A semiparametric Gaussian mixture model for chest CT-based 3D blood vessel reconstruction.

Computed tomography (CT) has been a powerful diagnostic tool since its emergence in the 1970s. Using CT data, 3D structures of human internal organs and tissues, such as blood vessels, can be reconstructed using professional software. This 3D reconstruction is crucial for surgical operations and can serve as a vivid medical teaching example. However, traditional 3D reconstruction heavily relies on manual operations, which are time-consuming, subjective, and require substantial experience. To address this problem, we develop a novel semiparametric Gaussian mixture model tailored for the 3D reconstruction of blood vessels. This model extends the classical Gaussian mixture model by enabling nonparametric variations in the component-wise parameters of interest according to voxel positions. We develop a kernel-based expectation-maximization algorithm for estimating the model parameters, accompanied by a supporting asymptotic theory. Furthermore, we propose a novel regression method for optimal bandwidth selection. Compared to the conventional cross-validation-based (CV) method, the regression method outperforms the CV method in terms of computational and statistical efficiency. In application, this methodology facilitates the fully automated reconstruction of 3D blood vessel structures with remarkable accuracy.

Performance of rapid on-site evaluation of touch imprints of bronchoscopic biopsies or lung tissue biopsies for the diagnosis of invasive pulmonary filamentous fungi infections in non-neutropenic patients.

The diagnosis of invasive pulmonary fungal disease depends on histopathology and mycological culture; there are few studies on touch imprints of bronchoscopic biopsies or lung tissue biopsies for the diagnosis of pulmonary filamentous fungi infections. The purpose of the present study was to explore the detection accuracy of rapid on-site evaluation of touch imprints of bronchoscopic biopsies or lung tissue biopsies for the filamentous fungi, and it aims to provide a basis for initiating antifungal therapy before obtaining microbiological evidence. We retrospectively analyzed the diagnosis and treatment of 44 non-neutropenic patients with invasive pulmonary filamentous fungi confirmed by glactomannan assay, histopathology, and culture from February 2017 to December 2023. The diagnostic positive rate and sensitivity of rapid on-site evaluation for these filamentous fungi identification, including diagnostic turnaround time, were calculated. Compared with the final diagnosis, the sensitivity of rapid on-site evaluation was 81.8%, and the sensitivity of histopathology, culture of bronchoalveolar lavage fluid, and glactomannan assay of bronchoalveolar lavage fluid was 86.4%, 52.3%, and 68.2%, respectively. The average turnaround time of detecting filamentous fungi by rapid on-site evaluation was 0.17 ± 0.03 hours, which was significantly faster than histopathology, glactomannan assay, and mycological culture. A total of 29 (76.3%) patients received earlier antifungal therapy based on ROSE diagnosis and demonstrated clinical improvement. Rapid on-site evaluation showed good sensitivity and accuracy that can be comparable to histopathology in identification of pulmonary filamentous fungi. Importantly, it contributed to the triage of biopsies for further microbial culture or molecular detection based on the preliminary diagnosis, and the decision on early antifungal therapy before microbiological evidence is available.

Combined brain topological metrics with machine learning to distinguish essential tremor and tremor-dominant Parkinson's disease.

BACKGROUND: Essential tremor (ET) and Parkinson's disease (PD) are the two most prevalent movement disorders, sharing several overlapping tremor clinical features. Although growing evidence pointed out that changes in similar brain network nodes are associated with these two diseases, the brain network topological properties are still not very clear. OBJECTIVE: The combination of graph theory analysis with machine learning (ML) algorithms provides a promising way to reveal the topological pathogenesis in ET and tremor-dominant PD (tPD). METHODS: Topological metrics were extracted from Resting-state functional images of 86 ET patients, 86 tPD patients, and 86 age- and sex-matched healthy controls (HCs). Three steps were conducted to feature dimensionality reduction and four frequently used classifiers were adopted to discriminate ET, tPD, and HCs. RESULTS: A support vector machine classifier achieved the best classification performance of four classifiers for discriminating ET, tPD, and HCs with 89.0% mean accuracy (mACC) and was used for binary classification. Particularly, the binary classification performances among ET vs. tPD, ET vs. HCs, and tPD vs. HCs were with 94.2% mACC, 86.0% mACC, and 86.3% mACC, respectively. The most power discriminative features were mainly located in the default, frontal-parietal, cingulo-opercular, sensorimotor, and cerebellum networks. Correlation analysis results showed that 2 topological features negatively and 1 positively correlated with clinical characteristics. CONCLUSIONS: These results demonstrated that combining topological metrics with ML algorithms could not only achieve high classification accuracy for discrimination ET, tPD, and HCs but also help to reveal the potential brain topological network pathogenesis in ET and tPD.

