Background
Computed tomography (CT) is essential for pulmonary nodule detection in diagnosing lung cancer. As deep learning algorithms have recently been regarded as a promising technique in medical ...fields, we attempt to integrate a well‐trained deep learning algorithm to detect and classify pulmonary nodules derived from clinical CT images.
Materials and Methods
Open‐source data sets and multicenter data sets have been used in this study. A three‐dimensional convolutional neural network (CNN) was designed to detect pulmonary nodules and classify them into malignant or benign diseases based on pathologically and laboratory proven results.
Results
The sensitivity and specificity of this well‐trained model were found to be 84.4% (95% confidence interval CI, 80.5%–88.3%) and 83.0% (95% CI, 79.5%–86.5%), respectively. Subgroup analysis of smaller nodules (<10 mm) have demonstrated remarkable sensitivity and specificity, similar to that of larger nodules (10–30 mm). Additional model validation was implemented by comparing manual assessments done by different ranks of doctors with those performed by three‐dimensional CNN. The results show that the performance of the CNN model was superior to manual assessment.
Conclusion
Under the companion diagnostics, the three‐dimensional CNN with a deep learning algorithm may assist radiologists in the future by providing accurate and timely information for diagnosing pulmonary nodules in regular clinical practices.
Implications for Practice
The three‐dimensional convolutional neural network described in this article demonstrated both high sensitivity and high specificity in classifying pulmonary nodules regardless of diameters as well as superiority compared with manual assessment. Although it still warrants further improvement and validation in larger screening cohorts, its clinical application could definitely facilitate and assist doctors in clinical practice.
摘要
背景。在肺癌的诊断中,计算机断层扫描 (CT) 对于肺结节的检测必不可少。近几年,随着医学领域逐渐认识到深度学习算法这种技术的价值,本研究试图集成一种训练有素的深度学习算法,对临床 CT 图像中的肺结节进行检测和分类。
材料和方法。本研究使用了开源数据集和多中心数据集。本文设计了一种三维卷积神经网络 (CNN) 来检测肺结节,然后根据病理和实验室证实的结果,判断为恶性或良性结节。
结果。这种训练有素的模型敏感性和特异性分别为 84.4% 95% 可信区间 (CI), 80.5%‐88.3%和83.0%(95% CI,79.5%‐86.5%)。小结节 (< 10mm) 亚组分析显示的敏感性和特异性显著,与大结节 (10‐30mm) 相似。对比不同级别医生的人工评估结果与三维 CNN 的评估结果,进行了额外的模型验证。结果表明,CNN 模型的表现优于人工评估。
结论。通过伴随诊断可知,加入深度学习算法的三维 CNN 能够提供准确、及时的信息,有助于放射科医生在常规临床实践中的肺结节诊断工作。
实践意义:在对各种直径的肺结节分类中,本文所述的三维卷积神经网络具有较高的敏感性和特异性,与人工评估结果相比具有优越性。虽然仍需在更大的筛选队列中进行进一步改进和验证,但可以肯定的是,临床应用三维卷积神经网络可以促进和协助医生的临床实践工作。
Interest in deep convolutional neural networks (CNN) is growing because of demonstrated accuracy with less manual intervention in computer vision tasks. This article describes efforts to use a pre‐trained CNN model integrating with multi‐centers datasets for detection and classification of pulmonary nodules.
We introduce two pioneering applications leveraging Distributed Fiber Optic Sensing (DFOS) and Machine Learning (ML) technologies. These innovations offer substantial benefits for fortifying telecom ...infrastructures and public safety. By harnessing existing telecom cables, our solutions excel in perimeter intrusion detection via buried cables and impulsive event classification through aerial cables. To achieve comprehensive intrusion detection, we introduce a label encoding strategy for multitask learning and evaluate the generalization performance of the proposed approach across various domain shifts. For accurate recognition of impulsive acoustic events, we compare several standard choices of representations for raw waveform data and neural network architectures, including convolutional neural networks (ConvNets) and vision transformers (ViT). We also study the effectiveness of the built-in inductive biases under both high- and low-fidelity sensing conditions and varying amounts of labeled training data. All computations are executed locally through edge computing, ensuring real-time detection capabilities. Furthermore, our proposed system seamlessly integrates with cameras for video analytics, significantly enhancing overall situation awareness of the surrounding environment.
We demonstrate a vibration detection and localization scheme based on bidirectional transmission of telecom signals with digital coherent detection at the receivers. Optical phase is extracted from ...the digital signal processing blocks of the coherent receiver, from which the vibration component is extracted by bandpass filtering, and the position along the cable closest to the vibration's epicenter is recovered by correlation. We demonstrate our scheme first using offline experiment with 200-Gb/s DP-16QAM, and we report field trial results over installed fiber to detect real-world vibration events.
