RESUMO
Chronic obstructive pulmonary disease (COPD) stands as a significant global health challenge, with its intricate pathophysiological manifestations often demanding advanced diagnostic strategies. The recent applications of artificial intelligence (AI) within the realm of medical imaging, especially in computed tomography, present a promising avenue for transformative changes in COPD diagnosis and management. This review delves deep into the capabilities and advancements of AI, particularly focusing on machine learning and deep learning, and their applications in COPD identification, staging, and imaging phenotypes. Emphasis is laid on the AI-powered insights into emphysema, airway dynamics, and vascular structures. The challenges linked with data intricacies and the integration of AI in the clinical landscape are discussed. Lastly, the review casts a forward-looking perspective, highlighting emerging innovations in AI for COPD imaging and the potential of interdisciplinary collaborations, hinting at a future where AI doesn't just support but pioneers breakthroughs in COPD care. Through this review, we aim to provide a comprehensive understanding of the current state and future potential of AI in shaping the landscape of COPD diagnosis and management.
Assuntos
Inteligência Artificial , Doença Pulmonar Obstrutiva Crônica , Tomografia Computadorizada por Raios X , Humanos , Doença Pulmonar Obstrutiva Crônica/diagnóstico por imagem , Inteligência Artificial/tendências , Tomografia Computadorizada por Raios X/métodos , Índice de Gravidade de DoençaRESUMO
Introductionï¼Exacerbations of chronic obstructive pulmonary disease (COPD) have a significant impact on hospitalizations, morbidity, and mortality of patients. This study aimed to develop a model for predicting acute exacerbation in COPD patient (AECOPD) based on deep learning (DL) features. METHODS: We performed a retrospective study on 219 patients with COPD who underwent inspiratory and expiratory HRCT scans. By recording the acute respiratory events of the previous year, these patients were further divided into Non-AECOPD group and AECOPD group according to the presence of acute exacerbation events. 69 Quantitative CT (QCT) parameters of emphysema and airway were calculated by NeuLungCARE software and 2000 DL features were extracted by VGG-16 method. The Logistic regression method was employed to identify AECOPD patients and 29 patients of external validation cohort were used to access the robustness of the results. RESULTS: The Model 3-B achieved an AUC of 0.933, and 0.865 in the testing cohort and external validation cohort respectively. Model 3-I obtained AUC of 0.895 in the testing cohort and AUC of 0.774 in the external validation cohort. Model 7-B combined clinical characteristics, QCT parameters, and DL features achieved the best performance with an AUC of 0.979 in the testing cohort and demonstrating robust predictability with an AUC of 0.932 in the external validation cohort. Likewise, Model 7-I achieving an AUC of 0.938, and 0.872 in the testing cohort and external validation cohort respectively. CONCLUSIONS: DL features extracted from HRCT scans can effectively predict acute exacerbation phenotype in COPD patients.
RESUMO
BACKGROUND: Chronic obstructive pulmonary disease (COPD) is a prevalent and debilitating respiratory condition that imposes a significant healthcare burden worldwide. Accurate staging of COPD severity is crucial for patient management and treatment planning. METHODS: The retrospective study included 530 hospital patients. A lobe-based radiomics method was proposed to classify COPD severity using computed tomography (CT) images. First, we segmented the lung lobes with a convolutional neural network model. Secondly, the radiomic features of each lung lobe are extracted from CT images, the features of the five lung lobes are merged, and the selection of features is accomplished through the utilization of a variance threshold, t-Test, least absolute shrinkage and selection operator (LASSO). Finally, the COPD severity was classified by a support vector machine (SVM) classifier. RESULTS: 104 features were selected for staging COPD according to the Global initiative for chronic Obstructive Lung Disease (GOLD). The SVM classifier showed remarkable performance with an accuracy of 0.63. Moreover, an additional set of 132 features were selected to distinguish between milder (GOLD I + GOLD II) and more severe instances (GOLD III + GOLD IV) of COPD. The accuracy for SVM stood at 0.87. CONCLUSIONS: The proposed method proved that the novel lobe-based radiomics method can significantly contribute to the refinement of COPD severity staging. By combining radiomic features from each lung lobe, it can obtain a more comprehensive and rich set of features and better capture the CT radiomic features of the lung than simply observing the lung as a whole.
