Abstract Objectives To develop and validate a radiomics model based on grayscale ultrasound (GSUS) images for characterizing subpleural pulmonary lesions (SPLs). Materials and methods In this prospective, multicenter study, 738 patients with CT-confirmed SPLs were enrolled from three institutions and assigned to training (n = 407), internal validation (n = 146), and external validation (n = 185) cohorts. A total of 1320 radiomics features were extracted from both lesion and perilesional regions on GSUS images. Feature selection was performed through intra- and inter-class correlation coefficients (ICCs) analyses, Pearson correlation analyses, and least absolute shrinkage and selection operator (LASSO) regression. Clinical–radiomics fusion models were subsequently constructed by integrating selected radiomics features with key clinical variables using multivariate logistic regression. Model performance was evaluated comprehensively using the area under the receiver-operating characteristic curve (AUC), sensitivity, specificity, F1-score, and additional diagnostic metrics. Results Five predictive models were constructed based on clinical, radiologic, and radiomics features. Among them, the integrated model combining lesion-based radiomics with clinical variables achieved the best diagnostic performance, with AUCs of 0.884 (95% CI: 0.828–0.940) in the internal validation cohort and 0.848 (95% CI: 0.791–0.904) in the external validation cohort. Calibration and decision curve analyses demonstrated good model calibration and favorable clinical utility. The diagnostic accuracy of the model was comparable to that of experienced lung ultrasound radiologists. Conclusions The GSUS-based radiomics model effectively differentiates between benign and malignant SPLs, demonstrating strong diagnostic performance and promising clinical applicability. Critical relevance statement The proposed ultrasound-based radiomics model provides a reproducible, noninvasive decision-support tool for characterizing subpleural pulmonary lesions, offering particular value in patients for whom invasive procedures are unsuitable or in settings where CT or biopsy is not readily available. Key Points Accurate characterization of subpleural pulmonary lesions remains challenging using conventional imaging techniques. The grayscale ultrasound radiomics model achieved accuracy comparable to expert radiologists. This model provides a noninvasive and accessible tool when CT or biopsy is limited. Graphical Abstract

Piezoelectric materials generate charges that directly interact with cancerous tissue or stimulate the production of reactive oxygen species (ROS) for innovative tumor therapies mediated by sonography. However, the precise optimization of piezoelectric nanomaterials, combined with overcoming apoptosis resistance, represents a substantial challenge that requires immediate attention. In this study, we have strategically designed cancer cell membrane-coated sodium metaniobate (NaNbO3; MNO) piezoelectric nanocubes with inherent homologous targeting capabilities and exceptional ROS-generating potential for the treatment of lung cancer. The cell membrane encapsulation technique significantly enhances the accumulation and retention of the nanopiezoelectric system at the tumor site. Crucially, in addition to its apoptotic induction properties, the increased ROS production activates pyroptosis via the ROS-NLRP3-Caspase-1-GSDMD signaling pathway, thereby augmenting therapeutic efficacy against tumors. Both in vitro and in vivo antineoplastic evaluations validate the advantages and potential of this biomimetic nanopiezoelectric system. This study underscores the role of biomimetic sonopiezoelectric engineering in catalytically inducing a dual mode of cell death, involving both apoptosis and pyroptosis, within lung cancer cells.

