Sample Dataset Research Articles

305. PBMC-Derived Transcriptomic Signatures Accurately Discriminate Between Viral, Bacterial, and Fungal Infections and can be Translated to Real-World Human Infections

Abstract Background Analysis of host gene expression patterns (‘signatures’) can provide diagnostic information to determine the etiology of acute infection. In vitro experiments can be used to supplement our knowledge about these signatures. Methods We performed in vitro human peripheral blood mononuclear cell (PBMC) challenges with influenza virus, gram-negative (Escherichia coli) and gram-positive (Streptococcus pneumoniae) bacteria, and fungal pathogens (Candida albicans, Cryptococcus neoformans and gattii). Exposed human cells were analyzed for differential gene expression utilizing Affymetrix microarrays. Logistic regression models were applied to the gene expression data to identify signatures capable of identifying each infection etiology. PBMC-derived signatures for each pathogen class were then applied to a microarray dataset of peripheral blood samples from human patients with candidemia. Results A 41-gene multinomial classifier was developed which correctly classified viral, bacterial, or fungal infection with 94-98% accuracy. Gene expression patterns in PBMC challenges show substantial overlap with gene expression changes in natural infection. For fungal infection, 31 of the top 50 differentially expressed genes (DEGs) from the PBMC challenge overlapped with responses seen in patients with candidemia. Twenty of these genes were upregulated, including those involved in IgG binding (FCGR1A and FCGR1B). A subset of these genes were able to accurately identify human cases of candidemia in an independent dataset with an area under the receiver operating characteristic curve (auROC) of 0.94. Pathway analysis of differentially expressed genes revealed pathogen class-specific immune responses. For fungal infections in particular, a marked upregulation of inflammatory responses was seen including cellular responses to stress, eosinophil chemotaxis (CCL24, SOCS1/2) and T cell proliferation and migration (IL2, CXCL9). Conclusion In vitro PBMC challenges can support ability to generate host transcriptomic signatures able to discriminate between bacterial, viral, and fungal infection, and their performance can be translated to real-world human infections. Disclosures Julie M. Steinbrink, MD, MHS, Duke University: Methods to Detect and Treat a Fungal Infection Ricardo Henao, PhD, Duke University: Methods to Diagnose and Treat Acute Respiratory Infections Ephraim L. Tsalik, MD, PhD, Danaher Diagnostics: VP and Chief Scientific Officer for Infectious Disease|Duke University: Methods to Diagnose and Treat Acute Respiratory Infections; Methods to Detect and Treat a Fungal Infection Geoffrey S. Ginsburg, MD, PhD, Duke University: Methods to Diagnose and Treat Acute Respiratory Infections Elizabeth Ramsburg, PhD, Spark Therapeutics: Head Of Central Nervous System & Research Christopher W. Woods, MD, MPH, Biomeme Inc: Methods to diagnose and treat acute respiratory infections|Biomeme Inc: Chief Medical Consultant|Biomeme Inc: Ownership Interest Micah T. McClain, MD, PhD, Biomeme Inc: Methods to diagnose and treat acute respiratory infections

DDoS2Vec: Flow-Level Characterisation of Volumetric DDoS Attacks at Scale

Volumetric Distributed Denial of Service (DDoS) attacks have been a severe threat to the Internet for more than two decades. Some success in mitigation has been achieved based on numerous defensive techniques created by the research community, implemented by the industry, and deployed by network operators. However, evolution is not a privilege of mitigations, and DDoS attackers have found better strategies and continue to cause harm. A key challenge in winning this race is understanding the various characteristics of DDoS attacks in network traffic at scale and in a realistic manner. In this paper, we propose DDoS2Vec, a novel approach to characterise DDoS attacks in real-world Internet traffic using Natural Language Processing (NLP) techniques. DDoS2Vec is a domain-specific application of Latent Semantic Analysis that learns vector representations of potential DDoS attacks. We look into the link between natural language and computer network communication in a way that has not been previously studied. Our approach is evaluated on a large-scale dataset of flow samples collected from an Internet eXchange Point (IXP) in one year. We evaluate the performance of DDoS2Vec via multi-label classification in a Machine Learning (ML) scenario. DDoS2Vec characterises DDoS attacks more clearly than other baselines - including NLP-based approaches inspired by recent networks research and a basic non-NLP solution.