Assuntos
Doença Pulmonar Obstrutiva Crônica , Índice de Gravidade de Doença , Máquina de Vetores de Suporte , Tomografia Computadorizada por Raios X , Humanos , Doença Pulmonar Obstrutiva Crônica/diagnóstico por imagem , Doença Pulmonar Obstrutiva Crônica/classificação , Tomografia Computadorizada por Raios X/métodos , Estudos Retrospectivos , Masculino , Feminino , Pessoa de Meia-Idade , Idoso , Pulmão/diagnóstico por imagem , Pulmão/patologia , Redes Neurais de Computação , RadiômicaRESUMO
Biosafety laboratories are critical in many fields. However, experimenters associated the infection risk from biological aerosols. In this study, by conducting experiments on the release and collection of bioaerosols within a typical BSL-2 + laboratory, the spatial distribution of bioaerosols was tracked. Numerical calculations were employed to obtain and visualize the airflow patterns and aerosol dispersion paths of four ventilation methods. The results indicated that equipment and tables led to uneven airflow distribution within the laboratory. The comparison results of the four evaluation indicators showed that the air age distribution of UU (Upward supply and upward return) mode and CD (Cross-supply and downward return) mode was superior, with air change efficiency values of 0.595 and 0.603, respectively. Additionally, the contaminant removal index of CD mode was 1.48, significantly higher than the other ventilation methods. The statistical results of the contaminant dispersion index also indicated that CD mode was most conducive to diluting aerosols in the spatial environment. The LD (lateral supply and downward return) mode may lead to airflow short-circuiting. The UD (upward supply and downward return) mode can provide balanced protection for laboratory. Overall, CD mode performed the best among the four ventilation methods, followed by UU mode.
Assuntos
Aerossóis , Contenção de Riscos Biológicos , Laboratórios , Ventilação , Aerossóis/análise , Contenção de Riscos Biológicos/métodosRESUMO
Little is known about the relationships between human genetics and the airway microbiome. Deeply sequenced airway metagenomics, by simultaneously characterizing the microbiome and host genetics, provide a unique opportunity to assess the microbiome-host genetic associations. Here we performed a co-profiling of microbiome and host genetics with the identification of over 5 million single nucleotide polymorphisms (SNPs) through deep metagenomic sequencing in sputum of 99 chronic obstructive pulmonary disease (COPD) and 36 healthy individuals. Host genetic variation was the most significant factor associated with the microbiome except for geography and disease status, with its top 5 principal components accounting for 12.11% of the microbiome variability. Within COPD individuals, 113 SNPs mapped to candidate genes reported as genetically associated with COPD exhibited associations with 29 microbial species and 48 functional modules (P < 1 × 10-5), where Streptococcus salivarius exhibits the strongest association to SNP rs6917641 in TBC1D32 (P = 9.54 × 10-8). Integration of concurrent host transcriptomic data identified correlations between the expression of host genes and their genetically-linked microbiome features, including NUDT1, MAD1L1 and Veillonella parvula, TTLL9 and Stenotrophomonas maltophilia, and LTA4H and Haemophilus influenzae. Mendelian randomization analyses revealed a potential causal link between PARK7 expression and microbial type III secretion system, and a genetically-mediated association between COPD and increased relative abundance of airway Streptococcus intermedius. These results suggest a previously underappreciated role of host genetics in shaping the airway microbiome and provide fresh hypotheses for genetic-based host-microbiome interactions in COPD.
