Stratification Of Data Research Articles

Development and validation of the post-CAR prognostic index for large B-cell lymphoma patients after CAR-T progression in third or later line treatment

Chimeric antigen receptor (CAR) T-cell therapy fails to achieve durable responses in over 60% of relapsed/refractory (R/R) large B-cell lymphoma (LBCL) patients in the third or later line setting. After CAR-T failure, survival outcomes are heterogeneous and a prognostic model in this patient population is lacking. A training cohort of 216 patients with progressive disease (PD) after CAR-T from 12 Spanish centers was used to develop the Post-CAR Prognostic Index (PC-PI); primary endpoint was overall survival (OS) from CAR-T progression. Validation was performed in an external cohort from three different European centers (n = 204). The prognostic score incorporated five variables, assessed at time of PD to CAR-T: ECOG (> 0), hemoglobin (< 10 g/dL), LDH (≥ 2xULN), number of extranodal sites (> 1) and time from CAR-T to PD (< 4 months). Patients were classified in four risk groups with distinct OS (p-value < 0.05 in all comparisons). In the validation cohort, median OS in the low (31%), intermediate-low (26%), intermediate-high (17%) and high risk (26%) were 15.7, 7.1, 1.8 and 1.0 months, respectively (p < 0.05 in all comparisons). Results were consistent following adjustment for subsequent treatment. In the external cohort, the PC-PI showed a C-statistic of 0.79 (95%CI 0.76–0.82), outperforming IPI and R-IPI. In conclusion, the PC-PI score is a novel tool for OS prediction and could facilitate risk-adapted management of LBCL patients relapsing after CAR T-cells. Additionally, these results will help stratification and interpretation of trials and real-world data incorporating CART-exposed patients.

The information layer: shaping data into meaning with interactive tools

Abstract Valorization and interpretation of health data are contingent upon a robust analytical framework. This framework encompasses essential processes such as data cleaning, validation, and the critical evaluation of bias and missing data. A nuanced understanding of these factors is crucial for contextualizing data through stratification, developing comprehensive analysis plans, and addressing interactions, covariates, and confounders. Such meticulous methodologies guard against ‘data fishing expeditions’ that could otherwise lead to erroneous conclusions. Healthybelgium.be exemplifies our approach for deciding on appropriate levels of data aggregation and stratification. This study serves as a foundation for deriving conclusions and formulating hypotheses regarding the health status of the national and various sub-populations. Our discussion will cover the limitations of the questions that can be answered with the available data, the resulting opportunities for data enhancement, and the subsequent valorization of findings for knowledge advancement. Furthermore, we will present the interactive tools (through the use of R Shiny) developed by Sciensano to disseminate information broadly, making it accessible and comprehensible to diverse audiences. These tools embody a balance of strength and limitation, facilitating an informed dialogue among stakeholders. The ongoing refinement of these tools, influenced by continuous learning and insights, exemplifies our commitment to evolving public health intelligence.

Effects of simulated smoke condensate generated from combustion of selected military burn pit contents on human airway epithelial cells.

Exposure to military burn pit smoke during deployment is associated with different respiratory and non-respiratory diseases. However, information linking smoke exposure to human pulmonary health is lacking. This study examined the effects of simulated burn pit smoke condensates on human airway epithelial cells (HAECs) from twelve donors (smokers/non-smokers, biological female/male) cultured at an air-liquid interface and exposed to condensates from three simulated burn pit waste materials (cardboard, plywood, and plastic) incinerated at two combustion conditions: smoldering and flaming. Cellular gene expression was analyzed using bulk RNA sequencing, and basolateral media cytokine levels were assessed using multiplex immunoassay. Flaming smoke condensates caused more significant differentially expressed genes (DEGs) with plywood flaming smoke being the most potent in altering gene expression and modulating cytokine release. Cardboard and plywood flaming condensates primarily activated detoxification pathways, whereas plastic flaming affected genes related to anti-microbial and inflammatory responses. Correlation analysis between smoke condensate chemicals and gene expression to understand the underlying mechanism revealed crucial role of oxygenated polycyclic aromatic hydrocarbons (PAHs) and aluminum, molybdenum, and silicon elements; IL6 expression was positively correlated with most PAHs. Stratification of data based on HAEC donor demographics suggests that these affect gene expression changes. Enrichment analysis indicated similarity with several deployment-related presumptive and reported diseases, including asthma, emphysema, and cancer of different organs. This study highlights that simulated burn pit smoke exposure of HAECs causes gene expression changes indicative of deployment-related diseases with more pronounced effects seen in smokers and females. Future studies are needed to further characterize how sex and smoking status affect deployment-related diseases.