Unsupervised domain adaptation for SAR target classification based on domain- and class-level alignment: From simulated to real data

Synthetic Aperture Radar (SAR) target classification methods based on convolutional neural networks have attracted wide attention recently. Typically, these methods necessitate a substantial quantity of labeled data to train the deep model, which is an obstacle for SAR target classification since annotating SAR images is costly and time-consuming. Generating large-scale labeled data in simulation to train the deep model is a feasible alternative. However, cross-domain recognition is challenging due to significant differences in distribution between simulated and real domains. In this paper, we propose an unsupervised domain adaptation method for SAR target classification by aligning the distribution between the simulated and unlabeled real domains. Specifically, we encode the dependencies across different granularity perspectives including domain-, and class-levels simultaneously to align two domains finely. On the one hand, a gradient-weighted adversarial alignment method is proposed to align the global distributions at the domain-level. On the other hand, a novel class-level alignment method based on the prototypical network is proposed to align the finer class structure. Technically, we respectively model the cross-domain prototype-prototype relations and cross-domain prototype-instance relations with contrastive learning to achieve intra-class prototype compact as well as inter-class prototype separation among cross-domain categories, where the prototypes of the real domain are learned with pseudo labels. To improve the quality of pseudo-labels, a pseudo-label filtering strategy combined with the attribute scattering center is further designed. The experiments on two simulation-to-reality datasets indicate that the proposed method outperforms state-of-the-art unsupervised domain adaptive methods, such as pixelfeat (+4.45% and +7.6% in accuracy on SAMPLE and S2M datasets respectively). Code and data will be available in https://github.com/YuShi1213/Sim2Real-Unsupervised-SAR-Target-Classification.

Auto-metric graph neural network for paddy leaf disease classification

In traditional methods, rice plant diseases can be identified individually by experts, but laboratory testing takes a lot of time. As a result, it reduces agricultural productivity and causes economic losses for farmers. To address these, rapid and effective systems for identifying and classifying rice plant diseases need to be developed. This article proposes an Auto-metric Graph Neural Network (AGNN) for the classification of paddy leaf disease. Initially, images are gathered via the dataset of Rice Leaf Disease Image Samples. The collected images are preprocessing by utilizing Anisotropic Diffusion Filter-Based Unsharp Masking with crispening. These preprocessed paddy leaf images are supplied to Bayesian fuzzy clustering for the purpose of segmentation. The segmented pictures are fed to AGNN for categorizing paddy leaf infection image as Bacterial Blight, Blast, Brown Spot and Tungro. The paddy leaf disorder classification-AGNN technique attains accuracies of 8.65%, 10.76%, 18.34% and 13.19% compared to the existing methods.

Smoke recognition in steelmaking converter images: Class-distance-based feature selection model

Controlling the smoke and dust generated in steelmaking efficiently and precisely is a significant task for green metallurgy. This study aims at recognizing the smoke and dust generated from converters in steelmaking processes, to reach this goal, a research about feature generation and selection is conducted. The comprehensive feature datasets include GLCM features, LBP features and gray statistics features as a feature vector of 212 dimension for each image. To obtain better classification accuracy and performance, a supervised Class-Distance-based Feature Selection (CDFS) algorithm is proposed which involved the Mahalanobis distance which is chosen as class distance, correlation, T-test and train with machine learning classifier SVM. Then classifying images as smoke or non-smoke, the best feature combination with high classification performance as well as smoke images classification model can be exported as an output. Applied results on 5 converter image datasets show that this feature selection algorithm after optimization of Z-Score data standardization can not only reach an excellent accuracy in range of 92.54%-99.18% and AUC from 0.88 to 0.99 with feature size of 6–12 dimension, but apply well on small sample and imbalance class datasets. Compared with PSO, GA and BA, CDFS obtain higher accuracies as well as lower feature size. An application result also proved the effectiveness and generalization of this algorithm which means it can achieve precise classification accuracies with less cost in training as well. It can provide the foundation for efficient and precise emission treatment of converters in steelmaking.