Assuntos
Microbiota , Doença Pulmonar Obstrutiva Crônica , Humanos , Doença Pulmonar Obstrutiva Crônica/diagnóstico , Doença Pulmonar Obstrutiva Crônica/genética , Doença Pulmonar Obstrutiva Crônica/complicações , Microbiota/genética , Escarro , Transcriptoma , Genética Humana , Proteínas Adaptadoras de Transdução de Sinal/genéticaRESUMO
Background: GST-HG171 is a potent, broad-spectrum, orally bioavailable small-molecule 3C like protease inhibitor that has demonstrated greater potency and efficacy compared to Nirmatrelvir in pre-clinical studies. We aimed to evaluate the efficacy and safety of orally administered GST-HG171 plus Ritonavir in patients with coronavirus disease 2019 (COVID-19) infected with emerging XBB and non-XBB variants. Methods: This randomised, double-blind, placebo-controlled phase 2/3 trial was conducted in 47 sites in China among adult patients with mild-to-moderate COVID-19 with symptoms onset ≤72 h. Eligible patients were randomised 1:1 to receive GST-HG171 (150 mg) plus Ritonavir (100 mg) or corresponding placebo tablets twice daily for 5 days, with stratification factors including the risk level of disease progression and vaccination status. The primary efficacy endpoint was time to sustained recovery of clinical symptoms within 28 days, defined as a score of 0 for 11 COVID-19-related target symptoms for 2 consecutive days, assessed in the modified intention-to-treat (mITT) population. This trial was registered at ClinicalTrials.gov (NCT05656443) and Chinese Clinical Trial Registry (ChiCTR2200067088). Findings: Between Dec 19, 2022, and May 4, 2023, 1525 patients were screened. Among 1246 patients who underwent randomisation, most completed basic (21.2%) or booster (74.9%) COVID-19 immunization, and most had a low risk of disease progression at baseline. 610 of 617 who received GST-HG171 plus Ritonavir and 603 of 610 who received placebo were included in the mITT population. Patients who received GST-HG171 plus Ritonavir showed shortened median time to sustained recovery of clinical symptoms compared to the placebo group (13.0 days [95.45% confidence interval 12.0-15.0] vs. 15.0 days [14.0-15.0], P = 0.031). Consistent results were observed in both SARS-CoV-2 XBB (45.7%, 481/1053 of mITT population) and non-XBB variants (54.3%, 572/1053 of mITT population) subgroups. Incidence of adverse events was similar in the GST-HG171 plus Ritonavir (320/617, 51.9%) and placebo group (298/610, 48.9%). The most common adverse events in both placebo and treatment groups were hypertriglyceridaemia (10.0% vs. 14.7%). No deaths occurred. Interpretation: Treatment with GST-HG171 plus Ritonavir has demonstrated benefits in symptom recovery and viral clearance among low-risk vaccinated adult patients with COVID-19, without apparent safety concerns. As most patients were treated within 2 days after symptom onset in our study, confirming the potential benefits of symptom recovery for patients with a longer duration between symptom onset and treatment initiation will require real-world studies. Funding: Fujian Akeylink Biotechnology Co., Ltd.
RESUMO
BACKGROUND: The prevalence of type 2 diabetes (T2DM) in China is over 10%, affecting around 114 million people. Despite the inclusion of T2DM in the National Basic Public Health Service Program (NBPHSP), most people with T2DM experience challenges in achieving optimal management targets. This study aimed to identify barriers and facilitators of diabetes management from the perspectives of primary health care (PHC) service providers and recipients. METHODS: This mixed-methods study was conducted in Shijiazhuang City, Hebei Province, China. A quantitative PHC facility assessment survey was conducted in all administrative districts and qualitative in-depth interviews were conducted in one district to government officials, medical staff, patients with T2DM, and their family members. Interviews were thematically analyzed, and all findings were synthesized using Michie's COM-B theory. RESULTS: A total of 197 village/community level PHC facilities and 66 township/street level PHC facilities answered the survey, and 42 in-depth interviews were conducted. The key facilitators stemmed from the NBPHSP policy, which standardized the basic infrastructure, medical equipment, and medication for the PHC facilities, provided training on NCD prevention and control, and compensated the PHC workers. However, we identified a detrimental cycle among PHC providers characterized by inadequate capacity, overwhelming workloads, insufficient income, limited career development opportunities, and challenges in attracting young talents. Although patients were covered by the national medical insurance schemes, they experienced capability constraints primarily driven by low education levels, advanced age, low health literacy, and a proliferation of misinformation. These factors influenced patients' motivation to be actively engaged in care and contributed to inertia to intensify treatment and achieve their clinical management goals. CONCLUSION: This study identifies several major facilitators and barriers from the perspectives of both PHC providers and patients with T2DM. Our findings suggest there are substantial opportunities to strengthen the NBPHSP, including improving the capacity and the income level of the PHC providers, attracting and retaining skilled health workers in rural areas, supporting patients to improve their health literacy and take a more active role in their health care, and improving access to high-quality care through digital health approaches. TRIAL REGISTRATION: ClinicalTrials.gov (record NCT02726100, 03/22/2016).