Identification of essential tremor based on resting-state functional connectivity.

Currently, machine-learning algorithms have been considered the most promising approach to reach a clinical diagnosis at the individual level. This study aimed to investigate whether the whole-brain resting-state functional connectivity (RSFC) metrics combined with machine-learning algorithms could be used to identify essential tremor (ET) patients from healthy controls (HCs) and further revealed ET-related brain network pathogenesis to establish the potential diagnostic biomarkers. The RSFC metrics obtained from 127 ET patients and 120 HCs were used as input features, then the Mann-Whitney U test and the least absolute shrinkage and selection operator (LASSO) methods were applied to reduce feature dimensionality. Four machine-learning algorithms were adopted to identify ET from HCs. The accuracy, sensitivity, specificity and the area under the curve (AUC) were used to evaluate the classification performances. The support vector machine, gradient boosting decision tree, random forest and Gaussian naïve Bayes algorithms could achieve good classification performances with accuracy at 82.8%, 79.4%, 78.9% and 72.4%, respectively. The most discriminative features were primarily located in the cerebello-thalamo-motor and non-motor circuits. Correlation analysis showed that two RSFC features were positively correlated with tremor frequency and four RSFC features were negatively correlated with tremor severity. The present study demonstrated that combining the RSFC matrices with multiple machine-learning algorithms could not only achieve high classification accuracy for discriminating ET patients from HCs but also help us to reveal the potential brain network pathogenesis in ET.

Performance of rapid on-site evaluation of touch imprints of lung tissue biopsies for the diagnosis of pulmonary cryptococcosis in patients without HIV infection.

BACKGROUND AND OBJECTIVE: The diagnosis of pulmonary cryptococcosis depends on serum testing, histopathology and mycological culture; there are few studies on touch imprints of lung tissue biopsies for the diagnosis of pulmonary cryptococcosis in patients without HIV infection. The purpose of the current study was to investigate the accuracy and timeliness of on-site touch imprint cytology in the diagnosis of pulmonary cryptococcosis during CT-guided percutaneous lung biopsy. METHODS: We retrospectively analysed the diagnosis and treatment of 56 patients with final proof of pulmonary cryptococcosis through histopathology and culture or surgical resection from September 2015 to February 2021. Diagnostic methods and treatment and the turnaround time for diagnosis were analysed. RESULTS: The sensitivity of rapid on-site evaluation was 89.3%, and the sensitivity of serology, histopathology and mycological culture was 53.6%, 91.1% and 61.5%, respectively, compared with the final diagnosis. The average turnaround time to diagnose pulmonary cryptococcosis by on-site touch imprint cytology was 8.3 ± 0.9 min, which was significantly faster than serum testing, histopathology and mycological culture. CONCLUSION: On-site touch imprint cytology showed good sensitivity and timeliness in the diagnosis of pulmonary cryptococcosis. In addition, it contributed to the triage of biopsies based on the preliminary diagnosis. On-site touch imprint cytology should be applied and promoted in the diagnosis of pulmonary cryptococcosis during biopsy.

Altered local and matrix functional connectivity in depressed essential tremor patients.