Abstract
Sepsis is a common complication of combat injuries and trauma, and is defined as a life-threatening organ dysfunction caused by a dysregulated host response to infection. It is also one of ...the significant causes of death and increased health care costs in modern intensive care units. The use of antibiotics, fluid resuscitation, and organ support therapy have limited prognostic impact in patients with sepsis. Although its pathophysiology remains elusive, immunosuppression is now recognized as one of the major causes of septic death. Sepsis-induced immunosuppression is resulted from disruption of immune homeostasis. It is characterized by the release of anti-inflammatory cytokines, abnormal death of immune effector cells, hyperproliferation of immune suppressor cells, and expression of immune checkpoints. By targeting immunosuppression, especially with immune checkpoint inhibitors, preclinical studies have demonstrated the reversal of immunocyte dysfunctions and established host resistance. Here, we comprehensively discuss recent findings on the mechanisms, regulation and biomarkers of sepsis-induced immunosuppression and highlight their implications for developing effective strategies to treat patients with septic shock.
Electronic-cigarettes (E-cigs) are marketed as a safe alternative to tobacco to deliver the stimulant nicotine, and their use is gaining in popularity, particularly among the younger population. We ...recently showed that mice exposed to short-term (12 wk) E-cig smoke (ECS) sustained extensive DNA damage in lungs, heart, and bladder mucosa and diminished DNA repair in lungs. Nicotine and its nitrosation product, nicotine-derived nitrosamine ketone, cause the same deleterious effects in human lung epithelial and bladder urothelial cells. These findings raise the possibility that ECS is a lung and bladder carcinogen in addition to nicotine. Given the fact that E-cig use has become popular in the past decade, epidemiological data on the relationship between ECS and human cancer may not be known for a decade to come. In this study, the carcinogenicity of ECS was tested in mice. We found that mice exposed to ECS for 54 wk developed lung adenocarcinomas (9 of 40 mice, 22.5%) and bladder urothelial hyperplasia (23 of 40 mice, 57.5%). These lesions were extremely rare in mice exposed to vehicle control or filtered air. Current observations that ECS induces lung adenocarcinomas and bladder urothelial hyperplasia, combined with our previous findings that ECS induces DNA damage in the lungs and bladder and inhibits DNA repair in lung tissues, implicate ECS as a lung and potential bladder carcinogen in mice. While it is well established that tobacco smoke poses a huge threat to human health, whether ECS poses any threat to humans is not yet known and warrants careful investigation.
This study aimed to challenge chemoresistance by curcumin (CUR) with drug-selected human lung cancer A549 sublines that continuously proliferate in the present of docetaxel (DOC) and vincristine ...(VCR). Their sensitivities to CUR were measured by MTT assay and the particular intracellular reactive oxygen species (ROS) was detected by fluorescence activated cell sorting (FACS) analysis. Apoptosis was analyzed by Annexin V assay of the flow cytometry. Inhibitors and RNA interference were used to examine the signaling pathway regulated by the kinases. The obtained data demonstrated that CUR induces chemoresistant cell apoptosis by generating ROS and application of N-acetylcysteine (NAC) blocks ROS production, resulting in apoptosis suppression. Phosphorylation of extracellular regulated kinase (ERK), p38 MAPK, and eIF-2α were increased but c-Jun N-terminal kinase (JNK) did not increase when chemoresistant cells were treated with CUR. Downregulation of ERK and p38 MAPK phosphorylation by their inhibitors had no effect on CUR-induced apoptosis. Interestingly, the knockdown of p38 MAPK with shRNA significantly reduced CUR-induced apoptosis on the chemoresistant sublines. Phosphorylation of the eIF-2α protein was inhibited when p38 MAPK was knocked down in DOC-resistant A549 cells, but a high level of phosphorylated eIF-2α protein remained on the VCR-resistant A549 cells when p38 MAPK was knocked down. These data confirmed that CUR-augmented ROS potently induced apoptosis via upregulated p38 MAPK phosphorylation. Therefore, activated p38 MAPK is considered a pro-apoptotic signal for CUR-induced apoptosis of chemoresistant human lung cancer cells.