Assuntos
Diabetes Mellitus Tipo 2 , Atenção Primária à Saúde , Humanos , Instituições de Assistência Ambulatorial , Atenção à Saúde , Diabetes Mellitus Tipo 2/epidemiologia , Diabetes Mellitus Tipo 2/terapia , Pesquisa QualitativaRESUMO
Ochratoxin A (OTA), a common mycotoxin, can contaminate food and feed and is difficult to remove. Astaxanthin (ASTA), a natural antioxidant, can effectively protect against OTA-induced hepatotoxicity; however, its mechanism of action remains unclear. In the present study, we elucidate the protective effects of ASTA on the OTA-induced damage of the endoplasmic reticulum and mitochondria in broiler liver samples by serum biochemical analysis, antioxidant analysis, qRT-PCR, and Western blot analysis. ASTA inhibited the expressions of ahr, pxr, car, cyp1a1, cyp1a5, cyp2c18, cyp2d6, and cyp3a9 genes, and significantly alleviated OTA-induced liver oxidative damage (SOD, GSH-Px, GSH, MDA). Furthermore, it inhibited OTA-activated endoplasmic reticulum stress genes and proteins (grp94, GRP78, atf4, ATF6, perk, eif2α, ire1, CHOP). ASTA alleviated OTA-induced mitochondrial dynamic imbalance, inhibited mitochondrial division (DRP1, mff), and promoted mitochondrial fusion (OPA1, MFN1, MFN2). In conclusion, ASTA can decrease OTA-induced oxidative damage, thereby alleviating endoplasmic reticulum stress and mitochondrial dynamic imbalance.
Assuntos
Doença Hepática Crônica Induzida por Substâncias e Drogas , Hepatopatias , Ocratoxinas , Animais , Antioxidantes/farmacologia , Dinâmica Mitocondrial , Galinhas , Estresse Oxidativo , Estresse do Retículo Endoplasmático , Apoptose , XantofilasRESUMO
The frequency characteristics of lung sounds have great significance for noninvasive diagnosis of respiratory diseases. The rales in the lower respiratory tract region that can provide rich information about symptoms of respiratory diseases are not clear. In this paper, a three-dimensional idealized bifurcated lower respiratory tract geometric model, which contains 3rd to 13th generation (G3-G13) bronchi is constructed, where Re â¼ 10 1 - 10 3 , and then the large eddy simulation and volume of fluid are used to study the fluid flow characteristics. Ffowcs Williams and Hawkings model are subsequently used to study the frequency characteristics of rale of different generations of bronchi. The results showed that bronchial blockage and sputum movement will enhance the turbulence intensity and vortex shedding intensity of flow. The dominant frequency and highest value of sound pressure level (SPL) of rhonchi/moist crackles decrease with the increase of bronchial generation. The change rates of dominant frequency of rhonchi / moist crackles in adjacent generations were 5.0 ± 0.1 ~ 9.1 ± 0.2% and 3.1 ± 0.1 ~ 11.9 ± 0.3%, respectively, which is concentrated in 290 ~ 420 Hz and 200 ~ 300 Hz, respectively. The change rates of SPL of rhonchi/moist crackles were 8.8 ± 0.1 ~ 15.7 ± 0.1% and 7.1 ± 0.1 ~ 19.5 ± 0.2%, respectively, which is concentrated in 28 ~ 50 dB and 16 ~ 32 dB, respectively. In the same generation of bronchus (e.g., G8, G9) with the same degree of initial blockage, the dominant frequency and SPL of moist crackles can be 3.7 ± 0.2% and 4.5 ± 0.3% slightly higher than that of rhonchi, respectively. This research is conducive to the establishment of a rapid and accurate noninvasive diagnosis system for respiratory diseases.
Assuntos
Sons Respiratórios , Doenças Respiratórias , Humanos , Sons Respiratórios/diagnóstico , Brônquios , Simulação por ComputadorRESUMO
This randomized controlled trial aimed to determine whether lung ultrasound-guided fluid resuscitation improves the clinical outcomes of neonates with septic shock. Seventy-two patients were randomly assigned to undergo treatment with lung ultrasound-guided fluid resuscitation (LUGFR), or with usual fluid resuscitation (Control) in the first 6 h since the start of the sepsis treatment. The primary study outcome was 14-day mortality after randomization. Fourteen-day mortalities in the two groups were not significantly different (LUGFR group, 13.89%; control group, 16.67%; p = 0.76; hazard ratio 0.81 [95% CI 0.27-2.50]). The LUGFR group experienced shorter length of neonatal intensive care unit (NICU) stays (21 vs. 26 days, p = 0.04) and hospital stays (32 vs. 39 days, p = 0.01), and less fluid was used in the first 6 h (77 vs. 106 mL/kg, p = 0.02). Further, our study found that ultrasound-guided fluid resuscitation can significantly reduce the incidence of acute kidney injury (25% vs. 47.2%, p = 0.05) and intracranial hemorrhage (grades I-II) within 72 h (13.9% vs. 36.1%, p = 0.03). However, no significant difference was found in the resolution of shock within 1 h or 6 h, use of mechanical ventilation or vasopressor support, time to achieve lactate level < 2 mmol/L, and the number of participants developing hepatomegaly in the first 6 h. CONCLUSION: Lung ultrasound is a noninvasive and convenient tool for predicting fluid overload in neonatal septic shock. Fluid resuscitation guided by lung ultrasound can shorten the length of hospital and NICU stays, reduce the amount of fluid used in the first 6 h, and reduce the risk of acute kidney injury and intracranial hemorrhage. TRIAL REGISTRATION: Registered in Guangdong Second Provincial General Hospital: 2021-IIT-156-EK, date of registration: November 13, 2021. And ClinicalTrials.gov: NCT06144463 (retrospectively registered). WHAT IS KNOWN: ⢠Excessive fluid resuscitation in neonates with septic shock had worse outcomes. WHAT IS NEW: ⢠Lung ultrasound should be routinely used to guide fluid resuscitation in neonatal septic shock.
