-
Medical Image Analysis Jun 2024The conventional pretraining-and-finetuning paradigm, while effective for common diseases with ample data, faces challenges in diagnosing data-scarce occupational...
The conventional pretraining-and-finetuning paradigm, while effective for common diseases with ample data, faces challenges in diagnosing data-scarce occupational diseases like pneumoconiosis. Recently, large language models (LLMs) have exhibits unprecedented ability when conducting multiple tasks in dialogue, bringing opportunities to diagnosis. A common strategy might involve using adapter layers for vision-language alignment and diagnosis in a dialogic manner. Yet, this approach often requires optimization of extensive learnable parameters in the text branch and the dialogue head, potentially diminishing the LLMs' efficacy, especially with limited training data. In our work, we innovate by eliminating the text branch and substituting the dialogue head with a classification head. This approach presents a more effective method for harnessing LLMs in diagnosis with fewer learnable parameters. Furthermore, to balance the retention of detailed image information with progression towards accurate diagnosis, we introduce the contextual multi-token engine. This engine is specialized in adaptively generating diagnostic tokens. Additionally, we propose the information emitter module, which unidirectionally emits information from image tokens to diagnosis tokens. Comprehensive experiments validate the superiority of our methods.
PubMed: 38941859
DOI: 10.1016/j.media.2024.103248 -
Occupational and Environmental Medicine Jun 2024Previous radiologic and histopathologic studies suggest respirable crystalline silica (RCS) overexposure has been driving the resurgence of pneumoconiosis among...
OBJECTIVES
Previous radiologic and histopathologic studies suggest respirable crystalline silica (RCS) overexposure has been driving the resurgence of pneumoconiosis among contemporary US coal miners, with a higher prevalence of severe disease in Central Appalachia. We sought to better understand RCS exposure among US underground coal miners.
METHODS
We analysed RCS levels, as measured by respirable quartz, from coal mine dust compliance data from 1982 to 2021.
RESULTS
We analysed 322 919 respirable quartz samples from 5064 US underground coal mines. Mean mine-level respirable quartz percentage and mass concentrations were consistently higher for Central Appalachian mines than the rest of the USA. Mean mine-level respirable quartz mass concentrations decreased significantly over time, from 0.116 mg/m in 1982 to as low as 0.017 mg/m for Central Appalachian mines, and from 0.089 mg/m in 1983 to 0.015 mg/m in 2020 for the rest of the USA. Smaller mine size, location in Central Appalachia, lack of mine safety committee and thinner coal seams were predictive of higher respirable quartz mass concentrations.
CONCLUSIONS
These data substantially support the association between RCS overexposure and the resurgence of coal workers' pneumoconiosis in the USA, particularly in smaller mines in Central Appalachia.
PubMed: 38937079
DOI: 10.1136/oemed-2023-109347 -
International Journal of Environmental... May 2024Engineered stone (ES) is a popular building product, due to its architectural versatility and generally lower cost. However, the fabrication of organic resin-based ES... (Review)
Review
Engineered stone (ES) is a popular building product, due to its architectural versatility and generally lower cost. However, the fabrication of organic resin-based ES kitchen benchtops from slabs has been associated with alarming rates of silicosis among workers. In 2024, fifteen years after the first reported ES-related cases in the world, Australia became the first country to ban the use and importation of ES. A range of interacting factors are relevant for ES-associated silicosis, including ES material composition, characteristics of dust exposure and lung cell-particle response. In turn, these are influenced by consumer demand, work practices, particle size and chemistry, dust control measures, industry regulation and worker-related characteristics. This literature review provides an evidence synthesis using a narrative approach, with the themes of product, exposure and host. Exposure pathways and pathogenesis are explored. Apart from crystalline silica content, consideration is given to non-siliceous ES components such as resins and metals that may modify chemical interactions and disease risk. Preventive effort can be aligned with each theme and associated evidence.
Topics: Silicosis; Humans; Occupational Exposure; Construction Materials; Dust
PubMed: 38928930
DOI: 10.3390/ijerph21060683 -
Advanced Materials (Deerfield Beach,... Jun 2024The compact three-dimensional structure of extracellular polymeric substances (EPS) within biofilms significantly hinders the penetration of antimicrobial agents, making...
