Hasil untuk "Diseases of the circulatory (Cardiovascular) system"

Siqi Ouyang, Xi Xu, Lei Li

This paper presents CMU's submission to the IWSLT 2025 Simultaneous Speech Translation (SST) task for translating unsegmented English speech into Chinese and German text in a streaming manner. Our end-to-end speech-to-text system integrates a chunkwise causal Wav2Vec 2.0 speech encoder, an adapter, and the Qwen2.5-7B-Instruct as the decoder. We use a two-stage simultaneous training procedure on robust speech segments curated from LibriSpeech, CommonVoice, and VoxPopuli datasets, utilizing standard cross-entropy loss. Our model supports adjustable latency through a configurable latency multiplier. Experimental results demonstrate that our system achieves 44.3 BLEU for English-to-Chinese and 25.1 BLEU for English-to-German translations on the ACL60/60 development set, with computation-aware latencies of 2.7 seconds and 2.3 seconds, and theoretical latencies of 2.2 and 1.7 seconds, respectively.

John Wu, Adam Cross, Jimeng Sun

Rare diseases affect 1 in 10 Americans, yet standard ICD coding systems fail to capture these conditions in electronic health records (EHR), leaving crucial information buried in clinical notes. Current approaches struggle with medical abbreviations, miss implicit disease mentions, raise privacy concerns with cloud processing, and lack clinical reasoning abilities. We present Rare Disease Mining Agents (RDMA), a framework that mirrors how medical experts identify rare disease patterns in EHR. RDMA connects scattered clinical observations that together suggest specific rare conditions. By handling clinical abbreviations, recognizing implicit disease patterns, and applying contextual reasoning locally on standard hardware, RDMA reduces privacy risks while improving F1 performance by upwards of 30\% and decreasing inferences costs 10-fold. This approach helps clinicians avoid the privacy risk of using cloud services while accessing key rare disease information from EHR systems, supporting earlier diagnosis for rare disease patients. Available at https://github.com/jhnwu3/RDMA.

Elliot Schumacher, Dhruv Naik, Anitha Kannan

Large language models (LLMs) have demonstrated impressive capabilities in disease diagnosis. However, their effectiveness in identifying rarer diseases, which are inherently more challenging to diagnose, remains an open question. Rare disease performance is critical with the increasing use of LLMs in healthcare settings. This is especially true if a primary care physician needs to make a rarer prognosis from only a patient conversation so that they can take the appropriate next step. To that end, several clinical decision support systems are designed to support providers in rare disease identification. Yet their utility is limited due to their lack of knowledge of common disorders and difficulty of use. In this paper, we propose RareScale to combine the knowledge LLMs with expert systems. We use jointly use an expert system and LLM to simulate rare disease chats. This data is used to train a rare disease candidate predictor model. Candidates from this smaller model are then used as additional inputs to black-box LLM to make the final differential diagnosis. Thus, RareScale allows for a balance between rare and common diagnoses. We present results on over 575 rare diseases, beginning with Abdominal Actinomycosis and ending with Wilson's Disease. Our approach significantly improves the baseline performance of black-box LLMs by over 17% in Top-5 accuracy. We also find that our candidate generation performance is high (e.g. 88.8% on gpt-4o generated chats).

Vivek Yadav, Anugrah Jain

India, as a predominantly agrarian economy, faces significant challenges in agriculture, including substantial crop losses caused by diseases, pests, and environmental stress. Early detection and accurate identification of diseases across different crops are critical for improving yield and ensuring food security. This paper proposes a deep learning based solution for detecting multiple diseases in multiple crops, aimed to cover India's diverse agricultural landscape. We first create a unified dataset encompassing images of 17 different crops and 34 different diseases from various available repositories. Proposed deep learning model is trained on this dataset and outperforms the state-of-the-art in terms of accuracy and the number of crops, diseases covered. We achieve a significant detection accuracy, i.e., 99 percent for our unified dataset which is 7 percent more when compared to state-of-the-art handling 14 crops and 26 different diseases only. By improving the number of crops and types of diseases that can be detected, proposed solution aims to provide a better product for Indian farmers.

Evan S. Manning, MD, MPP, Gautam R. Shroff, MD, David R. Jacobs, Jr., PhD et al.