The Operational and Climate Land Surface Temperature Products from the Sea and Land Surface Temperature Radiometers on Sentinel-3A and 3B

Land Surface Temperature (LST) is integral to our understanding of the radiative energy budget of the Earth’s surface since it provides the best approximation to the thermodynamic temperature that drives the outgoing longwave flux from surface to atmosphere. Since 5 July 2017, an operational LST product has been available from the Sentinel-3A mission, with the corresponding product being available from Sentinel-3B since 17 November 2018. Here, we present the first paper describing formal products, including algorithms, for the Sea and Land Surface Temperature Radiometer (SLSTR) instruments onboard Sentinel-3A and 3B (SLSTR-A and SLSTR-B, respectively). We evaluate the quality of both the Land Surface Temperature Climate Change Initiative (LST_cci) product and the Copernicus operational LST product (SL_2_LST) for the years 2018 to 2021. The evaluation takes the form of a validation against ground-based observations of LST across eleven well-established in situ stations. For the validation, the mean absolute daytime and night-time difference against the in situ measurements for the LST_cci product is 0.77 K and 0.50 K, respectively, for SLSTR-A, and 0.91 K and 0.54 K, respectively, for SLSTR-B. These are an improvement on the corresponding statistics for the SL_2_LST product, which are 1.45 K (daytime) and 0.76 (night-time) for SLSTR-A, and 1.29 K (daytime) and 0.77 (night-time) for SLSTR-B. The key influencing factors in this improvement include an upgraded database of reference states for the generation of retrieval coefficients, higher stratification of the auxiliary data for the biome and fractional vegetation, and enhanced cloud masking.

Developing a generalisable stratification approach for clerical review of linked data

ObjectiveData linkage is a vital process in the creation of many national statistics, but understanding the quality of linked data is currently highly inefficient. To find errors, data must be reviewed by humans which is costly and lengthy. Sampling is used to reduce the clerical burden. This research aims to develop a method for stratifying links to create representative samples while reducing the number reviewed. The final method will enable nuanced stratification of data for review whilst optimising resource efficiency. The objectives are to: ensure that the method is adaptable across diverse datasets, achieve full automation, ensure scalability to accommodate large datasets. ApproachOur approach centres on designing an algorithm that responds to the variability in the data distribution of probabilistic scores and stratify accordingly. The intention is for the developed method to automatically adjust its parameters, such as strata threshold and numbers based on the data’s characteristics. The research involves a comparative analysis of the performance of dynamic- and percentile-based stratification against the current standard practice of static threshold stratification. ResultsTests are ongoing to compare the above methods on a variety of metrics including homogeneity of strata, total variance, and between-strata distance. Findings will be presented at the conference. ConclusionsWe hope to design a robust, generalisable and scalable stratification method that can be integrated into a Linkage pipeline. Implications Implementing the method will help to improve the quality of national statistics, ensuring more accurate, reliable and timely outputs are produced in a resource efficient manner.

Metagenomics analysis of sewage for surveillance of antimicrobial resistance in South Africa.

Our 24-month study used metagenomics to investigate antimicrobial resistance (AMR) abundance in raw sewage from wastewater treatment works (WWTWs) in two municipalities in Gauteng Province, South Africa. At the AMR class level, data showed similar trends at all WWTWs, showing that aminoglycoside, beta-lactam, sulfonamide and tetracycline resistance was most abundant. AMR abundance differences were shown between municipalities, where Tshwane Metropolitan Municipality (TMM) WWTWs showed overall higher abundance of AMR compared to Ekurhuleni Metropolitan Municipality (EMM) WWTWs. Also, within each municipality, there were differing trends in AMR abundance. Notably, within TMM, certain AMR classes (macrolides and macrolides_streptogramin B) were in higher abundance at a WWTW serving an urban high-income area, while other AMR classes (aminoglycosides) were in higher abundance at a WWTW serving a semi-urban low income area. At the AMR gene level, all WWTWs samples showed the most abundance for the sul1 gene (encoding sulfonamide resistance). Following this, the next 14 most abundant genes encoded resistance to sulfonamides, aminoglycosides, macrolides, tetracyclines and beta-lactams. Notably, within TMM, some macrolide-encoding resistance genes (mefC, msrE, mphG and mphE) were in highest abundance at a WWTW serving an urban high-income area; while sul1, sul2 and tetC genes were in highest abundance at a WWTW serving a semi-urban low income area. Differential abundance analysis of AMR genes at WWTWs, following stratification of data by season, showed some notable variance in six AMR genes, of which blaKPC-2 and blaKPC-34 genes showed the highest prevalence of seasonal abundance differences when comparing data within a WWTW. The general trend was to see higher abundances of AMR genes in colder seasons, when comparing seasonal data within a WWTW. Our study investigated wastewater samples in only one province of South Africa, from WWTWs located within close proximity to one another. We would require a more widespread investigation at WWTWs distributed across all regions/provinces of South Africa, in order to describe a more comprehensive profile of AMR abundance across the country.