Objective: To develop a deep learning model based on dual-modality ultrasound (DMUS) video recognition for the differential diagnosis of benign and malignant subpleural pulmonary nodules (SPNs). Patients and Methods: Participant data (n=193, median age, 58 years [IQR, 34-66 years]; 123 men) with SPNs, prospectively collected from January 7, to December 21, 2020, were divided into training (n=154) and validation (n=39) sets in an 8:2 ratio. Additionally, independent internal (n=88) and external (n=91) test sets were prospectively collected from January 10 to June 25, 2021. The nature of the SPNs was determined through biopsy (n=306) and clinical follow-up (n=66). Our model integrated DMUS videos, time-intensity curves, and clinical information. The model’s performance was evaluated using area under the receiver operating characteristic curve, accuracy, sensitivity, and specificity and compared with state-of-the-art video classification models, as well as ultrasound and computed tomography diagnoses made by radiologists. Results: In the internal test set, our model accurately distinguished malignant from benign SPNs with an AUC, accuracy, sensitivity, and specificity of 0.91, 91% (80 of 88), 90% (27 of 30), and 91% (53 of 58), outperforming state-of-the-art video classification models (all P<.05). In the external test set, the model achieved the accuracy, sensitivity, and specificity of 89% (81 of 91), 84% (27 of 32), and 92% (54 of 59), which were higher than the parameters for radiologist interpretations of ultrasound (81% [74 of 91], 63% [20 of 32], and 92% [54 of 59]) and computed tomography (76% [69 of 91], 91% [29 of 32], and 68% [40 of 59]), respectively. Conclusion: This deep learning model based on DMUS video recognition enhances the performance of ultrasound in differentiating benign from malignant SPNs. Trial Registration: clinicaltrials.gov Identifier: ChiCTR1800019828

OBJECTIVE:This study aimed to evaluate the procedural safety and diagnostic efficacy of ultrasound-guided percutaneous transthoracic needle biopsy (US-PTNB), and to identify modifiable risk factors for associated complications. MATERIALS AND METHODS:We retrospectively analyzed 3043 consecutive US-PTNB procedures performed between July 2019 and September 2021. The pathological diagnostic yield and postprocedural complication rates were documented and analyzed for all 3043 patients. Demographic variables, pulmonary lesion characteristics, and biopsy-related technical variables were investigated to identify potential risk factors associated with complications. RESULTS:The overall diagnostic accuracy of US-PTNB was 94.08% (sensitivity: 93.21%; specificity: 95.48%; positive predictive value: 97.05%; negative predictive value: 89.81%). The overall complication rate was 4.83%. Independent risk factors for increased overall complications included: Benign (odds ratio [OR]=2.292, 95% CI [1.598, 3.287], P<0.001), lower lobe (OR=2.347, 95% CI [1.616, 3.410], P<0.001), lesion diameter ≤3cm (OR=1.993, 95% CI [1.387, 2.864], P<0.001), procedure duration >15 min (OR=3.069, 95% CI [2.134, 4.414], P<0.001), and inexperienced operators(OR=2.227, 95% CI [1.547, 3.206], P<0.001). Independent risk factors for hemoptysis included: Benign (OR=2.479, 95% CI [1.582, 3.884], P<0.001), lower lobe (OR=1.596, 95% CI [1.022, 2.491], P=0.040), procedure duration >15 min (OR=3.237, 95% CI [2.035, 5.149], P<0.001), and inexperienced operators (OR=6.429, 95% CI [3.536,11.690], P<0.001). Independent risk factors for pneumothorax included: Benign (OR=1.998, 95% CI [1.009, 3.958], P=0.047), lower lobe (OR=9.421, 95% CI [2.882, 30.790], P<0.001) and lesion diameter ≤3cm (OR=9.147, 95% CI [4.285, 19.522], P<0.001). CONCLUSION:US-PTNB demonstrates high diagnostic accuracy and a low complication rate. Benign, lower lobe, lesion diameter ≤3cm, procedure duration >15 min, and inexperienced operators were identified as independent risk factors for overall complications. Preprocedural assessment of these risk factors may help reduce the incidence of complications.

With advances in lung cancer management, there is a growing need for larger tissue samples to enable tumor genomic analysis and characterization.This study aims to determine whether the core number obtained during Ultrasound-guided percutaneous lung core needle biopsy(US-PLCNB) is associated with post-procedural complications. This retrospective study enrolled consecutive patients who underwent US-PLCNB for subpleural primary lung cancer at Shanghai Pulmonary Hospital between July 2019 and September 2021. Patient data were extracted from medical records, including demographics, lesion size, and core number. Post-procedural complications, including hemoptysis, pneumothorax, intolerable pain, pleural reaction, hemothorax, and delayed hemopneumothorax, were documented. Multivariate logistic regression models were used to evaluate whether the core number was an independent predictor of complications following US-PLCNB. A total of 1,151 patients (mean age, 64.47 ± 10.58 [SD] years; 278 [24.15