BACKGROUND: Depression in essential tremor (ET) has been constantly studied and reported, while the associated brain activity changes remain unclear. Recently, regional homogeneity (ReHo), a voxel-wise local functional connectivity (FC) analysis of resting-state functional magnetic resonance imaging, has provided a promising way to observe spontaneous brain activity. METHODS: Local FC analyses were performed in forty-one depressed ET patients, 49 non-depressed ET patients and 43 healthy controls (HCs), and then matrix FC and clinical depression severity correlation analyses were further performed to reveal spontaneous neural activity changes in depressed ET patients. RESULTS: Compared with the non-depressed ET patients, the depressed ET patients showed decreased ReHo in the bilateral cerebellum lobules IX, and increased ReHo in the bilateral anterior cingulate cortices and middle prefrontal cortices. Twenty-five significant changes of ReHo clusters were observed in the depressed ET patients compared with the HCs, and matrix FC analysis further revealed that inter-ROI FC differences were also observed in the frontal-cerebellar-anterior cingulate cortex pathway. Correlation analyses showed that clinical depression severity was positively correlated with the inter-ROI FC values between the anterior cingulate cortex and bilateral middle prefrontal cortices and was negatively correlated with the inter-ROI FC values of the anterior cingulate cortex and bilateral cerebellum lobules IX. CONCLUSION: Our findings revealed local and inter-ROI FC differences in frontal-cerebellar-anterior cingulate cortex circuits in depressed ET patients, and among these regions, the cerebellum lobules IX, middle prefrontal cortices and anterior cingulate cortices could function as pathogenic structures underlying depression in ET patients.

Rapid on-site evaluation of touch imprints of biopsies improves the diagnostic yield of transbronchial biopsy for endoscopically nonvisible malignancy: a retrospective study.

OBJECTIVE: Rapid on-site evaluation has long been used for transbronchial needle aspiration or fine-needle aspiration to evaluate the adequacy of biopsy materials for the diagnosis of peripheral lung lesions. However, research on rapid on-site evaluation combined with transbronchial forceps biopsy in the diagnosis of lung carcinoma is rarely reported. Therefore, we aimed to investigate the value of rapid on-site evaluation during transbronchial forceps biopsy for endoscopically visible (tumor, infiltrative and necrotic) or nonvisible (compressive, nonspecific and normal) malignancy. METHODS: A retrospective analysis was performed between January 2015 and January 2019 in Taihe Hospital with 1216 lung cancer patients who underwent bronchoscopy procedures, and these patients were allocated into the rapid on-site evaluation group and non-rapid on-site evaluation group, depending on the timing of the procedure. According to endoscopic features, bronchoscopic appearance was described as endoscopically visible malignancy (tumor, infiltrative and necrotic) and endoscopically nonvisible malignancy (compressive, nonspecific and normal). The diagnostic yield was compared, and the concordance between the rapid on-site evaluation results and the final histology was analyzed. RESULTS: There was a statistically significant difference in the diagnostic yield between the rapid on-site evaluation and non-rapid on-site evaluation groups for endoscopically nonvisible malignancy (74.3% vs. 51.7%, P < 0.05). However, we found no significant improvement in terms of diagnostic yield for endoscopically visible malignancy (95.2% vs. 91.2%, P > 0.05). The rapid on-site evaluation results showed high-level concordance with histology in the diagnosis of squamous cell carcinoma, adenocarcinoma and small cell carcinoma, with kappa values of 0.749 (P < 0.05), 0.728 (P < 0.05) and 0.940 (P < 0.05), respectively. CONCLUSIONS: The findings demonstrated that the diagnostic yield of transbronchial biopsy for endoscopically nonvisible malignancy (compressive, nonspecific and normal) was significantly improved when rapid on-site evaluation was implemented. In addition, the rapid on-site evaluation results had high-level concordance with the final histological diagnosis.

Feature Screening for Network Autoregression Model.

Network analysis has drawn great attention in recent years. It is applied to a wide range disciplines. These include but are not limited to social science, finance and genetics. It is typical that one collects abundant covariates along the response variable in practice. Since the network structure makes the responses at different nodes no longer independent, existing screening methods may not perform well for network data. We propose a network-based sure independence screening (NW-SIS) method. This approach explicitly takes the network structure into consideration. The strong screening consistency property of the NW-SIS is rigorously established. We further investigated the estimation of the network effect and establish the n -consistency of the estimator. The finite sample performance of the proposed method is assessed by simulation study and illustrated by an empirical analysis of a dataset from Chinese stock market.

Bronchial Dieulafoy's disease: a retrospective analysis of 73 cases.