Coupled space-time scale analysis for obtaining enhanced accuracy in species dynamics

Scale effects significantly affect the rationality and accuracy of ecological models, so temporal and spatial scales are included in the construction and application of ecological studies. However, few ecological pattern analyses focus on both of these scales simultaneously, which is especially important in terms of synchrony. In this study, based on fishery catch data and simulated data, we propose a framework for coupled space-time scale analysis and reveal possible deficiencies in conventional scale studies. We verified the feasibility and reliability of the scheme using simulated data with different combinations of temporal and spatial scales. The results showed that our scheme can simultaneously identify and localize scale features in the variation in species spatial-temporal patterns, specifically regarding the synchrony and amplitude of temporal dynamics (or spatial patterns) of multiple sites (or times). Our scheme revealed that spatial scales had an important effect on the intrinsic 13-year scale features, which significantly improved our understanding of global Rajiformes fishery dynamics. This showed that the spatial-temporal patterns and scale features obtained based on the coupled space-time scale analysis were complete and more accurate. At the same time, our scheme correctly identified complex pattern structures, such as data stratification and multiple combinations of time and space scales, which can reduce possible errors in practical applications. This scheme can be applied to the identification and prediction of the spatial-temporal patterns in biodiversity in the future and will help in formulating effective policies promoting sustainable fishery resources management and comprehensive conservation of endangered marine species.

Implementation of sustainability activities in food retail chains

PurposeThis study aims to examine the implementation of sustainability activities in food retail chains operating in the Czech market and discusses the specific activities that influence their implementation.Design/methodology/approachUsing the neo-institutional theory as a framework, the research employs qualitative content analysis with an explanatory and inductive approach based on the stratification of data collected by multiple-round, personal, face-to-face interviews by the researchers with managers of the Czech headquarters of multinational food retail chains.FindingsThere is lively development in activities in the fields of the environment and social welfare, while the activities in governance and economic resilience are more stabilised. To remain ahead of the competition, retail chains aim to implement and communicate as many activities as possible where at least some links to sustainability can be found. A lack of benchmarks and clear definitions reduce the ability to determine their degree of engagement. Thus, market actors are inundated with sustainability claims leading to inflation of the concept, while significant achievements are drowned.Originality/valueThis paper, based on empirical research, contributes to the current literature by showing inflation of the sustainability concept as one of the key effects of retail-chain competition. In this context, it also strengthens previous findings on the profitability motive behind sustainability activities. Moreover, we have developed our own stratification method, enabling differentiation between activities getting the most attention due to their progressive phase and those where development is in the background.

Insight into the Predictive Power of Surrogate Diagnostic Indices for Identifying Individuals with Metabolic Syndrome.