Predicting pedestrian trajectories plays an important role in improving the safety of driverless driving, while pedestrians have extremely high uncertainty. It is difficult to correctly predict trajectories using traditional physics prediction, and deep learning neural networks have more advantages. In order to analyze and improve the prediction of pedestrian crossing trajectory based on deep learning, this paper constructs LSTM network model, TCN network model and Transformer network model, and applies them to a virtual 3D dataset for trajectory prediction, and compares and analyzes the training performance and prediction performance of the three models. Additionally, the Transformer network is enhanced with an autoregressive mechanism, a positional embedding mechanism, and optimized training strategies, resulting in the Transformer model achieving the best performance for pedestrian trajectory prediction on this dataset. This paper summarizes the characteristics of different neural network models in pedestrian trajectory prediction and outlines future research directions.

Objective To analyze Necrotic Area Features of subpleural pulmonary lesions (SPLs) demonstrated by contrast-enhanced ultrasound (CEUS) and investigate their value in differentiating between malignant and benign SPLs. Methods Patients with SPLs who underwent CEUS at our hospital from January to May 2021. The following patient information was recorded: (i) age, (ii) sex, (iii) lesion size, (iv) lesion location, (v) size of necrotic areas and (vi) necrotic area morphology, including sieve-like necrosis, necrotic area with septal enhancement, necrotic area with annular enhancement margins, and necrotic area with burr-like enhancement margins. These parameters were analyzed using univariate and multivariate logistic regression. Subgroup analyses based on lesion size were further conducted using the collected data. Results A total of 212 patients with 212 SPLs were enrolled, comprising 99 benign and 113 malignant cases. Significant differences were observed between malignant and benign groups in terms of age, sex, lesion size and necrotic area morphology (all, p < 0.05). Conclusion Necrotic area's features observed on CEUS were valuable for distinguishing between benign and malignant SPLs. Age, sex, lesion size and the presence of burr-like enhancement margins are identified as independent predictors of malignant lesions.

AimsTo develop and validate an individualized nomogram for differentiating the histologic subtypes (adenocarcinoma and squamous cell carcinoma) of subpleural non-small cell lung cancer (NSCLC) based on ultrasound parameters and clinical data.MethodsThis study was conducted retrospectively between March 2018 and December 2019. Patients were randomly assigned to a development cohort (DC, n=179) and a validation cohort (VC, n=77). A total of 7 clinical parameters and 16 ultrasound parameters were collected. Least absolute shrinkage and selection operator regression analysis was employed to identify the most significant predictors utilizing a 10-fold cross-validation. The multivariate logistic regression model was applied to investigate the relevant factors. An individualized nomogram was then developed. Receiver operating characteristic (ROC) curve, calibration plot and decision curve analysis (DCA) were applied for model validation in both DC and VC.ResultsFollowing the final regression analysis, gender, serum carcinoembryonic antigen, lesion size and perfusion defect in contrast-enhanced ultrasound were entered into the nomogram. The model showed moderate predictive ability, with an area under the ROC curve of 0.867 for DC and 0.838 for VC. The calibration curves of the model showed good agreement between actual and predicted probabilities. The ROC and DCA curves demonstrated that the nomogram exhibited a good predictive performance.ConclusionWe developed a nomogram that can predict the histologic subtypes of subpleural NSCLC. Both internal and external validation revealed optimal discrimination and calibration, indicating that the nomogram may have clinical utility. This model has the potential to assist clinicians in making treatment recommendations.