Assuntos
Injúria Renal Aguda , Choque Séptico , Recém-Nascido , Humanos , Choque Séptico/terapia , Hidratação , Ressuscitação , Pulmão/diagnóstico por imagem , Hemorragias Intracranianas , Ultrassonografia de IntervençãoRESUMO
Considering the significance of multipath transmissions of respiratory pathogens in the post-epidemic era, surprisingly little is acknowledged regarding the susceptibility, short-term aerodynamics, and exposure risk of children in indoor environments. Here, experimental and computational investigations were conducted to evaluate the exposure risks associated with respiratory pathogens. The dominant effect of recirculation structure originating from indoor ventilation, including air supply modes and air change rate, on aerosol dispersion was quantitatively proved. A large proportion of deposited aerosol particles was captured by the human body, and deposited particles may further increase under high air change rate, which required a balanced ventilation strategy. Little discrepancies existed between adults and children in exposure risk by airborne transmission. The infection probability by contact transmission for children, however, may be surprisingly high due to high frequency in interactive activities, deposition on upper and lower limbs of accompanying parents, and the wall within arm span.
RESUMO
Background: Lung cancer combined by chronic obstructive pulmonary disease (LC-COPD) is a common comorbidity and their interaction with each other poses significant clinical challenges. However, there is a lack of well-established consensus on the diagnosis and treatment of LC-COPD. Methods: A panel of experts, comprising specialists in oncology, respiratory medicine, radiology, interventional medicine, and thoracic surgery, was convened. The panel was presented with a comprehensive review of the current evidence pertaining to LC-COPD. After thorough discussions, the panel reached a consensus on 17 recommendations with over 70% agreement in voting to enhance the management of LC-COPD and optimize the care of these patients. Results: The 17 statements focused on pathogenic mechanisms (n=2), general strategies (n=4), and clinical application in COPD (n=2) and lung cancer (n=9) were developed and modified. These statements provide guidance on early screening and treatment selection of LC-COPD, the interplay of lung cancer and COPD on treatment, and considerations during treatment. This consensus also emphasizes patient-centered and personalized treatment in the management of LC-COPD. Conclusions: The consensus highlights the need for concurrent treatment for both lung cancer and COPD in LC-COPD patients, while being mindful of the mutual influence of the two conditions on treatment and monitoring for adverse reactions.
RESUMO
Progressive lung function decline is a hallmark of chronic obstructive pulmonary disease (COPD). Airway dysbiosis occurs in COPD, but whether it contributes to disease progression remains unknown. Here, we show, through a longitudinal analysis of two cohorts involving four UK centers, that baseline airway dysbiosis in COPD patients, characterized by the enrichment of opportunistic pathogenic taxa, associates with a rapid forced expiratory volume in 1 s (FEV1) decline over 2 years. Dysbiosis associates with exacerbation-related FEV1 fall and sudden FEV1 fall at stability, contributing to long-term FEV1 decline. A third cohort in China further validates the microbiota-FEV1-decline association. Human multi-omics and murine studies show that airway Staphylococcus aureus colonization promotes lung function decline through homocysteine, which elicits a neutrophil apoptosis-to-NETosis shift via the AKT1-S100A8/A9 axis. S. aureus depletion via bacteriophages restores lung function in emphysema mice, providing a fresh approach to slow COPD progression by targeting the airway microbiome.