This study aimed to assess the diagnostic capability of insulin surrogate measurements in identifying individuals with metabolic syndrome (MetS) and propose applicable indices derived from fasting values, particularly in large study populations. Data were collected from the datasets of the Surveillance of Risk Factors of NCDs in Iran Study (STEPS). MetS was defined based on the National Cholesterol Education Program (NCEP) criteria. Various insulin surrogate indices, including Homeostasis Model Assessment (HOMA), Quantitative Insulin Sensitivity Check Index (QUICKI), Fasting glucose to insulin ratio (FGIR), Reynaud, Reciprocal insulin, McAuley, Metabolic Score for Insulin Resistance (METS-IR), Triglyceride-glucose index (TyG), TG/ HDL-C, TG/ BMI, and TG/ WC ratio were assessed. Receiver Operating Characteristic (ROC) curves were used to assess pathologic conditions and determine the optimal cut-off through the highest score of the Youden index. Also, Area Under the Curve (AUC) values were established for each index totally and according to sex, age, and BMI differences. The study population consisted of 373 individuals (49.9% women; 75.1% middle age, 39.1% obese, and 27.3% overweight), of whom 117 (31.4%) had MetS. The METS-IR (AUC: 0.856; 95% CI: 0.817-0.895), TG/ HDL-C (AUC: 0.820; 95% CI: 0.775-0.886), TyG (AUC: 0.808; 95% CI: 0.759-0.857), and McAuley (AUC: 0.804; 95% CI: 0.757-0.852) indices provided the greatest AUC respectively for detection of MetS. The values of AUC for all the indices were higher in men than women. This trend was consistent after data stratification based on BMI categories, middle age, and senile individuals. The present study indicated that indices of insulin, including METS-IR, TG/HDLC, TyG, and McAuley, have an equal or better capacity in determining the risk of MetS than HOMA-IR, are capable of identifying individuals with MetS and may provide a simple approach for identifying populations at risk of insulin resistance.

Physician and Practice Characteristics Influencing Telemedicine Uptake Among Frontline Clinicians in the Early COVID-19 Pandemic Response: National Survey Study.

Telemedicine expanded rapidly during the COVID-19 pandemic, as key policy changes, financial support, and pandemic fears tipped the balance toward internet-based care. Despite this increased support and benefits to patients and clinicians, telemedicine uptake was variable across clinicians and practices. Little is known regarding physician and institutional characteristics underlying this variability. This study aimed to evaluate factors influencing telemedicine uptake among frontline physicians in the early pandemic response. We surveyed a national stratified sample of frontline clinicians drawn from the American Medical Association Physician Professional Data in June or July 2020. The survey inquired about the first month and most recent month (June 2020) of pandemic telemedicine use; sample data included clinician gender, specialty, census region, and years in practice. Local pandemic conditions were estimated from county-level data on COVID-19 rates at the time of survey response. Data were analyzed in a weighted logistic regression, controlling for county-specific pandemic data, and weighted to account for survey data stratification and nonresponse. Over the first 3-4 months of the pandemic, the proportion of physicians reporting use of telemedicine in >30% of visits increased from 29.2% (70/239) to 35.7% (85/238). Relative to primary care, odds of substantial telemedicine use (>30%) both during the first month of the pandemic and in June 2020 were increased among infectious disease and critical care physicians and decreased among hospitalists and emergency medicine physicians. At least minimal prepandemic telemedicine use (odds ratio [OR] 11.41, 95% CI 1.34-97.04) and a high 2-week moving average of local COVID-19 cases (OR 10.16, 95% CI 2.07-49.97) were also associated with substantial telemedicine use in June 2020. There were no significant differences according to clinician gender, census region, or years in practice. Prepandemic telemedicine use, high local COVID-19 case counts, and clinician specialty were associated with higher levels of substantial telemedicine use during the early pandemic response. These results suggest that telemedicine uptake in the face of the pandemic may have been heavily influenced by the level of perceived threat and the resources available for implementation. Such understanding has important implications for reducing burnout and preparation for future public health emergencies.

What sperm parameters effect blastocyst formation and quality during ICSI with severe male infertility

Sperm-derived genetic material contributes half of the genome to the embryo, hence it’s crucial to investigate which sperm parameter influences blastocyst formation in the intracytoplasmic sperm injection (ICSI) cycles with severe male infertility. The retrospective study analyzed 296 ICSI cycles with severe oligoasthenoteratozoospermia (OAT) and 99 ICSI cycles with preimplantation genetic testing for aneuploidy (PGT-A). Following the correlation analysis, data stratifications were performed in the OAT ICSI subgroup. The results showed that the matching blastocyst in the OAT ICSI cycles had inferior sperm parameters. DFI and sperm morphology had an influence on the blastocyst formation rate and the high-quality blastocysts formation rate on Day6, but no significant effect on the blastocyst development on Day 5. The high-quality blastocysts formation rate and ratio of high-quality blastocyst on Day 6 were demonstrably better in the subgroup of the teratozoospermic morphology when DFI was within the normal range. In the case of the normal sperm morphology, no statistically significant difference was found in blastocyst development, although there were numerical differences within different DFI subgroups. It was concluded that the blastocyst quality and development declined with the decreased sperm qualities.