Objectives: To develop and validate a deep learning (DL) model based on multi-scale features of Lung ultrasound (LUS) and attention mechanism to detect A-line, B-line, pulmonary consolidation, and pleural effusion caused by pulmonary gas-liquid ratio variationsMethods: A total of 6000 LUS images were prospectively collected from 3966 patients, of which 5545 images were selected. All the images were randomly divided into the training set (4,436 images) and the testing set (1,109 images) with a ratio of 4:1. Faced on multi-scale features of LUS, an end-to-end deep learning model based on multi-scale split attention and Mish function was proposed to automatically identify the four LUS features.Results: The overall prediction AUC, accuracy, specificity, and sensitivity of the independent test set were 99.76%, 98.20%, 99.41%, and 98.27%, respectively, and achieved significant and consistent improvement as compared to other deep learning baselines.Conclusions: Our proposed model could interpret the four important LUS features intelligently and be adopted as a support system in the routine diagnosis of an emergency clinician. Significance: This study can not only assist clinicians in recognizing common lung lesions but also provide a new method for the realization of high-quality intelligent diagnosis.

目的:探讨Markov模型在社区骨质疏松人群健康状态预测中的应用.方法:以社区专病数据库中2199名有骨密度数据的人群为对象,年龄40～99岁,平均年龄为(70.97±9.03)岁;其中男性572例,女性1627例.根据骨密度测定结果将其划分为健康、低骨量、骨质疏松三种状态,并进行Markov模型模拟,分析不同骨密度人群未来状态分布的发展趋势.结果:通过建立Markov模型的结果发现,健康转换为低骨量的概率为11.8％,低骨量转化为健康概率为4.2％,低骨量转化为骨质疏松概率为17.7％,骨质疏松转化为低骨量的概率为10.4％.该模型预测未来30年间,低骨量病例所占比例将减少,骨质疏松病例所占比例将增加,健康人数所占比例将基本保持不变.结论:Markov模型适用于骨质疏松健康状态预测研究,对临床工作有实际指导意义.

Objective: This study aims to demonstrate our surgical experience for bronchiectasis and analyze the risk factors related with the surgery outcome. Methods: We retrospectively reviewed medical records of 260 consecutive patients who underwent surgery for bronchiectasis between January 2000 and December 2010. The factors related with the outcome were analyzed and the candidate factors were screened by χ2 test and t test. Furthermore, logistic regression analysis was used for multiple factor analysis to obtain the independent factors that affected the surgical outcome. Results: Complications occurred in 30 (11.5%) patients during perioperative period. The univariate analysis showed that significant differences can be observed in age (P = 0.000), sputum volume (P = 0.000), smoking history (P = 0.033), pulmonary function (P = 0.003), Gram-negative bacillus infection (P = 0.000), bronchial stump coverage (P = 0.016) using intercostals muscles or pedicle pleura embedding and surgical approach (P = 0.003) between the patients with excellent and poor outcome. The multivariable analyses showed that sputum volume (P = 0.000), Gram-negative bacillus infection (P = 0.000) and bronchial stump coverage (P = 0.000) were the three independent factors related with surgical outcome. Conclusion: Surgery is an effective treatment option for bronchiectasis. Sputum volumes lower than 30 mL, negative proof of Gram-negative bacteria and bronchial stump coverage using intercostals muscles or pedicle pleura embedding are the key factors for successful treatment. Special attention has to be given to any complications in elderly patients.