BACKGROUND: Bronchial Dieulafoy's disease (BDD) is a rare disease that is known to be a cause of hemorrhage. The characteristics of this disease are still unknown. The present study describes the disorder based on a review of the world's literature, emphasizing the diagnostic and therapeutic views. METHODS: A comprehensive research of BDD of the PubMed, Google Scholar, and Web of Science databases was performed. The following data were collected: patient characteristics; chest imaging, bronchoscopy, vascular angiography, and histopathologic examination findings; and treatment rendered. RESULTS: 73 cases of BDD have been reported from 1995 to 2019. Most of the cases occurred in Asia (52.1%), followed by Europe (31.5%). Chest imaging findings were non-specific. The main bronchoscopy finding was a nodular or protruding lesion (60.9%). 19 patients underwent bronchoscopic biopsies, 17 had bleeding, and 6 died. Four patients were successfully shown to have vascular malformations under mucosal protrusion by endobronchial ultrasound scan (EBUS). Vascular angiography mainly showed tortuous, dilated bronchial arteries. Vascular angiography mainly showed tortuous, dilated bronchial arteries. The arterial supply was mainly provided by bronchial arteries (48 cases) and the pulmonary circulation (4 cases). The lesions were mainly located in the right bronchus (53 cases). Selective bronchial artery embolization (BAE) was attempted in 38 patients and 20 patients underwent lobectomies. Emergency resection was performed in 15 patients, all of whom survived and had no recurrent hemoptysis. CONCLUSIONS: Massive hemoptysis was the common manifestation of BDD. Vascular angiography and EBUS is a very useful examination before biopsy. BAE may be used in stable patients, or patients who cannot tolerate surgery, while surgical resection should be considered in patients who are unstable, patients with uncontrolled hemoptysis, or following BAE failure.

Portal Nodes Screening for Large Scale Social Networks.

Network autoregression model (NAM), as a powerful tool to study user social behaviors on large scale social networks, has drawn great attention in recent years. In this paper, we are interested in identifying the influential users (i.e., portal nodes) in a social network under the framework of NAM. Especially, we consider the autoregression model that allows to have a heterogenous and sparse network effect coefficients. Therefore, the portal nodes take influential powers which are corresponding to the nonzero network effect coefficients. A screening procedure is designed to screen out the portal nodes and the strong screening consistency is established theoretically. A quasi maximum likelihood method is applied to estimate the influential powers. The asymptotic normality of the resulting estimator is established. Further selection procedure is given by taking advantage of the local linear approximation algorithm. Extensive numerical studies are conducted by using a Sina Weibo dataset for illustration purpose.

Dynamic contrast-enhanced magnetic resonance imaging for differentiating osteomyelitis from acute neuropathic arthropathy in the complicated diabetic foot.

OBJECTIVE: The main purpose of this study was to investigate the diagnostic value of dynamic contrast-enhanced MRI (DCE-MRI) in differentiating osteomyelitis from acute neuropathic arthropathy in the diabetic foot. MATERIALS AND METHODS: This prospective study was carried out on 30 diabetic foot patients, with a mean age of 51 years. The patients all underwent clinical examinations, laboratory examinations and DCE-MRI. The DCE-MRI parameters (Ktrans, Kep and Ve) of the regions of acute neuropathic arthropathy and osteomyelitis were calculated. Receiver operating characteristic curves (ROCs) were used to identify the DCE-MRI parameters that showed the highest accuracy in differentiating the acute neuropathic arthropathy from the osteomyelitic regions. Pearson correlation coefficients were used to assess the correlations among the DCE-MRI parameters, the level of C-reactive protein (CRP) and the erythrocyte sedimentation rate (ESR). RESULTS: The Ktrans, Kep and Ve values of the osteomyelitic regions were higher than those of the acute neuropathic arthropathy regions, and significant differences were found between the two groups (P = 0.000, P = 0.000, P = 0.000). The ROC analysis showed that Ktrans and Ve performed best in differentiating osteomyelitis from acute neuropathic arthropathy, both with an area under the curve of 0.938. The Pearson correlation coefficients showed that the DCE-MRI parameters correlated significantly with the level of CRP and ESR (P = 0.000, P = 0.014, P = 0.000; P = 0.000, P = 0.000, P = 0.013). CONCLUSIONS: Our results showed that DCE-MRI may provide reproducible parameters that can reliably differentiate osteomyelitis from acute neuropathic arthropathy.

Essential tremor is associated with disruption of functional connectivity in the ventral intermediate Nucleus--Motor Cortex--Cerebellum circuit.