The compact three-dimensional structure of extracellular polymeric substances (EPS) within biofilms significantly hinders the penetration of antimicrobial agents, making biofilm eradication challenging and resulting in persistent biofilm-associated infections. To address this challenge, we propose a solution: a probiotic bio-heterojunction (P-bioHJ) combining Lactobacillus rhamnosus with MXene (TiC) quantum dots (MQDs)/FeS heterojunction. This innovation aims to break down the saccharides in EPS, enabling effective combat against biofilm-associated infections. Initially, the P-bioHJ targets saccharides through metabolic processes, causing the collapse of EPS and allowing infiltration into bacterial colonies. Simultaneously, upon exposure to near-infrared irradiation, the P-bioHJ produces reactive oxygen species (ROS) and thermal energy, deploying physical mechanisms to combat bacterial biofilms effectively. Following antibiofilm treatment, the P-bioHJ adjusts the oxidative environment, reduces wound inflammation by scavenging ROS, boosts antioxidant enzyme activity, and mitigates the NF-κB inflammatory pathway, thereby accelerating wound healing. In vitro and in vivo experiments confirm the exceptional antibiofilm, antioxidant/anti-inflammatory, and wound-regeneration properties of P-bioHJ. In conclusion, this study provides a promising approach for treating biofilm-related infections. This article is protected by copyright. All rights reserved.
PubMed: 38924628
DOI: 10.1002/adma.202402530 -
Artificial Intelligence in Medicine Jun 2024Early detection of pneumoconiosis by routine health screening of workers in the mining industry is critical for preventing the progression of this incurable disease....
Early detection of pneumoconiosis by routine health screening of workers in the mining industry is critical for preventing the progression of this incurable disease. Automated pneumoconiosis classification in chest X-ray images is challenging due to the low contrast of opacities, inter-class similarity, intra-class variation and the existence of artifacts. Compared to traditional methods, convolutional neural networks have shown significant improvement in pneumoconiosis classification tasks, however, accurate classification remains challenging due to mainly the inability to focus on semantically meaningful lesion opacities. Most existing networks focus on high level abstract information and ignore low level detailed object information. Different from natural images where an object occupies large space, the classification of pneumoconiosis depends on the density of small opacities inside the lung. To address this issue, we propose a novel two-stage adaptive multi-scale feature pyramid network called AMFP-Net for the diagnosis of pneumoconiosis from chest X-rays. The proposed model consists of 1) an adaptive multi-scale context block to extract rich contextual and discriminative information and 2) a weighted feature fusion module to effectively combine low level detailed and high level global semantic information. This two-stage network first segments the lungs to focus more on relevant regions by excluding irrelevant parts of the image, and then utilises the segmented lungs to classify pneumoconiosis into different categories. Extensive experiments on public and private datasets demonstrate that the proposed approach can outperform state-of-the-art methods for both segmentation and classification.
PubMed: 38917599
DOI: 10.1016/j.artmed.2024.102917 -
BMJ Case Reports Jun 2024This case series sheds light on the pulmonary diseases afflicting artisanal gold miners in Chunya district, Mbeya, Tanzania. We present 3 cases from a group of 21...
This case series sheds light on the pulmonary diseases afflicting artisanal gold miners in Chunya district, Mbeya, Tanzania. We present 3 cases from a group of 21 miners. The patients, ranging in age and mining exposure, exhibited symptoms of severe pulmonary conditions, including pneumoconiosis, pulmonary hypertension and Cor pulmonale, attributed to prolonged exposure to dust and inadequate protective measures in mining environments. These cases underscore the urgent need for enhanced occupational health standards and preventive strategies in artisanal mining communities.
Topics: Humans; Tanzania; Male; Pneumoconiosis; Mining; Middle Aged; Adult; Occupational Exposure; Miners; Hypertension, Pulmonary
PubMed: 38914531
DOI: 10.1136/bcr-2024-260481 -
The American Journal of the Medical... Jun 2024Previous studies have shown that hyponatremia was strongly associated with a poor prognosis of type 1 pulmonary hypertension, and our team's antecedent studies found...
BACKGROUND
Previous studies have shown that hyponatremia was strongly associated with a poor prognosis of type 1 pulmonary hypertension, and our team's antecedent studies found that low serum sodium was associated with the severity and the length of hospitalization of pulmonary hypertension associated with left ventricular disease (PH-LHD). However, the relationship between serum sodium and the prognosis of PH-LHD remains unclear. This study aims to determine the clinical value of serum sodium in evaluating poor prognosis in patients with PH-LHD.
METHODS
We successfully followed 716 patients with PH-LHD. Kaplan-Meier was used to plot survival in PH-LHD patients with different serum sodium levels. The effect of serum sodium on poor prognosis was analyzed using a Cox proportional risk model. The trends between patients serum sodium and survival were visualized by restricted cubic spline (RCS).