Background: Inflammation plays a role in cardiovascular disease (CVD). We defined various noncardiovascular and noncancer conditions, both infectious and noninfectious, with a common basis of inflammation, collectively termed chronic inflammatory-related disease (ChrIRD). We describe ChrIRD and its interplay with CVD during follow-up in the Multi-Ethnic Study of Atherosclerosis. Objectives: The aim of the study was to describe ChrIRD, its associations with CVD, and its association with mortality. Methods: Participants were free of overt CVD at baseline with median 17.9 (Q1-Q3: 14.9-18.6) years of follow-up. ChrIRD was determined by review of hospitalization and death records of International Classification of Diseases codes. CVD diagnosis was adjudicated based on medical records. We performed time-dependent proportional hazard regressions to identify risks related to ChrIRD or CVD events. Results: MESA (Multi-Ethnic Study of Atherosclerosis) participants (n = 6,791) had a mean age of 62 ± 10 years, with 47% (3,201/6,791) men, 39% (2,617/6,791) White, 28% (1,882/6,791) Black, 22% (1,489/6,791) Hispanic, and 12% (803/6,791) Chinese race/ethnicity. ChrIRD was observed in 29% (1,965/6,791) and CVD in 21% (1,420/6,791); including 11% (761/6,791) with both conditions. Mortality after ChrIRD only was 47% (567/1,204; 95% CI: 44%-49%); after CVD only was 45% (300/659; 95% CI: 41%-49%); and after both conditions was 67% (510/761; 95% CI: 63%-70%). CVD was associated with increased risk of ChrIRD (HR: 1.48, 1.23-1.77) and ChrIRD was associated with increased risk of CVD (HR: 2.23, 1.97-2.52). Baseline inflammatory markers predicted both conditions. Conclusions: ChrIRD is common, present in all organ systems, and is associated with significant mortality, particularly in combination with CVD. The association between CVD and ChrIRD is bidirectional, and baseline inflammatory markers are associated with ChrIRD and CVD.

Shaelyn Cavanaugh, MD, Paul Lawrence Feingold, MD, MHS

Mario Padilla Rodriguez, Mohamed Nafea

Cardiovascular diseases are a leading cause of mortality worldwide, highlighting the need for accurate diagnostic methods. This study benchmarks centralized and federated machine learning algorithms for heart disease classification using the UCI dataset which includes 920 patient records from four hospitals in the USA, Hungary and Switzerland. Our benchmark is supported by Shapley-value interpretability analysis to quantify features' importance for classification. In the centralized setup, various binary classification algorithms are trained on pooled data, with a support vector machine (SVM) achieving the highest testing accuracy of 83.3\%, surpassing the established benchmark of 78.7\% with logistic regression. Additionally, federated learning algorithms with four clients (hospitals) are explored, leveraging the dataset's natural partition to enhance privacy without sacrificing accuracy. Federated SVM, an uncommon approach in the literature, achieves a top testing accuracy of 73.8\%. Our interpretability analysis aligns with existing medical knowledge of heart disease indicators. Overall, this study establishes a benchmark for efficient and interpretable pre-screening tools for heart disease while maintaining patients' privacy. This work is available at https://github.com/padillma1/Heart-Disease-Classification-on-UCI-dataset-and-Shapley-Interpretability-Analysis.

Ashraf Mohammed Anwar, Hattan H. Alshawkani, Ibrahim Albakri et al.

A unicuspid aortic valve (UAV) in adults is a very rare form of aortic valve (AV) malformation. UAV has two distinct subtypes, acommissural UAV and unicommissural, and can be differentiated by anatomical features, imaging modalities, and clinical presentation. With the development of significant AV lesion (s), surgical or transcatheter intervention will be required. The first part is a summarized review of UAV (anatomical features, clinical presentation, diagnostic modalities, and management). In the second part, we present a series of four patients diagnosed with UAV (3 unicommissural and 1 acommissural). The first case underwent balloon aortic valvuloplasty during childhood and surgical AV replacement later, with the progression to severe aortic stenosis (AS). The second case underwent a Ross procedure. The third and fourth cases were asymptomatic with moderate AS and mild-to-moderate AR and were kept on follow-up. In all the cases, transesophageal echocardiography confirmed the diagnosis of UAV with detailed morphological and functional assessment of AV.

Tian Bian, Yuli Jiang, Jia Li et al.