BACKGROUND:To determine the current role of surgery and the risks and operative outcomes in patients with massive hemoptysis caused by pulmonary tuberculosis. METHODS:A retrospective review of patients treated in the Department of Thoracic Surgery, Shanghai Pulmonary Disease Hospital between January 2001 and December 2010 was undertaken. The inclusion criteria were massive hemoptysis defined as expectoration of at least 200 mL of blood on each occasion or more than 600 mL in 24 hours; anatomic major pulmonary resection including lobectomy, bilobectomy, and pneumonectomy; histologic or bacteriologic pulmonary tuberculosis. RESULTS:Sixty-eight men (76.4%) and 21 women (23.6%) aged 41.3 ± 11.7 years met the criteria. The cumulative amount of bleeding on admission was 558 ± 272 mL. A total of 36 patients (40.4%) underwent an emergency operation and 53 (59.6%) had a delayed operation. The operative morbidity rate was 31.5% (28 of 89) and mortality was 2.2% (2 of 89). Only 2 (2.2%) of the 89 patients had a recurrent episode of hemoptysis. Multivariate analysis showed that patients who received antituberculous therapy before surgery (odds ratio [OR], 0.12; 95% confidence interval [CI], 0.04-0.38; P = .0003) had a decreased risk of developing postoperative complications and that patients who underwent an emergency operation (OR, 3.9; 95% CI, 1.29-11.5; P = .0154) had an increased risk. CONCLUSIONS:In the present study, the immediate and long-term outcomes after surgery for patients with massive hemoptysis caused by pulmonary tuberculosis were fairly satisfactory. The morbidity and mortality were also acceptable. However, delayed elective surgery is preferred and emergency surgery should be reserved for patients with persistent life-threatening hemoptysis despite a multidisciplinary intervention.

目的分析含左氧氟沙星四联方案补救治疗幽门螺杆菌的临床疗效。方法对50例幽门螺杆菌感染患者的临床资料进行回顾性分析。结果和对照组相比,观察组幽门螺杆菌根除率和溃疡愈合率较高,但无显著性差异(P>0.05);观察组不良反应发生率为32%,对照组为56%,观察组不良反应发生率明显低于对照组(P<0.05);观察组总有效率为92%,对照组为80%,2组有显著性差异(P<0.05)。结论含左氧氟沙星四联方案补救治疗幽门螺杆菌能够有效降低患者的不良反应发生率,提升患者治疗的总有效率,具有良好的治疗效果,值得在临床广为推广。

Objective The purpose of the present study was to analyses video-assisted thoracic surgery (VATS) lobectomy for early lung cancer and to provide evidence in guiding clinical practice.Methods From May 1997 to October 2009,VATS lobectomy for early lung cancer was performed in 518 patients.All patient data was reviewed retrospectively.Results The data group consisted of 297 male patients and 221 female patients with a mean age of (58.9 ± 10.6) years ( 19 - 89 years).Morbidity was 10.8％ and mortality was 0.4％.Multivariate analysis identified patient’ s age ( P =0.0300,OR =2.0148,95％ CI 1.0700-3.7940) and operation duration (P=0.0007,OR =1.0086,95％ CI 1.0036-1.0136) as the statistically significant predictors of postoperative complications.Overall 1,3 and 5-year survival rates were 98％,81％ and 66％.And postoperative patbological staging ( P =0.0036,OR =1.6071,95 ％ CI 1.1677 -2.2118 ) is a prognostic determinant.Conclusion VATS lobectomy is a safe and effective therapeutic method for early lung cancer.However,patient selection plays key role in VATS.Operation duration should be shortened as possible,otherwise,it may result in increased postoperative morbidity.It is important to deal with the accident situation rationally and converse to thoracotomy decidedly if necessary.

Aspiration via a congenital broncho-esophageal fistula in an adult thoracotomy patient has not been previously reported. Repeated aspiration and subsequent respiratory failure if the fistula is not recognized could be life-threatening in these postoperative patients. We describe one such critical case, in which a broncho-esophageal fistula was discovered weeks after aspiration and the onset of respiratory failure after left lower lobectomy. This unusual case suggests that repeated localized pulmonary infections can indicate a broncho-esophageal fistula and that further investigations should be performed, including detailed history.

To alleviate increasingly worsening urban transportation problems and promote optimal travel mode,it is necessary to study the relationship between urban land use and transportation system.Taking Shenzhen City as an example,this paper develops a multinomial logit(MNL) model using land use and travelers’household,social and economic attributes as variables,and analyzes the impact of these variables on commuting and non-commuting travel mode choice.The analysis results indicate that the location of origin is an important factor in choosing travel mode.Comparing travelers from centers of communities and from suburbs,it is clear that trip-makers in urban centers and sub-centers are more likely to travel by public transportation.Finally,the paper provides suggestions on how to promote optimal travel modes through urban planning and design.