RESULTS
The survival rates at 1, 2, 3 and 4 years were 52%, 41%, 31% and 31% for the patients with hyponatremia associated with PH-LHD and 71%, 71%, 71% and 54% for the patients with hypernatremia, respectively. The observed mortality rate in the hyponatremia and hypernatremia groups surpassed that of the normonatremic group. The adjusted risks of death (risk ratio) for patients with hyponatremia and hypernatremia were found to be 2.044 and 1.877. Furthermore, the restricted cubic spline demonstrated an L-shaped correlation between serum sodium and all-cause mortality in patients with PH-LHD.
CONCLUSIONS
Abnormal serum sodium level is strongly associated with poor prognosis in PH-LHD. Serum sodium may play an important pathogenic role in PH-LHD occurrence and could be used as a marker to assess the survival in patients.
PubMed: 38909900
DOI: 10.1016/j.amjms.2024.06.016 -
Medicine Jun 2024The diagnosis of pneumoconiosis is complex and subjective, leading to inevitable variability in readings. This is especially true for inexperienced doctors. To improve... (Observational Study)
Observational Study
The diagnosis of pneumoconiosis is complex and subjective, leading to inevitable variability in readings. This is especially true for inexperienced doctors. To improve accuracy, a computer-assisted diagnosis system is used for more effective pneumoconiosis diagnoses. Three models (Resnet50, Resnet101, and DenseNet) were used for pneumoconiosis classification based on 1250 chest X-ray images. Three experienced and highly qualified physicians read the collected digital radiography images and classified them from category 0 to category III in a double-blinded manner. The results of the 3 physicians in agreement were considered the relative gold standards. Subsequently, 3 models were used to train and test these images and their performance was evaluated using multi-class classification metrics. We used kappa values and accuracy to evaluate the consistency and reliability of the optimal model with clinical typing. The results showed that ResNet101 was the optimal model among the 3 convolutional neural networks. The AUC of ResNet101 was 1.0, 0.9, 0.89, and 0.94 for detecting pneumoconiosis categories 0, I, II, and III, respectively. The micro-average and macro-average mean AUC values were 0.93 and 0.94, respectively. The accuracy and Kappa values of ResNet101 were 0.72 and 0.7111 for quadruple classification and 0.98 and 0.955 for dichotomous classification, respectively, compared with the relative standard classification of the clinic. This study develops a deep learning based model for screening and staging of pneumoconiosis is using chest radiographs. The ResNet101 model performed relatively better in classifying pneumoconiosis than radiologists. The dichotomous classification displayed outstanding performance, thereby indicating the feasibility of deep learning techniques in pneumoconiosis screening.
Topics: Humans; Pneumoconiosis; Deep Learning; Radiography, Thoracic; Male; Middle Aged; Reproducibility of Results; Female; Diagnosis, Computer-Assisted; Aged; Neural Networks, Computer
PubMed: 38905434
DOI: 10.1097/MD.0000000000038478 -
The European Respiratory Journal Jun 2024
Topics: Humans; Australia; Silicon Dioxide; Occupational Exposure; Silicosis
PubMed: 38901887
DOI: 10.1183/13993003.00138-2024 -
International Journal of Molecular... May 2024Silicosis caused by engineered stone (ES-silicosis) is an emerging worldwide issue characterized by inflammation and fibrosis in the lungs. To our knowledge, only a few...
Silicosis caused by engineered stone (ES-silicosis) is an emerging worldwide issue characterized by inflammation and fibrosis in the lungs. To our knowledge, only a few reports have investigated leukocyte/lymphocyte subsets in ES-silicosis patients. The present study was designed to explore the proportions of the main lymphocyte subsets in ES-silicosis patients stratified into two groups, one with simple silicosis (SS) and the other with a more advanced state of the disease, defined as progressive massive fibrosis (PMF). The proportions of B (memory and plasmablasts) cells, T (helper, cytotoxic, regulatory) cells, and natural killer (NK) (regulatory and cytotoxic) cells were investigated by multiparameter flow cytometry in 91 ES-silicosis patients (53 SS patients and 38 PMF patients) and 22 healthy controls (HC). Although the total number of leukocytes did not differ between the groups studied, lymphopenia was observed in patients compared to healthy controls. Compared with those in healthy controls, the proportions of memory B cells, naïve helper T cells, and the CD4/CD8 T cells' ratio in the peripheral blood of patients with silicosis were significantly decreased, while the percentages of plasma cells, memory helper T cells, and regulatory T cells were significantly increased. For the NK cell subsets, no significant differences were found between the groups studied. These results revealed altered cellular immune processes in the peripheral blood of patients with ES-silicosis and provided further insight into silicosis pathogenesis.
Topics: Humans; Male; Silicosis; Middle Aged; Female; Silicon Dioxide; Adult; Killer Cells, Natural; Lymphocyte Subsets; Pneumonia; Aged; Case-Control Studies
PubMed: 38891910
DOI: 10.3390/ijms25115722