Many patients with chronic diseases resort to multiple medications to relieve various symptoms, which raises concerns about the safety of multiple medication use, as severe drug-drug antagonism can lead to serious adverse effects or even death. This paper presents a Decision Support System, called DSSDDI, based on drug-drug interactions to support doctors prescribing decisions. DSSDDI contains three modules, Drug-Drug Interaction (DDI) module, Medical Decision (MD) module and Medical Support (MS) module. The DDI module learns safer and more effective drug representations from the drug-drug interactions. To capture the potential causal relationship between DDI and medication use, the MD module considers the representations of patients and drugs as context, DDI and patients' similarity as treatment, and medication use as outcome to construct counterfactual links for the representation learning. Furthermore, the MS module provides drug candidates to doctors with explanations. Experiments on the chronic data collected from the Hong Kong Chronic Disease Study Project and a public diagnostic data MIMIC-III demonstrate that DSSDDI can be a reliable reference for doctors in terms of safety and efficiency of clinical diagnosis, with significant improvements compared to baseline methods.

Guillaume Fahrni, Guillaume Fahrni, Guillaume Fahrni et al.

IntroductionTo evaluate a reduced iodine volume protocol for lower extremity CT angiography (CTA) using dual-energy CT (DECT).MethodsThis retrospective study included consecutive patients who underwent lower extremity CTA from June to December 2022. A 10 ml 1:1 mixed test bolus was performed, followed by a 40 ml full bolus at a 2.5/s injection rate, using 400 mg/ml iodine contrast media. Conventional and 40 keV virtual monoenergetic images (VMI) were reconstructed. For both reconstructions, five main artery segments were assessed with a 3-point image quality score as well as quantitative attenuation, signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) measurements with diagnostic quality thresholds (respectively >150 HU and >3).ResultsForty patients were included in the study (mean age 68 ± 12 yo). 200 artery segments were assessed. Median qualitative image scores were 3 [IQR, 3, 3] for both reconstructions. 40 keV VMI upgraded qualitative scores for 51 (26%) of patients, including 9 (5%) from nondiagnostic to diagnostic quality. 40 keV VMI obtained attenuation and CNR diagnostic quality for respectively 100% and 100% of segments, compared with 96% and 98% for conventional images (p < 0.001). Distal artery segments showed the most differences between 40 keV VMI and conventional images.ConclusionA low-iodine lower extremity CTA protocol is feasible, with 40 keV virtual monoenergetic spectral reconstruction enabling maintained diagnostic image quality at the distal artery segments.

Ana Beatrice Magalhães deOliveira, Beatriz Luchiari, Isabella Bonilha et al.

Bailey M. DeBarmore, Jessica K. Zègre-Hemsey, Anna M. Kucharska-Newton et al.

Background: Our objective was to describe characteristics of patients presenting with and without ischemic pain among those diagnosed with acute myocardial infarction (MI) using individual-level data from the Atherosclerosis Risk in Communities Study from 2005 to 2019. Methods: Acute MI included events deemed definite or probable MI by a physician panel based on ischemic pain, cardiac biomarkers, and ECG evidence. Patient characteristics included age at hospitalization, sex, race/ethnicity, comorbidities (smoking status, diabetes, hypertension, history of previous stroke, MI, or cardiovascular procedure, and history of valvular disease or cardiomyopathy) and in-hospital complications occurring during the event of interest (pulmonary edema, pulmonary embolism, in-hospital stroke, pneumonia, cardiogenic shock, ventricular fibrillation). Analyses were stratified by MI subtype (STEMI, NSTEMI, Unclassified) and patient characteristics and 28-day case fatality was compared between MI presenting with or without ischemic pain. Results: Between 2005 and 2019, there were 1711 hospitalized definite/probable MI events (47 % female, 26 % black, and age of 78 [6.7 years]). A smaller proportion of STEMI patients presented without ischemic pain compared to NSTEMI patients (20 % vs 32 %). Race, sex, age, and comorbidity profiles did not differ significantly across ischemic pain presentations. Patients presenting without ischemic pain had a higher 28-day all-cause case fatality after adjusting for age, race, sex, and comorbidities. However, after further adjustment, time from symptom onset to hospital arrival, time to treatment, and in-hospital complications explained the difference in 28-day case fatality between ischemic pain presentations. Conclusions: Future research should focus on differences in treatment delay across ischemic pain presentations rather than sex differences in acute coronary syndrome presentation.

Hisanori Kosuge, Shoko Hachiya, Yasuhiro Fujita et al.