Abstract Background Increasing evidence supports the use of magnetic resonance (MR)–targeted prostate biopsy. The optimal method for such biopsy remains undefined, however. Objective To prospectively ...compare targeted biopsy outcomes between MR imaging (MRI)–ultrasound fusion and visual targeting. Design, setting, and participants From June 2012 to March 2013, prospective targeted biopsy was performed in 125 consecutive men with suspicious regions identified on prebiopsy 3-T MRI consisting of T2-weighted, diffusion-weighted, and dynamic-contrast enhanced sequences. Intervention Two MRI–ultrasound fusion targeted cores per target were performed by one operator using the ei-Nav|Artemis system. Targets were then blinded, and a second operator took two visually targeted cores and a 12-core biopsy. Outcome measurements and statistical analysis Biopsy information yield was compared between targeting techniques and to 12-core biopsy. Results were analyzed using the McNemar test. Multivariate analysis was performed using binomial logistic regression. Results and limitations Among 172 targets, fusion biopsy detected 55 (32.0%) cancers and 35 (20.3%) Gleason sum ≥7 cancers compared with 46 (26.7%) and 26 (15.1%), respectively, using visual targeting ( p = 0.1374, p = 0.0523). Fusion biopsy provided informative nonbenign histology in 77 targets compared with 60 by visual ( p = 0.0104). Targeted biopsy detected 75.0% of all clinically significant cancers and 86.4% of Gleason sum ≥7 cancers detected on standard biopsy. On multivariate analysis, fusion performed best among smaller targets. The study is limited by lack of comparison with whole-gland specimens and sample size. Furthermore, cancer detection on visual targeting is likely higher than in community settings, where experience with this technique may be limited. Conclusions Fusion biopsy was more often histologically informative than visual targeting but did not increase cancer detection. A trend toward increased detection with fusion biopsy was observed across all study subsets, suggesting a need for a larger study size. Fusion targeting improved accuracy for smaller lesions. Its use may reduce the learning curve necessary for visual targeting and improve community adoption of MR-targeted biopsy.
We have proposed and experimentally demonstrated a novel architecture for orthogonal frequency-division- multiplexing (OFDM) wavelength-division-multiplexing passive optical network with centralized ...lightwave. In this architecture, 16 quadrature amplitude modulation intensity-modulated OFDM signals at 10 Gb/s are utilized for downstream transmission. A wavelength-reuse scheme is employed to carry the upstream data to reduce the cost at optical network unit. By using one intensity modulator, the downstream signal is remodulated for upstream on-off keying (OOK) data at 2.5 Gb/s based on its return-to-zero shape waveform. We have also studied the fading effect caused by double-sideband (DSB) downstream signals. Measurement results show that 2.5-dB power penalty is caused by the fading effect. The fading effect can be removed when the DSB OFDM downstream signals are converted to single sideband (SSB) after vestigial filtering. The power penalty is negligible for both SSB OFDM downstream and the remodulated OOK upstream signals after over 25-km standard single-mode-fiber transmission. Index
Cinnamic acid is one of the phenolic compounds that is isolated from cinnamon, or other natural plants, and has a wide range of physiological activities. However, the application of cinnamic acid is ...limited due to its poor solubility and low oral bioavailability. In this study, the feasibility of producing octyl cinnamate by ultrasonic assistance, combined with a rotary evaporation under vacuum, was studied using methyl cinnamate and octanol as the starting materials. A Box–Behnken design (BBD) was employed to evaluate the effects of the operation parameters, including reaction temperature (55–75 °C), reaction time (4–12 h), and ultrasonic power (90–150 W) on the production of octyl cinnamate. Meanwhile, the synthesis process was further optimized by the modeling response surface methodology (RSM). The data indicated that octyl cinnamate was efficiently synthesized from methyl cinnamate and octanol using the ultrasound plus vacuum system; further, this system was superior to the conventional method. According to the RSM model for the actual experiments, a reaction temperature of 74.6 °C, a reaction time of 11.1 h, and an ultrasound power of 150 W were determined to be the best conditions for the maximum molar conversion of octyl cinnamate (93.8%). In conclusion, the highly efficient synthesis of octyl cinnamate by a rotary evaporator with an ultrasound plus vacuum system was achieved via RSM optimization.
We review recent advances in distributed fiber optic sensing (DFOS) and their applications. The scattering mechanisms in glass, which are exploited for reflectometry-based DFOS, are Rayleigh, ...Brillouin, and Raman scatterings. These are sensitive to either strain and/or temperature, allowing optical fiber cables to monitor their ambient environment in addition to their conventional role as a medium for telecommunications. Recently, DFOS leveraged technologies developed for telecommunications, such as coherent detection, digital signal processing, coding, and spatial/frequency diversity, to achieve improved performance in terms of measurand resolution, reach, spatial resolution, and bandwidth. We review the theory and architecture of commonly used DFOS methods. We provide recent experimental and field trial results where DFOS was used in wide-ranging applications, such as geohazard monitoring, seismic monitoring, traffic monitoring, and infrastructure health monitoring. Events of interest often have unique signatures either in the spatial, temporal, frequency, or wavenumber domains. Based on the temperature and strain raw data obtained from DFOS, downstream postprocessing allows the detection, classification, and localization of events. Combining DFOS with machine learning methods, it is possible to realize complete sensor systems that are compact, low cost, and can operate in harsh environments and difficult-to-access locations, facilitating increased public safety and smarter cities.