Vaccination Status in Children presenting with severe complicated Measles

ABSTRACT Background: Measles occurs worldwide and remains a leading cause of mortality especially among children ≤5 years of age. Present study will determine this association in our local population. Objective: The aim of the study was to determine the frequency of unvaccinated children in patients presenting with severe Measles. Material and Methods: After permission from Hospital ethical committee. Total 145 patients of severe measles were enrolled, and frequency of unvaccinated children was noted. This study was conducted in Children Ward, Women &amp; Children Hospital, MTI Bannu from October 2022 to April 2023. The results obtained were presented in the form of tables and graphs and level of significance was determined statistically. Results: In our study, total 145 patients were enrolled. Mean age of patients was 7.07±2.3 years. Minimum age was 3 years and maximum age was 12 years. There were 46.2% males and 53.8% female patients. Most of patients belonged to middle socioeconomic status i.e. 57.2. Regarding maternal education 50.3% were having 6-10 years of education, 20.9% had less than 5 years of education, 16.6% had 11-12 years of education while only 4.1% had more than 12 years of education. In our study participants 47.6% lived in urban area and 52.4% lived in rural area. There were 66.2% unvaccinated children and 33.8% were vaccinated. Data stratification for age group, BMI, residential status and socioeconomic status was significant i.e. p-value≤ 0.05. Conclusion: Severe measles infections are prevalent among unvaccinated children due to their susceptibility to complications such as pneumonia and encephalitis. Vaccination remains the most effective measure to prevent measles outbreaks and safeguard public health.

CROSS SECTIONAL STUDY ON PROTECTIVE FACTORS AGAINST SUICIDE IN INDIVIDUALS WITH SEVERE DEPRESSION

OBJECTIVE The aim of this study is to find out “factors that prohibit those suffering from severe depression with suicidal ideation of committing suicide”, admitted to the psychiatric unit of a tertiary care hospital in Khyber Pakhtunkhwa, Peshawar. STUDY DESIGN Cross sectional study PLACE AND DURATION OF STUDY This study was conducted in psychiatry department of Khyber Teaching Hospital, Peshawar from May 2023 till November 2023. SUBJECTS AND METHODS We conducted this study on 230 admitted patients, and they were diagnosed with severe depression according to ICD-10 and beck scale for suicidal ideation was applied and a score of ≥ 8 was considered positive for the presence of suicidal ideations and then a “modified version of reasons for living scale” to identify protective factors against suicide. Data was analyzed using SPSS version 22.0. RESULTS The most common protective factor found in our study was religious and moral objections in 37.4% patients followed by fear of social disapproval in 17.4%, hopefulness and fear of suicide in 15.7% each and responsibility towards family was present in 13.9% patients. Data stratification for protective factors and other demographic characteristics including gender, residential, socioeconomic, marital, educational statuses and family system were statistically significant with their P-values of &lt;0.001. CONCLUSION This study demonstrated that religious and moral objections were the most common protective factors in prevention of suicide, followed by fear of social disapproval, hopefulness, fear of suicide and responsibility towards family. Gender, residential, socioeconomic, marital, and educational statuses along with family system were significantly associated with the presence of protective factors against suicide.

Deep learning ensembles for detecting brain metastases in longitudinal multi-modal MRI studies

Metastatic brain cancer is a condition characterized by the migration of cancer cells to the brain from extracranial sites. Notably, metastatic brain tumors surpass primary brain tumors in prevalence by a significant factor, they exhibit an aggressive growth potential and have the capacity to spread across diverse cerebral locations simultaneously. Magnetic resonance imaging (MRI) scans of individuals afflicted with metastatic brain tumors unveil a wide spectrum of characteristics. These lesions vary in size and quantity, spanning from tiny nodules to substantial masses captured within MRI. Patients may present with a limited number of lesions or an extensive burden of hundreds of them. Moreover, longitudinal studies may depict surgical resection cavities, as well as areas of necrosis or edema. Thus, the manual analysis of such MRI scans is difficult, user-dependent and cost-inefficient, and – importantly – it lacks reproducibility. We address these challenges and propose a pipeline for detecting and analyzing brain metastases in longitudinal studies, which benefits from an ensemble of various deep learning architectures originally designed for different downstream tasks (detection and segmentation). The experiments, performed over 275 multi-modal MRI scans of 87 patients acquired in 53 sites, coupled with rigorously validated manual annotations, revealed that our pipeline, built upon open-source tools to ensure its reproducibility, offers high-quality detection, and allows for precisely tracking the disease progression. To objectively quantify the generalizability of models, we introduce a new data stratification approach that accommodates the heterogeneity of the dataset and is used to elaborate training-test splits in a data-robust manner, alongside a new set of quality metrics to objectively assess algorithms. Our system provides a fully automatic and quantitative approach that may support physicians in a laborious process of disease progression tracking and evaluation of treatment efficacy.