Assuntos
Pulmão , Doença Pulmonar Obstrutiva Crônica , Humanos , Animais , Camundongos , Disbiose , Staphylococcus aureus , Volume Expiratório Forçado , Progressão da DoençaRESUMO
In the orthodontics process, intervention and sliding of an orthodontic bracket during the orthodontic process can arise large response of the labio-cheek soft tissue. Soft tissue damage and ulcers frequently happen at the early stage of orthodontic treatment. In the field of orthodontic medicine, qualitative analysis is always carried out through statistics of clinical cases, while quantitative explanation of bio-mechanical mechanism is lacking. For this purpose, finite element analysis of a three-dimensional labio-cheek-bracket-tooth model is conducted to quantify the bracket-induced mechanical response of the labio-cheek soft tissue, which involves complex coupling of contact nonlinearity, material nonlinearity and geometric nonlinearity. Firstly, based on the biological composition characteristics of labio-cheek, a second-order Ogden model is optimally selected to describe the adipose-like material of the labio-cheek soft tissue. Secondly, according to the characteristics of oral activity, a two-stage simulation model of bracket intervention and orthogonal sliding is established, and the key contact parameters are optimally set. Finally, the two-level analysis method of overall model and submodel is used to achieve efficient solution of high-precision strains in submodels based on the displacement boundary obtained from the overall model calculation. Calculation results with four typical tooth morphologies during orthodontic treatment show that: â the maximum strain of soft tissue is distributed along the sharp edges of the bracket, consistent with the clinically observed profile of soft tissue deformation; â¡ the maximum strain of soft tissue is reduced as the teeth align, consistent with the clinical manifestation of common damage and ulcers at the beginning of orthodontic treatment and reduced patient discomfort at the end of treatment. The method in this paper can provide reference for relevant quantitative analysis studies in the field of orthodontic medical treatment at home and abroad, and further benefit to the product development analysis of new orthodontic devices.
Assuntos
Ligamento Periodontal , Dente , Humanos , Ligamento Periodontal/fisiologia , Fios Ortodônticos , Bochecha , Úlcera , Análise de Elementos FinitosRESUMO
RATIONALE: Chronic obstructive pulmonary disease (COPD) is a complicated chronic inflammatory disease. It is important to investigate the characteristics of acute exacerbation of COPD to develop new therapeutic strategies. OBJECTIVE: This study aimed to determine the relationship between the human beta-defensin-2 (hBD-2) levels and aggravation of COPD. METHODS: We detected the sputum hBD-2 level of 254 patients from Guangzhou, China, for 2 years. The study participants were categorized into the COPD group (n = 203, GOLD 0-4) and the control group (n = 51, 40-79 years old). At baseline, 12th month, and 24th month, we detected the sputum hBD-2 level and levels of cytokines, such as CXCL10, CXCL11, and IFN. RESULTS: At baseline, there were no significant differences in the sputum and serum hBD-2 levels between the patients and the controls. However, the sputum hBD-2 levels of patients who had at least one symptom aggravation over the next 2 years were significantly lower than those of patients without any exacerbations (1130.9 ± 858.4 pg/mL vs. 2103.7 ± 1294.2 pg/mL, respectively; p = 0.001). Nevertheless, there were no statistically significant differences in the sputum hBD-2 levels between patients (no aggravation history) and controls (2084.9 ± 1317.6 pg/mL vs. 2152.5 ± 1251.6 pg/mL, respectively; p = 0.626). We used a logistic regression model to assess the relationship between aggravation and sputum hBD-2 levels. Interestingly, we found that low hBD-2 level (< 1000 pg/mL) was significantly associated with exacerbations. Specifically, patients with low hBD-2 levels were more likely to experience exacerbations in the next 12 months (0.333 vs. 0.117; p = 0.001). Moreover, we compared the hBD-2 levels between controls and patients with GOLD 3-4 and found that participants with bacteria (+) and/or viruses (+) had an association between hBD-2 level and disease severity (p = 0.02). CONCLUSION: Patients at risk of exacerbations are more likely to have lower sputum hBD-2 levels. These results have important implications for future therapies for COPD.