Warming and wetting-induced soil acidification triggers methanotrophic diversity loss and species turnover in an alpine ecosystem

Increases in temperature and precipitation have continually occurred in the past few decades, both globally and in China, which have probably impacted the dynamics of net methane (CH4) emissions by regulating soil CH4 oxidization. An alpine ecosystem is quite sensitive to climate changes, but how these two climatic factors regulate the community of soil aerobic methane-oxidizing bacteria (aMOB) remains elusive in such an ecosystem because of the lack of high-resolution soil sampling datasets. Here, we employed a hybrid approach of gridded field surveys and random forest prediction to discern the above question in the Qinghai-Tibetan Plateau (QTP). Results indicated that type Ic aMOBs exhibited a preference for drier, colder areas, while type IIa, TUSC-like and RA21-like aMOBs probably occurred in wetter and relatively warmer areas. There were no remarkable shifts in α-diversity and bacterial community compositions from the driest and coldest regions to moderate ones, while the α-diversity decreased and community compositions changed substantially from moderate arid and cold areas to wetter, relatively warmer ones. These trends were mainly regulated by soil pH, a potentially secondary effect shaped by the interactions of mean annual precipitation (MAP) and temperature (MAT) and an aridity index (AI) in the QTP. A slight decline in soil pH from the driest and coldest areas to moderate ones might regulate aMOB diversity and community structure by the stochastic extinction of individuals adapted to a higher pH and the unstable colonization of individuals preferring a moderate pH. A great decline in soil pH from the moderate arid, cold to wetter, relatively warmer areas might lead to the deterministic exclusion of species with a high pH adaptability and the stable colonization of species preferring the acidic habitat. This study implies that soil acidification probably induces aMOB diversity loss and species turnover in the arid, cold areas of the QTP under climatic warming and wetting, shedding a new light on projecting the impact of climate changes on an alpine ecosystem’s structure and function on unprecedented spatial scales.

Multivariate multiscale dispersion Lempel–Ziv complexity for fault diagnosis of machinery with multiple channels

Lempel–Ziv complexity (LZC), as a nonlinear feature in information science, has shown great promise in detecting correlations and capturing dynamic changes in single-channel time series. However, its application to multichannel data has been largely unexplored, while the complexity of real-world systems demands the utilization of data collected from multiple sensors or channels so as to extract distinguishable fault features for fault diagnosis. This paper proposes a novel method called multivariate multiscale dispersion Lempel–Ziv complexity (mvMDLZC) to extract the fault features hidden in multi-source information. First, multivariate embedding theory is applied to obtain multivariate embedded vectors and multivariate dispersion patterns, which can reflect the inherent relationships in the multichannel series. Second, by assigning labels to these patterns, the original multichannel time series can be transformed into a symbolic sequence with multiple symbols instead of the original binary conversion, enabling the accurate recovery of the system dynamics. Finally, the complexity counter value and normalized LZC are calculated for the complexity measure. Experimental results using synthetic and real-world datasets demonstrate that mvMDLZC outperforms existing LZC-based methods and multivariate dispersion entropy in recognizing different states of mechanical systems. Additionally, mvMDLZC exhibits robustness in handling challenges such as small sample datasets and noise interference, making it suitable for real industrial applications. These findings highlight the potential of mvMDLZC as a valuable approach for dissecting multichannel systems across various real-world scenarios.

Unraveling Molecular Composition in Biological Samples-Systematic Evaluation of Statistical Methods for the Analysis of Hyperspectral Raman Data.

Recently, confocal Raman microscopy has gained popularity in biomedical research for studying tissues in healthy and diseased state due to its ability to acquire chemically selective data in a noninvasive approach. However, biological samples, such as brain tissue, are inherently difficult to analyze due to the superposition of molecules in the Raman spectra and low variation of spectral features within the sample. The analysis is further impeded by pathological hallmarks, for example beta-amyloid (Aβ) plaques in Alzheimer's disease, which are often solely characterized by subtle shifts in the respective Raman peaks. To unravel the underlying molecular information, convoluted statistical procedures are inevitable. Unfortunately, such statistical methods are often inadequately described, and most natural scientists lack knowledge of their appropriate use, causing unreproducible results and stagnation in the application of hyperspectral Raman imaging. Therefore, we have set out to provide a comprehensive guide to address these challenges with the example of a complex hyperspectral data set of brain tissue samples with Aβ plaques. Our study encompasses established as well as novel statistical methods, including univariate analysis, principal component analysis, cluster analysis, spectral unmixing, and 2D correlation spectroscopy, and critically compares the outcomes of each analysis. Moreover, we transparently demonstrate the effect of preprocessing decisions like denoising and scaling techniques, providing valuable insights into implications of spectral quality for data evaluation. Thereby, this study provides a comprehensive evaluation of analysis approaches for complex hyperspectral Raman data, laying out a blueprint for elucidating meaningful information from biological samples in chemical imaging.