The clinical benefits of targeting the ventral intermediate nucleus (VIM) for the treatment of tremors in essential tremor (ET) patients suggest that the VIM is a key hub in the network of tremor generation and propagation and that the VIM can be considered as a seed region to study the tremor network. However, little is known about the central tremor network in ET patients. Twenty-six ET patients and 26 matched healthy controls (HCs) were included in this study. After considering structural and head-motion factors and establishing the accuracy of our seed region, a VIM seed-based functional connectivity (FC) analysis of resting-state functional magnetic resonance imaging (RS-fMRI) data was performed to characterize the VIM FC network in ET patients. We found that ET patients and HCs shared a similar VIM FC network that was generally consistent with the VIM anatomical connectivity network inferred from normal nonhuman primates and healthy humans. Compared with HCs, ET patients displayed VIM-related FC changes, primarily within the VIM-motor cortex (MC)-cerebellum (CBLM) circuit, which included decreased FC in the CBLM and increased FC in the MC. Importantly, tremor severity correlated with these FC changes. These findings provide the first evidence that the pathological tremors observed in ET patients might be based on a physiologically pre-existing VIM - MC - CBLM network and that disruption of FC in this physiological network is associated with ET. Further, these findings demonstrate a potential approach for elucidating the neural network mechanisms underlying this disease.

Testing a single regression coefficient in high dimensional linear models.

In linear regression models with high dimensional data, the classical z-test (or t-test) for testing the significance of each single regression coefficient is no longer applicable. This is mainly because the number of covariates exceeds the sample size. In this paper, we propose a simple and novel alternative by introducing the Correlated Predictors Screening (CPS) method to control for predictors that are highly correlated with the target covariate. Accordingly, the classical ordinary least squares approach can be employed to estimate the regression coefficient associated with the target covariate. In addition, we demonstrate that the resulting estimator is consistent and asymptotically normal even if the random errors are heteroscedastic. This enables us to apply the z-test to assess the significance of each covariate. Based on the p-value obtained from testing the significance of each covariate, we further conduct multiple hypothesis testing by controlling the false discovery rate at the nominal level. Then, we show that the multiple hypothesis testing achieves consistent model selection. Simulation studies and empirical examples are presented to illustrate the finite sample performance and the usefulness of the proposed method, respectively.

Multiple Resting-State Networks Are Associated With Tremors and Cognitive Features in Essential Tremor.

INTRODUCTION: The heterogeneous clinical features of essential tremor indicate that the dysfunctions of this syndrome are not confined to motor networks, but extend to nonmotor networks. Currently, these neural network dysfunctions in essential tremor remain unclear. In this study, independent component analysis of resting-state functional MRI was used to study these neural network mechanisms. METHODS: Thirty-five essential tremor patients and 35 matched healthy controls with clinical and neuropsychological tests were included, and eight resting-state networks were identified. After considering the structure and head-motion factors and testing the reliability of the selected resting-state networks, we assessed the functional connectivity changes within or between resting-state networks. Finally, image-behavior correlation analysis was performed. RESULTS: Compared to healthy controls, essential tremor patients displayed increased functional connectivity in the sensorimotor and salience networks and decreased functional connectivity in the cerebellum network. Additionally, increased functional network connectivity was observed between anterior and posterior default mode networks, and a decreased functional network connectivity was noted between the cerebellum network and the sensorimotor and posterior default mode networks. Importantly, the functional connectivity changes within and between these resting-state networks were correlated with the tremor severity and total cognitive scores of essential tremor patients. CONCLUSIONS: The findings of this study provide the first evidence that functional connectivity changes within and between multiple resting-state networks are associated with tremors and cognitive features of essential tremor, and this work demonstrates a potential approach for identifying the underlying neural network mechanisms of this syndrome.

Multimodal deep learning-based drought monitoring research for winter wheat during critical growth stages.

Wheat is a major grain crop in China, accounting for one-fifth of the national grain production. Drought stress severely affects the normal growth and development of wheat, leading to total crop failure, reduced yields, and quality. To address the lag and limitations inherent in traditional drought monitoring methods, this paper proposes a multimodal deep learning-based drought stress monitoring S-DNet model for winter wheat during its critical growth periods. Drought stress images of winter wheat during the Rise-Jointing, Heading-Flowering and Flowering-Maturity stages were acquired to establish a dataset corresponding to soil moisture monitoring data. The DenseNet-121 model was selected as the base network to extract drought features. Combining the drought phenotypic characteristics of wheat in the field with meteorological factors and IoT technology, the study integrated the meteorological drought index SPEI, based on WSN sensors, and deep image learning data to build a multimodal deep learning-based S-DNet model for monitoring drought stress in winter wheat. The results show that, compared to the single-modal DenseNet-121 model, the multimodal S-DNet model has higher robustness and generalization capability, with an average drought recognition accuracy reaching 96.4%. This effectively achieves non-destructive, accurate, and rapid monitoring of drought stress in winter wheat.