Abstract Background Ischemia of the hypertrophied myocardium due to microvascular dysfunction is related to a worse prognosis in hypertrophic cardiomyopathy (HCM). Stress and rest T1 mapping without contrast agents can be used to assess myocardial blood flow. Herein, we evaluated the potential of non-contrast stress T1 mapping in assessing myocardial injury in patients with HCM. Methods Forty-five consecutive subjects (31 HCM patients and 14 control subjects) underwent cardiac magnetic resonance (CMR) at 3T, including cine imaging, T1 mapping at rest and during adenosine triphosphate (ATP) stress, late gadolinium enhancement (LGE), and phase-contrast (PC) cine imaging of coronary sinus flow at rest and during stress to assess coronary flow reserve (CFR). PC cine imaging was performed on 25 subjects (17 patients with HCM and 8 control subjects). Native T1 values at rest and during stress were measured using the 16-segment model, and T1 reactivity was defined as the change in T1 values from rest to stress. Results ATP stress induced a significant increase in native T1 values in both the HCM and control groups (HCM: p < 0.001, control: p = 0.002). T1 reactivity in the HCM group was significantly lower than that in the control group (4.2 ± 0.3% vs. 5.6 ± 0.5%, p = 0.044). On univariate analysis, T1 reactivity correlated with native T1 values at rest, left ventricular mass index, and CFR. Multiple linear regression analysis demonstrated that only CFR was independently correlated with T1 reactivity (β = 0.449; 95% confidence interval, 0.048–0.932; p = 0.032). Furthermore, segmental analysis showed decreased T1 reactivity in the hypertrophied myocardium and the non-hypertrophied myocardium with LGE in the HCM group. Conclusions T1 reactivity was lower in the hypertrophied myocardium and LGE-positive myocardium compared to non-injured myocardium. Non-contrast stress T1 mapping is a promising CMR method for assessing myocardial injury in patients with HCM. Trial registration Retrospectively registered.

Warren Clements, Mark Fitzgerald, S. Murthy Chennapragada et al.

Abstract Background Splenic artery embolisation (SAE) has become a vital strategy in the modern landscape of multidisciplinary trauma care, improving splenic salvage rates in patients with high-grade injury. However, due to a lack of prospective data there remains contention amongst stakeholders as to whether SAE should be performed at the time of presentation (prophylactic or pSAE), or whether patients should be observed, and SAE only used only if a patient re-bleeds. This systematic review aimed to assess published practice management guidelines which recommend pSAE, stratified according to their quality. Methods The study was registered and reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. Medline, PubMed, Cochrane, Embase, and Google Scholar were searched by the study authors. Identified guidelines were graded according to the Appraisal of Guidelines Research and Evaluation II (AGREE-II) instrument. Results Database and internet searches identified 1006 results. After applying exclusion criteria, 28 guidelines were included. The use of pSAE was recommended in 15 guidelines (54%). This included 6 out of 9 guidelines that were high quality (66.7%), 4 out of 9 guidelines that were moderate quality (44.4%), and 3 out of 10 (30%) guidelines that were low quality, p = 0.275. Conclusions This systematic review showed that recommendation of pSAE is more common in guidelines which are of high quality. However, there is vast heterogeneity of recommended practice guidelines, likely based on individual trauma systems rather than the available evidence. This reflects biases with interpretation of data and lack of multidisciplinary system inputs, including from interventional radiologists.

Jiechun Zhu, Guangdong Yang

M. A. Rodriguez, H. AlMarzouqi, P. Liatsis

Early detection of retinal diseases is one of the most important means of preventing partial or permanent blindness in patients. In this research, a novel multi-label classification system is proposed for the detection of multiple retinal diseases, using fundus images collected from a variety of sources. First, a new multi-label retinal disease dataset, the MuReD dataset, is constructed, using a number of publicly available datasets for fundus disease classification. Next, a sequence of post-processing steps is applied to ensure the quality of the image data and the range of diseases, present in the dataset. For the first time in fundus multi-label disease classification, a transformer-based model optimized through extensive experimentation is used for image analysis and decision making. Numerous experiments are performed to optimize the configuration of the proposed system. It is shown that the approach performs better than state-of-the-art works on the same task by 7.9% and 8.1% in terms of AUC score for disease detection and disease classification, respectively. The obtained results further support the potential applications of transformer-based architectures in the medical imaging field.

Javad Hassannataj Joloudari, Hamid Saadatfar, Mohammad GhasemiGol et al.