Label-Free CD34+ Cell Identification Using Deep Learning and Lens-Free Shadow Imaging Technology.

Accurate and efficient classification and quantification of CD34+ cells are essential for the diagnosis and monitoring of leukemia. Current methods, such as flow cytometry, are complex, time-consuming, and require specialized expertise and equipment. This study proposes a novel approach for the label-free identification of CD34+ cells using a deep learning model and lens-free shadow imaging technology (LSIT). LSIT is a portable and user-friendly technique that eliminates the need for cell staining, enhances accessibility to nonexperts, and reduces the risk of sample degradation. The study involved three phases: sample preparation, dataset generation, and data analysis. Bone marrow and peripheral blood samples were collected from leukemia patients, and mononuclear cells were isolated using Ficoll density gradient centrifugation. The samples were then injected into a cell chip and analyzed using a proprietary LSIT-based device (Cellytics). A robust dataset was generated, and a custom AlexNet deep learning model was meticulously trained to distinguish CD34+ from non-CD34+ cells using the dataset. The model achieved a high accuracy in identifying CD34+ cells from 1929 bone marrow cell images, with training and validation accuracies of 97.3% and 96.2%, respectively. The customized AlexNet model outperformed the Vgg16 and ResNet50 models. It also demonstrated a strong correlation with the standard fluorescence-activated cell sorting (FACS) technique for quantifying CD34+ cells across 13 patient samples, yielding a coefficient of determination of 0.81. Bland-Altman analysis confirmed the model's reliability, with a mean bias of -2.29 and 95% limits of agreement between 18.49 and -23.07. This deep-learning-powered LSIT offers a groundbreaking approach to detecting CD34+ cells without the need for cell staining, facilitating rapid CD34+ cell classification, even by individuals without prior expertise.

Iron inhibits glioblastoma cell migration and polarization.

Glioblastoma is one of the deadliest malignancies facing modern oncology today. The ability of glioblastoma cells to diffusely spread into neighboring healthy brain makes complete surgical resection nearly impossible and contributes to the recurrent disease faced by most patients. Although research into the impact of iron on glioblastoma has addressed proliferation, there has been little investigation into how cellular iron impacts the ability of glioblastoma cells to migrate-a key question, especially in the context of the diffuse spread observed in these tumors. Herein, we show that increasing cellular iron content results in decreased migratory capacity of human glioblastoma cells. The decrease in migratory capacity was accompanied by a decrease in cellular polarization in the direction of movement. Expression of CDC42, a Rho GTPase that is essential for both cellular migration and establishment of polarity in the direction of cell movement, was reduced upon iron treatment. We then analyzed a single-cell RNA-seq dataset of human glioblastoma samples and found that cells at the tumor periphery had a gene signature that is consistent with having lower levels of cellular iron. Altogether, our results suggest that cellular iron content is impacting glioblastoma cell migratory capacity and that cells with higher iron levels exhibit reduced motility.

The Path of Improving the Effectiveness of Class Teachers’ Parenting in Private Colleges and Universities Based on Decision Tree Models

Abstract This paper firstly researches the path to enhance the effectiveness of class teacher’s collaborative parenting in private colleges and universities, mainly studies the two collaborative parenting modes of “Counselor + Freshman Class Teacher” and “Class Teacher + Academic Mentor” and summarizes the roles and main links in the two modes. The two modes’ roles and subject connections are summarized. On the basis of the second decision tree, the C4.5 algorithm is used to calculate the sample data set and then get the results of student characteristic information gain, and then the decision tree is generated according to the results of student characteristic information gain, disaggregated attributes and split points. Then, the student characteristic information gain is used as the selection measure of attributes, and then the decision tree-based classroom teacher teaching and educating model is constructed, and the effectiveness of classroom teacher education in private colleges and universities is statistically analyzed. The results show that there are significant differences in the dimensions of caring for students [F(2, 86)=5.58, P=0.005] and instructing students’ ability [F(2, 86)=3.42, P=0.038] among classroom teachers of different classroom teacher’s tenure in private colleges and universities, and there are no significant differences in the other dimensions of educating people’s ability. This research knot promotes the improvement of classroom teachers’ parenting ability in private colleges and universities, which is important for the improvement of students’ learning ability.