Case report: Endobronchial closure of postoperative bronchopleural fistula with embolization coil: a sandwich-like approach.

Background: Embolization Coil has been reported to effectively treat postoperative bronchopleural fistula (BPF). Little detailed information was available on computer tomography (CT) imaging features of postoperative BPF and treating procedures with pushable Embolization Coil. Objective: We aimed to specify the imaging characteristics of postoperative BPFs and present our experience treating them with the pushable Embolization Coil. Methods: Six consecutive patients (four males and two females aged 29-56 years) diagnosed with postoperative BPF receiving bronchoscopic treatment with the pushable Nester® Embolization Coil (Cook Medical, Bloomington, Indiana) were included in this single-center, retrospective study. Multiplanar reconstruction of multidetector CT scans was reviewed for the presence, location, size, and radiological complications of each BPF, including air collection, pneumothorax, bronchiectasis, and chest tube. Using standardized data abstraction forms, demographic traits and clinical outcomes were extracted from the medical files of these patients. Results: The underlying diseases for lung resection surgery were pulmonary tuberculosis (n = 3), lung adenocarcinoma (n = 2), and pulmonary aspergillosis (n = 1). All patients had air or air-fluid collection with chest tubes on radiological findings. Multiplanar reconstruction identified the presence of postoperative BPF in all patients. Five fistulas were central, located proximal to the main or lobar bronchus, while one was peripheral, distant from the lobar bronchus. Fistula sizes ranged from 0.8 to 5.8 mm. Subsequent bronchoscopy and occlusion testing confirmed fistula openings in the bronchial stump: right main bronchus (n = 1), right upper lobe (n = 2), and left upper lobe (n = 3). The angioplasty catheter-based procedure allows precise fistula occlusion "like a sandwich" with the pushable Embolization Coil. Five patients with BPF sizes ranging from 0.8 to 1.5 mm were successfully treated with a pushable Embolization Coil, except for one with a BPF size of 5.8 mm. No adverse events or complications were observed throughout follow-up, ranging from 29 to 1,307 days. Conclusion: The pushable Nester® Embolization Coil seems a minimally invasive, cost-effective, and relatively easy-to-perform bronchoscopic treatment for postoperative BPF with a size less than 2 mm. Further studies are required to ensure the use of pushable Embolization Coil in treating postoperative BPF.

Pulmonary cryptococcosis coexisting with lung adenocarcinoma: A case report and review of the literature.

Pulmonary cryptococcosis (PC) is an invasive pulmonary fungal disease caused by Cryptococcus neoformans or Cryptococcus gattii. It often presents as a single nodule or mass on radiology, which is easily misdiagnosed as lung cancer or metastases. However, cases of PC coexisting with lung cancer are rare and when this scenario is encountered in clinical practice, it is easy to be misdiagnosed as metastatic lung cancer. The present study reported the case of a 65-year-old immunocompetent patient with PC coexisting with lung adenocarcinoma. Percutaneous lung biopsy was performed on the nodule in the anterior segment of the left upper lobe and the nodule in the posterior basal segment of the left lower lobe, which were diagnosed as primary adenocarcinoma and cryptococcus, respectively. Lung cancer was treated by surgery and PC was treated successfully by antifungal treatment. During the 5-year follow-up, contrast-enhanced CT showed no recurrence of either disease. This case reminds us of the possibility of dualism in the diagnosis of multiple pulmonary nodules based on CT examination, such as the coexistence of lung carcinoma and PC. In addition, early diagnosis and treatment contribute to good prognosis.

Clinical values of different specimen preparation methods for the diagnosis of lung cancer by EBUS-TBNA.

