hospital mortality was higher during the COVID-19 pandemic in unmatched patients in more severe conditions. These findings imply collateral damage to non-COVID-19 patients due to shortages in medical resources during the COVID-19 pandemic. Thus, strategic management of medical resources is required to avoid these consequences.