Deep Learning‐Based Bluetooth Low‐Energy 5.1 Multianchor Indoor Positioning with Attentional Data Filtering

Indoor positioning system (IPS) technologies have widespread applications in logistics, intelligent manufacturing, healthcare monitoring, etc. The recently released Bluetooth low‐energy (BLE) 5.1 specification enables in‐phase and quadrature‐phase (I/Q) data measurements. It allows angle of arrival estimation and becomes a natural choice for IPS implementation. Conventional BLE 5.1 IPSs use multiple anchors to provide massive redundancy to improve system robustness. It however demands effective approaches to leverage redundancy. Besides, interference due to various environmental factors can introduce severe errors to I/Q data and affect positioning accuracy. Facing these challenges, herein, a novel deep learning‐based multianchor BLE 5.1 IPS is proposed. The system aggregates measurements from multiple anchors and makes them available at regular time steps. Then, a novel attentional filtering network tailored to infer high‐quality I/Q sample data is developed and a spatial regularization loss incorporating spatial location relationships to strengthen the feature embedding discrimination is proposed. Two multianchor BLE 5.1 I/Q sample datasets are developed and released for public download. Numerical experiments are carried out to compare the proposed method with previous BLE 5.1 IPS methods and methods utilizing other radio frequency data. Results indicate that the proposed method consistently achieves submeter accuracy and significantly outperforms the state‐of‐the‐art approaches.

Cumulative spectrum distribution entropy for rotating machinery fault diagnosis

Entropy-based methods have shown promise in detecting dynamic changes in non-linear signals and have been widely applied in fault diagnosis for rotating machinery. However, these methods have limitations when it comes to capturing frequency-domain information of fault features, as they are primarily based on time-domain signals. To address this issue, this paper proposes a new entropy measure called cumulative spectrum distribution entropy (CSDEn), which is based on the cumulative distribution of the spectrum and considers both frequency probability and frequency values in the spectrum domain. The proposed method is evaluated using synthetic signals and experimental data from different bearing and gear working states. The results show that CSDEn outperforms other widely used entropy measures in detecting dynamic changes and measuring signal complexity with low noise sensitivity and high computing efficiency. Nonparametric Mann–Whitney U tests reveal significant differences between different working states for proposed CSDEn method, and compared with other entropy methods, CSDEn achieves the highest recognition rates in diagnosing different bearing and gear working states. Moreover, proposed CSDEn method demonstrates its effectiveness in addressing the challenges of small sample datasets and strong noise interference, making it highly competitive in real industrial applications.

Flow Prediction of a Measurement and Control Gate Based on an Optimized Back Propagation Neural Network

The measurement and control gate, as a new type of measurement and control equipment, has been widely used for water quantity control in irrigation areas. However, there is a lack of methods for calibrating the flow inside the measurement box at present. This paper establishes a flow prediction model based on a back propagation (BP) neural network and its optimization algorithm by using 450 sets of sample data obtained from the indoor gate overflow test and verified the effectiveness and accuracy of the prediction model by using another 205 sets of sample data. The results show that the gate flow prediction model based on a BP neural network and its optimization algorithm has self-adaptability to different flow patterns, and its prediction accuracy is significantly higher than that of the traditional water measurement method. Compared to the unoptimized BP model, the BP model optimized by the genetic algorithm (GA) or particle swarm optimization (PSO) has higher prediction accuracy and better error distribution. Both GA and PSO algorithms can be used to optimize the initial weights and thresholds of the BP flow prediction model. However, by comprehensively analyzing the prediction accuracy, error distribution, and running time, the PSO algorithm has better optimization performance compared to the GA algorithm. The prediction model can provide a reference for flow rate calibration and the anomaly rejection of measurement and control gates in the irrigation area.