BACKGROUND AND OBJECTIVE: EBUS-TBNA has emerged as an important minimally invasive procedure for the diagnosis and staging of lung cancer. Our objective was to evaluate the effect of different specimen preparation from aspirates on the diagnosis of lung cancer. METHODS: 181 consecutive patients with known or suspected lung cancer accompanied by hilar / mediastinal lymphadenopathy underwent EBUS-TBNA from January 2019 to December 2022. Specimens obtained by EBUS-TBNA were processed by three methods: Traditional smear cytology of aspirates (TSC), liquid-based cytology of aspirates (LBC) and histopathology of core biopsies. RESULTS: EBUS-TBNA was performed in 181 patients on 213 lymph nodes, the total positive rate of the combination of three specimen preparation methods was 80.7%. The diagnostic positive rate of histopathology was 72.3%, TSC was 68.1%, and LBC was 65.3%, no significant differences was observed (p = 0.29); however, statistically significant difference was noted between the combination of three preparation methods and any single specimen preparation methods (p = 0.002). The diagnostic sensitivity of histopathology combined with TSC and histopathology combined with LBC were 96.5 and 94.8%, the specificity was 95.0% and 97.5%, the PPV was 98.8% and 99.4%, the NPV was 86.4% and 81.2%, the diagnostic accuracy was 96.2% and 95.3%, respectively; The sensitivity and accuracy of above methods were higher than that of single specimen preparation, but lower than that of combination of three preparation methods. CONCLUSION: When EBUS-TBNA is used for the diagnosis and staging of lung cancer, histopathology combined with TSC can achieve enough diagnostic efficiency and better cost-effectiveness.

Machine learning models for diagnosis of essential tremor and dystonic tremor using grey matter morphological networks.

BACKGROUND: Essential tremor (ET) and dystonic tremor (DT) are the two most common tremor disorders, and misdiagnoses are very common due to similar tremor symptoms. In this study, we explore the structural network mechanisms of ET and DT using brain grey matter (GM) morphological networks and combine those with machine learning models. METHODS: 3D-T1 structural images of 75 ET patients, 71 DT patients, and 79 healthy controls (HCs) were acquired. We used voxel-based morphometry to obtain GM images and constructed GM morphological networks based on the Kullback-Leibler divergence-based similarity (KLS) method. We used the GM volumes, morphological relations, and global topological properties of GM-KLS morphological networks as input features. We employed three classifiers to perform the classification tasks. Moreover, we conducted correlation analysis between discriminative features and clinical characteristics. RESULTS: 16 morphological relations features and 1 global topological metric were identified as the discriminative features, and mainly involved the cerebello-thalamo-cortical circuits and the basal ganglia area. The Random Forest (RF) classifier achieved the best classification performance in the three-classification task, achieving a mean accuracy (mACC) of 78.7%, and was subsequently used for binary classification tasks. Specifically, the RF classifier demonstrated strong classification performance in distinguishing ET vs. HCs, ET vs. DT, and DT vs. HCs, with mACCs of 83.0 %, 95.2 %, and 89.3 %, respectively. Correlation analysis demonstrated that four discriminative features were significantly associated with the clinical characteristics. CONCLUSION: This study offers new insights into the structural network mechanisms of ET and DT. It demonstrates the effectiveness of combining GM-KLS morphological networks with machine learning models in distinguishing between ET, DT, and HCs.

Identification of winter wheat pests and diseases based on improved convolutional neural network.

Wheat pests and diseases are one of the main factors affecting wheat yield. According to the characteristics of four common pests and diseases, an identification method based on improved convolution neural network is proposed. VGGNet16 is selected as the basic network model, but the problem of small dataset size is common in specific fields such as smart agriculture, which limits the research and application of artificial intelligence methods based on deep learning technology in the field. Data expansion and transfer learning technology are introduced to improve the training mode, and then attention mechanism is introduced for further improvement. The experimental results show that the transfer learning scheme of fine-tuning source model is better than that of freezing source model, and the VGGNet16 based on fine-tuning all layers has the best recognition effect, with an accuracy of 96.02%. The CBAM-VGGNet16 and NLCBAM-VGGNet16 models are designed and implemented. The experimental results show that the recognition accuracy of the test set of CBAM-VGGNet16 and NLCBAM-VGGNet16 is higher than that of VGGNet16. The recognition accuracy of CBAM-VGGNet16 and NLCBAM-VGGNet16 is 96.60 and 97.57%, respectively, achieving high precision recognition of common pests and diseases of winter wheat.