Application of a data stratification approach in computer medical monitoring systems

The research object is the processes occurring in the data stratification subsystem in the medical monitoring computer system, which is part of the decision support system. Such a subsystem aims to solve data analysis and processing problems in the medical monitoring system. Among them, the problems of anomaly detection, data marking, state determination, selection of the most informative variables, and justification of decision-making are selected for solving. The paper proposes the structure and implementation of the data stratification subsystem in the decision support system. The subsystem contains modules for anomaly detection and an autoencoder, a clustering module using an advanced multi-agent clustering method, and a state detection module with a modified neural network training procedure. Modules of the stratification subsystem have been tested using diabetes monitoring data. The results showed that the clustering module provides 25.7 % lower accuracy than the achieved neural network. The accuracy difference is explained by the complexity of the data and the lack of adaptability of the proposed method to solving such problems. It is shown that the method of determining the overall informativeness of variables covers 90 % informativeness with 10 variables, comparable to the variability data. In general, the flexible nature of the proposed stratification subsystem allows for solving the problems. The proposed stratification subsystem offers a robust solution for improving treatment strategies and decision-making in computer medical monitoring systems. Its versatility allows it to be used in any system where decision support is needed, providing valuable information about informative variables and decision-making features for clinicians and researchers

Investigation of performance limitations in continuous hot-dip galvanizing units associated with product defects

The goals of the work were to search and systematize the speed limits of the steel strip during hot-dip galvanizing, associated with the threat of product defects. Since speed can be combined with many other factors, this paper provides an overview of the known causes of common defects. The causes were grouped taking into account the operations of individual sections of continuous hot-dip galvanizing units. To determine the circumstances in which defects occur, a method was used that involved step-by-step stratification of retrospective data and comparison of the distribution density of influencing factors for defective and non-defective products. The method was applied in the analysis of the data on defects at the MMK Metallurgy plant in Turkey, obtained in 2020–2021. Twenty-one types of acceptable and unacceptable defects that occur during galvanizing of DX51D steel were selected for analysis. Twenty-two technological parameters were taken as factors, including the strip speed. For each selected type of the defects, a set of influencing factors is determined, and for some types of defects, the alleged causes of their occurrence are indicated. It is shown that the relationship observed for many types of defects with the strip speed can actually be caused by other factors. We determined the types of defects, the probability of which increases with an increase in the speed or the level of its change. The paper proposes measures aimed at preventing the increase in the proportion of defective products along with the performance gain.

Application of the screening method for estimating COVID-19 vaccine effectiveness using routine surveillance data: Germany's experience during the COVID-19 pandemic, July 2021 to March 2023.

The screening method represents a simple, quick, and practical tool for estimating vaccine effectiveness (VE) using routine disease surveillance and vaccine coverage data, even if these data cannot be linked. In Germany, where notification data, laboratory testing data, and vaccine coverage data cannot be linked due to strict data protection requirements, the screening method was used to assess COVID-19 VE continuously between July 2021 and March 2023. During this period, when Delta and Omicron variants circulated, VE estimates were produced in real-time for different age groups and clinical outcomes. Here we describe the country's overall positive experience using the screening method, including its strengths and limitations, and provide practical guidance regarding a few issues, such as case definition stringency, testing behaviour, and data stratification, that require careful consideration during data analysis and the interpretation of the results.

A comprehensive overview of microbiome data in the light of machine learning applications: categorization, accessibility, and future directions.