Rising trends in the burden of cardiovascular comorbidities and major adverse cardiovascular events among young female smokers in two national cohorts studied ten years apart (2017 vs. 2007)

BackgroundAmple evidence suggests that female smokers face a greater risk of smoking-related health problems than male smokers. Due to the growing number of young smokers in the United States, there has been limited information on the effects of smoking on young female smokers over the past decade. MethodsHospitalizations of young (18-44 years) female tobacco smokers were identified using the National Inpatient Sample datasets from 2007 and 2017. We compared differences in admission frequency, comorbidity burden, in-hospital outcomes [all-cause mortality and major adverse cardiac events (MACE)], and resource utilization between two young cohorts separated by 10 years. ResultsIn 2007, there were 665,901 admissions among young female smokers (median age: 35), compared to 1,224,479 admissions (median age: 32) in 2017. In both cohorts, white female smokers accounted for most admissions, followed by blacks. In 2017, the prevalence of alcohol abuse, hyperlipidemia, uncomplicated diabetes, and chronic pulmonary disease decreased relative to the 2007 cohort, whereas the prevalence of deficiency and chronic blood loss anemias, diabetes with complications, drug abuse, hypertension, congestive heart failure, depression, liver disease, and obesity increased significantly (p<0.001). The 2017 cohort had significantly higher odds of all-cause mortality [aOR 1.25 (95%CI: 1.16-1.35)] and a higher risk of MACE [aOR 1.17 (95%CI:1.14-1.20)] upon multivariable adjustment. (p<0.001). Comparatively, the 2017 cohort had fewer routine discharges and higher home healthcare needs than the 2007 cohort. ConclusionsIn this decade-apart analysis, the study reveals rising trends in the burden of comorbidities, MACE, and healthcare resource utilization in admissions (regardless of the primary cause) among relatively younger female smokers. It is crucial to educate young female smokers about the detrimental effects of tobacco and polysubstance abuse on cardiovascular outcomes.

Soft Sensor Modeling Method for the Marine Lysozyme Fermentation Process Based on ISOA-GPR Weighted Ensemble Learning.

Due to the highly nonlinear, multi-stage, and time-varying characteristics of the marine lysozyme fermentation process, the global soft sensor models established using traditional single modeling methods cannot describe the dynamic characteristics of the entire fermentation process. Therefore, this study proposes a weighted ensemble learning soft sensor modeling method based on an improved seagull optimization algorithm (ISOA) and Gaussian process regression (GPR). First, an improved density peak clustering algorithm (ADPC) was used to divide the sample dataset into multiple local sample subsets. Second, an improved seagull optimization algorithm was used to optimize and transform the Gaussian process regression model, and a sub-prediction model was established. Finally, the fusion strategy was determined according to the connectivity between the test samples and local sample subsets. The proposed soft sensor model was applied to the prediction of key biochemical parameters of the marine lysozyme fermentation process. The simulation results show that the proposed soft sensor model can effectively predict the key biochemical parameters with relatively small prediction errors in the case of limited training data. According to the results, this model can be expanded to the soft sensor prediction applications in general nonlinear systems.

Malicious Application Detection in Windows Using Machine Learning

As the proliferation of digital technology continues, the threat landscape for Windows operating systems has become increasingly complex. Malicious applications, including viruses, ransomware, and spyware, pose a significant risk to both individuals and organizations. To combat this growing threat, there is a pressing need for effective and efficient methods for detecting and mitigating malicious applications. This research paper presents an innovative approach to Malicious Application Detection in Windows using Support Vector Machine (SVM) algorithms. SVM is a powerful machine learning technique that has been successfully applied in various classification tasks, including malware detection. The primary objective of this study is to develop a robust and reliable system that can differentiate between benign and malicious applications in a Windows environment. We start by collecting a comprehensive dataset of Windows applications, comprising both legitimate and malicious software samples. Feature extraction techniques are employed to convert the application data into a suitable format for SVM analysis. These features may include file attributes, system call sequences, and behaviour analysis metrics