Assuntos
Doença Pulmonar Obstrutiva Crônica , Vírus , beta-Defensinas , Humanos , Adulto , Pessoa de Meia-Idade , Idoso , Escarro/microbiologia , beta-Defensinas/uso terapêutico , CitocinasRESUMO
Sensors have been used in various agricultural production scenarios due to significant advances in the Agricultural Internet of Things (Ag-IoT), leading to smart agriculture. Intelligent control or monitoring systems rely heavily on trustworthy sensor systems. Nonetheless, sensor failures are likely due to various factors, including key equipment malfunction or human error. A faulty sensor can produce corrupted measurements, resulting in incorrect decisions. Early detection of potential faults is crucial, and fault diagnosis techniques have been proposed. The purpose of sensor fault diagnosis is to detect faulty data in the sensor and recover or isolate the faulty sensors so that the sensor can finally provide correct data to the user. Current fault diagnosis technologies are based mainly on statistical models, artificial intelligence, deep learning, etc. The further development of fault diagnosis technology is also conducive to reducing the loss caused by sensor failures.
RESUMO
BACKGROUND: The coronavirus disease 2019 (COVID-19) and community-acquired pneumonia (CAP) present a high degree of similarity in chest computed tomography (CT) images. Therefore, a procedure for accurately and automatically distinguishing between them is crucial. METHODS: A deep learning method for distinguishing COVID-19 from CAP is developed using maximum intensity projection (MIP) images from CT scans. LinkNet is employed for lung segmentation of chest CT images. MIP images are produced by superposing the maximum gray of intrapulmonary CT values. The MIP images are input into a capsule network for patient-level pred iction and diagnosis of COVID-19. The network is trained using 333 CT scans (168 COVID-19/165 CAP) and validated on three external datasets containing 3581 CT scans (2110 COVID-19/1471 CAP). RESULTS: LinkNet achieves the highest Dice coefficient of 0.983 for lung segmentation. For the classification of COVID-19 and CAP, the capsule network with the DenseNet-121 feature extractor outperforms ResNet-50 and Inception-V3, achieving an accuracy of 0.970 on the training dataset. Without MIP or the capsule network, the accuracy decreases to 0.857 and 0.818, respectively. Accuracy scores of 0.961, 0.997, and 0.949 are achieved on the external validation datasets. The proposed method has higher or comparable sensitivity compared with ten state-of-the-art methods. CONCLUSIONS: The proposed method illustrates the feasibility of applying MIP images from CT scans to distinguish COVID-19 from CAP using capsule networks. MIP images provide conspicuous benefits when exploiting deep learning to detect COVID-19 lesions from CT scans and the capsule network improves COVID-19 diagnosis.
Assuntos
COVID-19 , Aprendizado Profundo , Pneumonia , Humanos , COVID-19/diagnóstico por imagem , Teste para COVID-19 , SARS-CoV-2 , Pneumonia/diagnóstico por imagem , Tomografia Computadorizada por Raios X/métodosRESUMO
BACKGROUND AND OBJECTIVES: Non-contrast CT (NCCT) and contrast-enhanced CT (CECT) are important diagnostic tools with distinct features and applications for chest diseases. We developed two synthesizers for the mutual synthesis of NCCT and CECT and evaluated their applications. METHODS: Two synthesizers (S1 and S2) were proposed based on a generative adversarial network. S1 generated synthetic CECT (SynCECT) from NCCT and S2 generated synthetic NCCT (SynNCCT) from CECT. A new training procedure for synthesizers was proposed. Initially, the synthesizers were pretrained using self-supervised learning (SSL) and dual-energy CT (DECT) and then fine-tuned using the registered NCCT and CECT images. Pulmonary vessel segmentation from NCCT was used as an example to demonstrate the effectiveness of the synthesizers. Two strategies (ST1 and ST2) were proposed for pulmonary vessel segmentation. In ST1, CECT images were used to train a segmentation model (Model-CECT), NCCT images were converted to SynCECT through S1, and SynCECT was input to Model-CECT for testing. In ST2, CECT data were converted to SynNCCT through S2. SynNCCT and CECT-based annotations were used to train an additional model (Model-NCCT), and NCCT was input to Model-NCCT for testing. Three datasets, D1 (40 paired CTs), D2 (14 NCCTs and 14 CECTs), and D3 (49 paired DECTs), were used to evaluate the synthesizers and strategies. RESULTS: For S1, the mean absolute error (MAE), mean squared error (MSE), peak signal-to-noise ratio (PSNR), and structural similarity index (SSIM) were 14.60± 2.19, 1644± 890, 34.34± 1.91, and 0.94± 0.02, respectively. For S2, they were 12.52± 2.59, 1460± 922, 35.08± 2.35, and 0.95± 0.02, respectively. Our synthesizers outperformed the counterparts of CycleGAN, Pix2Pix, and Pix2PixHD. The results of ablation studies on SSL pretraining, DECT pretraining, and fine-tuning showed that performance worsened (for example, for S1, MAE increased to 16.53± 3.10, 17.98± 3.10, and 20.57± 3.75, respectively). Model-NCCT and Model-CECT achieved dice similarity coefficients (DSC) of 0.77 and 0.86 on D1 and 0.77 and 0.72 on D2, respectively. CONCLUSIONS: The proposed synthesizers realized mutual and high-quality synthesis between NCCT and CECT images; the training procedures, including SSL pretraining, DECT pretraining, and fine-tuning, were critical to their effectiveness. The results demonstrated the usefulness of synthesizers for pulmonary vessel segmentation from NCCT images.