Cardiovascular disease is one of the most challenging diseases in middle-aged and older people, which causes high mortality. Coronary artery disease (CAD) is known as a common cardiovascular disease. A standard clinical tool for diagnosing CAD is angiography. The main challenges are dangerous side effects and high angiography costs. Today, the development of artificial intelligence-based methods is a valuable achievement for diagnosing disease. Hence, in this paper, artificial intelligence methods such as neural network (NN), deep neural network (DNN), and Fuzzy C-Means clustering combined with deep neural network (FCM-DNN) are developed for diagnosing CAD on a cardiac magnetic resonance imaging (CMRI) dataset. The original dataset is used in two different approaches. First, the labeled dataset is applied to the NN and DNN to create the NN and DNN models. Second, the labels are removed, and the unlabeled dataset is clustered via the FCM method, and then, the clustered dataset is fed to the DNN to create the FCM-DNN model. By utilizing the second clustering and modeling, the training process is improved, and consequently, the accuracy is increased. As a result, the proposed FCM-DNN model achieves the best performance with a 99.91% accuracy specifying 10 clusters, i.e., 5 clusters for healthy subjects and 5 clusters for sick subjects, through the 10-fold cross-validation technique compared to the NN and DNN models reaching the accuracies of 92.18% and 99.63%, respectively. To the best of our knowledge, no study has been conducted for CAD diagnosis on the CMRI dataset using artificial intelligence methods. The results confirm that the proposed FCM-DNN model can be helpful for scientific and research centers.

Mirko Hu, Guido Caldarelli, Tommaso Gili

Inflammatory bowel diseases (IBD) are complex diseases in which the gut microbiota is attacked by the immune system of genetically predisposed subjects when they are exposed to yet unclear environmental factors. The complexity of this class of diseases makes them suitable to be represented and studied with network science. In the project, the metagenomic data of the gut microbiota of control, Crohn's disease, and ulcerative colitis subjects were divided in three ranges (prevalent, common, uncommon). Then, correlation networks and co-expression networks were used to represent this data. The former networks involved the calculation of the Pearson's correlation and the use of the percolation threshold to binarize the adjacency matrix, whereas the latter involved the construction of the bipartite networks and the monopartite projection after binarization of the biadjacency matrix. Then, centrality measures and community detection were used on the so-built networks. The main results obtained were about the modules of "Bacteroides", which were connected in control subjects' correlation network, "Faecalibacterium prausnitzii", where co-enzyme A became central in IBD correlation networks and "Escherichia coli", which module has different position in the different diagnoses networks.

Ting Wang, Gui-Yun Li, Xin-Hui Li et al.

The main aim to build models capable of simulating the spreading of infectious diseases is to control them. And along this way, the key to find the optimal strategy for disease control is to obtain a large number of simulations of disease transitions under different scenarios. Therefore, the models that can simulate the spreading of diseases under scenarios closer to the reality and are with high efficiency are preferred. In the realistic social networks, the random contact, including contacts between people in the public places and the public transits, becomes the important access for the spreading of infectious diseases. In this paper, a model can efficiently simulate the spreading of infectious diseases under random contacts is proposed. In this approach, the random contact between people is characterized by the random matrix with elements randomly generated and the spread of the diseases is simulated by the Markov process. We report an interesting property of the proposed model: the main indicators of the spreading of the diseases such as the death rate are invariant of the size of the population. Therefore, representative simulations can be conducted on models consist of small number of populations. The main advantage of this model is that it can easily simulate the spreading of diseases under more realistic scenarios and thus is able to give a large number of simulations needed for the searching of the optimal control strategy. Based on this work, the reinforcement learning will be introduced to give the optimal control strategy in the following work.

Kush Vora, Dishant Padalia

Apple diseases, if not diagnosed early, can lead to massive resource loss and pose a serious threat to humans and animals who consume the infected apples. Hence, it is critical to diagnose these diseases early in order to manage plant health and minimize the risks associated with them. However, the conventional approach of monitoring plant diseases entails manual scouting and analyzing the features, texture, color, and shape of the plant leaves, resulting in delayed diagnosis and misjudgments. Our work proposes an ensembled system of Xception, InceptionResNet, and MobileNet architectures to detect 5 different types of apple plant diseases. The model has been trained on the publicly available Plant Pathology 2021 dataset and can classify multiple diseases in a given plant leaf. The system has achieved outstanding results in multi-class and multi-label classification and can be used in a real-time setting to monitor large apple plantations to aid the farmers manage their yields effectively.