PATH-37. PROGNOSTIC IMPACT OF TUMOR MUTATIONAL BURDEN AT BASELINE IN IDH-WILDTYPE GLIOBLASTOMA

Abstract Tumor mutational burden (TMB) has emerged as a potential biomarker for survival and response to immune checkpoint therapy. However, limited data are available regarding the prognostic impact of TMB at baseline in glioblastoma (GBM). This study aimed to evaluate the somatic mutation status and survival outcomes of isocitrate dehydrogenase (IDH) wildtype GBM using a dataset of GBM samples from cBioPortal and single tertiary hospital. A total of 840 samples were extracted from cBioPortal, including 410 samples from Memorial Sloan Kettering Cancer Center (MSKCC), 234 samples from The Cancer Genome Atlas (TCGA), 106 samples from Glioma Longitudinal Analysis Consortium (GLASS), and 90 samples from Clinical Proteomic Tumor Analysis Consortium (CPTAC). Additionally, we enrolled 341 GBM patients with targeted sequencing data (OncoPanel AMC) from Asan Medical Center. The TMBHigh group was defined as the top 10% value of TMB within each cohort. Among the total of 1,081 IDH-wt GBM samples, 104 samples were classified as TMBHigh. No significance difference in overall survival (OS) was observed between the TMBHigh group and the non-TMBHigh group across the 5 cohorts. The mutations of DNA mismatch repair genes (MLH1, MSH2, MSH6, PMS2, and POLE) was significantly higher in the TMBHigh group compared with the non-TMBHigh group in 3 of 5 cohorts (pASAN=1.02x10-4, pMSKCC=1.5x10-2, and pCPTAC=1.26x10-2). However, when we calculated the indel to total mutation ratio (I-index) to dedect microsatellite Instability, only the MSKCC cohort showed a higher I-index in the TMBHigh group compared to the non-TMBHigh group (16.7% vs 13.7%, p=2.67x10-3). In summary, there is no significant association between TMB at baseline and poor survival outcomes in IDH-wildtype GBM. Further research is needed to explore the potential implications of TMB as a prognostic marker and its relevance for personalized treatment strategies in GBM patients.

A novel multimodal framework for early diagnosis and classification of COPD based on CT scan images and multivariate pulmonary respiratory diseases

Background and objectiveChronic Obstructive Pulmonary Disease (COPD) is one of the world's worst diseases; its early diagnosis using existing methods like statistical machine learning techniques, medical diagnostic tools, conventional medical procedures, and other methods is challenging due to misclassification results of COPD diagnosis and takes a long time to perform accurate prediction. Due to the severe consequences of COPD, detection and accurate diagnosis of COPD at an early stage is essential. This paper aims to design and develop a multimodal framework for early diagnosis and accurate prediction of COPD patients based on prepared Computerized Tomography (CT) scan images and lung sound/cough (audio) samples using machine learning techniques, which are presented in this study. MethodThe proposed multimodal framework extracts texture, histogram intensity, chroma, Mel-Frequency Cepstral Coefficients (MFCCs), and Gaussian scale space from the prepared CT images and lung sound/cough samples. Accurate data from All India Institute Medical Sciences (AIIMS), Raipur, India, and the open respiratory CT images and lung sound/cough (audio) sample dataset validate the proposed framework. The discriminatory features are selected from the extracted feature sets using unsupervised ML techniques, and customized ensemble learning techniques are applied to perform early classification and assess the severity levels of COPD patients. ResultsThe proposed framework provided 97.50%, 98%, and 95.30% accuracy for early diagnosis of COPD patients based on the fusion technique, CT diagnostic model, and cough sample model. ConclusionFinally, we compare the performance of the proposed framework with existing methods, current approaches, and conventional benchmark techniques for early diagnosis.

Estimation of target occupancy in repeated dosing design studies using positron emission tomography: Biases due to target upregulation.

Positron emission tomography (PET) has become indispensable in the quantification of target engagement by brain targeting medications. The relationship between the drug plasma concentration (or drug dose administered) and target occupancy determined during a PET occupancy study has provided valuable information for the assessment of novel pharmaceuticals in the early phases of drug development. Such information is also critical for the understanding of the mechanisms of action and side-effect profile of approved medication commonly used in the clinic. Occupancy studies conducted following repeated drug dosing (RD) can produce systematic differences from those conducted following single drug dose (SD), differences that have not been adequately explored. We have hypothesised that when differences are observed between RD and SD studies, they are related to changes in target density induced by repeated drug accumulation. We have developed a modified occupancy model to account for potential changes in target density and tested it on a sample dataset. We found that target upregulation can parsimoniously explain the differences in drug affinity estimated in SD and RD studies. Our findings have implications for the interpretation of RD occupancy data in the literature and the relationship between specific target occupancy levels and drug efficacy and tolerability.