Assuntos
Tórax , Tomografia Computadorizada por Raios X , Tomografia Computadorizada por Raios X/métodos , Razão Sinal-Ruído , Processamento de Imagem Assistida por Computador/métodosRESUMO
Background: Diurnal temperature range (DTR) has been increasingly recognized as a risk factor for mortality and morbidity, but the association between DTR and acute lower respiratory infection (ALRI) outpatient visits has not been examined among children in China. Methods: A total of 79,416 ALRI outpatient visits among children were obtained from the Guangdong Second Provincial General Hospital between 2013 and 2019. DTR was calculated by taking the difference between the maximum and the minimum temperatures. Generalized additive models using a quasi-Poisson distribution were used to model the relationship between DTR and ALRI outpatient visits. Results: Diurnal temperature range was significantly associated with elevated risks of ALRI outpatient visits: the excess risks (ERs) and 95% confidence intervals (CIs) were 2.31% (1.26, 3.36%) for ALRI, 3.19% (1.86, 4.54%) for pneumonia, and 1.79% (0.59, 3.01%) for bronchiolitis, respectively. Subgroup analyses suggested that the associations were significantly stronger during rainy seasons (ER for ALRI: 3.02%, 95% CI: 1.43, 4.64%) than those in dry seasons (ER for ALRI: 2.21%, 95% CI: 0.65, 3.81%), while no significant effect modifications were found in sex and age groups. Conclusion: Diurnal temperature range may elevate the risk of ALRI outpatient visits among children in China, especially during rainy seasons. Public health policies are needed to mitigate the adverse health impacts of DTR on children.
Assuntos
Pacientes Ambulatoriais , Infecções Respiratórias , Criança , Humanos , Temperatura , Infecções Respiratórias/epidemiologia , Estações do Ano , China/epidemiologiaRESUMO
Background: Pulmonary vascular alteration is an important feature of chronic obstructive pulmonary disease (COPD), which is characterized by distal pulmonary vascular pruning in angiography. We aimed to further investigate the clinical relevance of pulmonary vasculature in COPD patients using non-contrast computed tomography (CT). Methods: Seventy-one control subjects and 216 COPD patients completed the questionnaires, spirometry, and computed tomography (CT) scans within 1 month and were included in the study. Small pulmonary vessels represented by percentage of cross-sectional area of pulmonary vessels smaller than 5 mm2 or 5-10 mm2 to the total lung fields (%CSA<5 or %CSA5-10, respectively) were measured using ImageJ software. Spearman correlation was used to investigate the relationship between %CSA<5 and airflow limitation. A receiver operating characteristic (ROC) curve was built to evaluate the value of %CSA<5 in discriminating COPD patients from healthy control subjects. Segmented regression was used to analyze the relationship between %CSA<5 and %LAA-950 (percentage of low-attenuation areas less than -950 HU). Results: We found a significant correlation between %CSA<5 and forced expiratory volume in one second (FEV1) percentage of predicted value (%pred) (r = 0.564, P < 0.001). The area under the ROC curve for the value of %CSA<5 in distinguishing COPD was 0.816, with a cut-off value of 0.537 (Youden index J, 0.501; sensitivity, 78.24%; specificity, 71.83%). Since the relationship between %CSA<5 and %LAA-950 was not constant, performance of segmented regression was better than ordinary linear regression (adjusted R2, 0.474 vs 0.332, P < 0.001 and P < 0.001, respectively). As %CSA<5 decreased, %LAA-950 slightly increased until an inflection point (%CSA<5 = 0.524) was reached, after which the %LAA-950 increased apparently with a decrease in %CSA<5. Conclusion: %CSA<5 was significantly correlated with both airflow limitation and emphysema, and we identified an inflection point for the relationship between %CSA<5 and %LAA-950.