Metagenomics, Metabolomics, and Metaproteomics have significantly advanced our knowledge of microbial communities by providing culture-independent insights into their composition and functional potential. However, a critical challenge in this field is the lack of standard and comprehensive metadata associated with raw data, hindering the ability to perform robust data stratifications and consider confounding factors. In this comprehensive review, we categorize publicly available microbiome data into five types: shotgun sequencing, amplicon sequencing, metatranscriptomic, metabolomic, and metaproteomic data. We explore the importance of metadata for data reuse and address the challenges in collecting standardized metadata. We also, assess the limitations in metadata collection of existing public repositories collecting metagenomic data. This review emphasizes the vital role of metadata in interpreting and comparing datasets and highlights the need for standardized metadata protocols to fully leverage metagenomic data's potential. Furthermore, we explore future directions of implementation of Machine Learning (ML) in metadata retrieval, offering promising avenues for a deeper understanding of microbial communities and their ecological roles. Leveraging these tools will enhance our insights into microbial functional capabilities and ecological dynamics in diverse ecosystems. Finally, we emphasize the crucial metadata role in ML models development.

Prognostic nomogram in patients with gastrointestinal stromal tumors: a SEER-based study.

Gastrointestinal stromal tumor (GIST) is a common mesenchymal tumor of the gastrointestinal system. They originate from the interstitial cells of Cajal located within the muscle layer and are characterized by over-expression of the tyrosine kinase receptor KIT. Data from the Surveillance Epidemiology, and End Results (SEER) database of 1,213 patients diagnosed with GIST between 2010 and 2019 were dichotomized into a modeling set and a validation set at a 2:1 ratio. For the modeling set, both univariate and multivariate Cox regression analyses were used to identify independent prognostic factors. A nomogram was then constructed based on these determinants. Model efficacy was tested using receiver operating characteristic (ROC) curves, calibration curves, clinical decision curves, and risk stratification analysis in both subsets. Identified prognostic determinants included age, sex, pathological differentiation level, tumor-node-metastasis (TNM) stage, surgical intervention, radiotherapy, and marital status. The constructed nomogram showed area under the ROC curve (AUC) values of 0.822, 0.793, and 0.779 for 1-, 3-, and 5-year overall survival (OS) in the modeling set, respectively, while in the validation set, the values were 0.796, 0.823, and 0.806, respectively. Calibration plots from both sets confirmed the concordance between predicted and observed survival. Decision curve analysis (DCA) indicated significant clinical utility for the nomogram. Risk stratification of the patient data revealed distinct survival differences between high-risk and low-risk cohorts in both sets (P<0.001). A novel and potent nomogram for the prognosis of GIST has been introduced. This model's precision offers crucial insights for clinical decisions, yet further external validation remains essential.

Towards a global impact-based forecasting model for tropical cyclones

Abstract. Tropical cyclones (TCs) produce strong winds and heavy rains accompanied by consecutive events such as landslides and storm surges, resulting in losses of lives and livelihoods, particularly in regions with high socioeconomic vulnerability. To proactively mitigate the impacts of TCs, humanitarian actors implement anticipatory action. In this work, we build upon such an existing anticipatory action for the Philippines, which uses an impact-based forecasting model for housing damage based on eXtreme Gradient Boosting (XGBoost) to release funding and trigger early action. We improve it in three ways. First, we perform a correlation and selection analysis to understand if Philippines-specific features can be left out or replaced with features from open global data sources. Secondly, we transform the target variable (percentage of completely damaged houses) and not yet grid-based global features to a 0.1∘ grid resolution by de-aggregation using Google Open Buildings data. Thirdly, we evaluate XGBoost regression models using different combinations of global and local features at grid and municipality spatial levels. We first introduce a two-stage model to predict if the damage is above 10 % and then use a regression model trained on all or only high-damage data. All experiments use data from 39 typhoons that impacted the Philippines between 2006–2020. Due to the scarcity and skewness of the training data, specific attention is paid to data stratification, sampling, and validation techniques. We demonstrate that employing only the global features does not significantly influence model performance. Despite excluding local data on physical vulnerability and storm surge susceptibility, the two-stage model improves upon the municipality-based model with local features. When applied to anticipatory action, our two-stage model would show a higher true-positive rate, a lower false-negative rate, and an improved false-positive rate, implying that fewer resources would be wasted in anticipatory action. We conclude that relying on globally available data sources and working at the grid level holds the potential to render a machine-learning-based impact model generalizable and transferable to locations outside of the Philippines impacted by TCs. Also, a grid-based model increases the resolution of the predictions, which may allow for a more targeted implementation of anticipatory action. However, it should be noted that an impact-based forecasting model can only be as good as the forecast skill of the TC forecast that goes into it. Future research will focus on replicating and testing the approach in other TC-prone countries. Ultimately, a transferable model will facilitate the scaling up of anticipatory action for TCs.