SEARCH · PubMed Health

Results for “Dataset”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 19 recordsLinked to original sources

PubMed2006

Are clusters found in one dataset present in another dataset?

In many microarray studies, a cluster defined on one dataset is sought in an independent dataset. If the cluster is found in the new dataset, the cluster is said to be "reproducible" and may be biologically significant. Classifying a new datum to a previously defined cluster can be seen as predicting which of the previously defined clusters is most similar to the new datum. If the new data classified to a cluster are similar, molecularly or clinically, to the data already present in the cluster, then the cluster is reproducible and the corresponding prediction accuracy is high. Here, we take advantage of the connection between reproducibility and prediction accuracy to develop a validation procedure for clusters found in datasets independent of the one in which they were characterized. We define a cluster quality measure called the "in-group proportion" (IGP) and introduce a general procedure for individually validating clusters. Using simulations and real breast cancer datasets, the IGP is compared to four other popular cluster quality measures (homogeneity score, separation score, silhouette width, and weighted average discrepant pairs score). Moreover, simulations and the real breast cancer datasets are used to compare the four versions of the validation procedure which all use the IGP, but differ in the way in which the null distributions are generated. We find that the IGP is the best measure of prediction accuracy, and one version of the validation procedure is the more widely applicable than the other three. An implementation of this algorithm is in a package called "clusterRepro" available through The Comprehensive R Archive Network (http://cran.r-project.org).

Breast Neoplasms↗

PubMed2005

Comprehensive analysis of a multidimensional liquid chromatography mass spectrometry dataset acquired on a quadrupole selecting, quadrupole collision cell, time-of-flight mass spectrometer: II. New developments in Protein Prospector allow for reliable and comprehensive automatic analysis of large datasets.

A thorough analysis of the protein interaction partners of the yeast GTPase Gsp1p was carried out by a multidimensional chromatography strategy of strong cation exchange fractionation of peptides followed by reverse phase LC-ESI-MSMS using a QSTAR instrument. This dataset was then analyzed using the latest developmental version of Protein Prospector. The Prospector search results were also compared with results from the search engine "Mascot" using a new results comparison program within Prospector named "SearchCompare." The results from this study demonstrate that the high quality data produced on a quadrupole selecting, quadrupole collision cell, time-of-flight (QqTOF) geometry instrument allows for confident assignment of the vast majority of interpretable spectra by current search engines.

Autoanalysis↗

PubMed2002

Visualization of conserved structures by fusing highly variable datasets.

INTRODUCTION: Skill, effort, and time are required to identify and visualize anatomic structures in three-dimensions from radiological data. Fundamentally, automating these processes requires a technique that uses symbolic information not in the dynamic range of the voxel data. We were developing such a technique based on mutual information for automatic multi-modality image fusion (MIAMI Fuse, University of Michigan). This system previously demonstrated facility at fusing one voxel dataset with integrated symbolic structure information to a CT dataset (different scale and resolution) from the same person. The next step of development of our technique was aimed at accommodating the variability of anatomy from patient to patient by using warping to fuse our standard dataset to arbitrary patient CT datasets. METHODS: A standard symbolic information dataset was created from the full color Visible Human Female by segmenting the liver parenchyma, portal veins, and hepatic veins and overwriting each set of voxels with a fixed color. Two arbitrarily selected patient CT scans of the abdomen were used for reference datasets. We used the warping functions in MIAMI Fuse to align the standard structure data to each patient scan. The key to successful fusion was the focused use of multiple warping control points that place themselves around the structure of interest automatically. The user assigns only a few initial control points to align the scans. Fusion 1 and 2 transformed the atlas with 27 points around the liver to CT1 and CT2 respectively. Fusion 3 transformed the atlas with 45 control points around the liver to CT1 and Fusion 4 transformed the atlas with 5 control points around the portal vein. The CT dataset is augmented with the transformed standard structure dataset, such that the warped structure masks are visualized in combination with the original patient dataset. This combined volume visualization is then rendered interactively in stereo on the ImmersaDesk in an immersive Virtual Reality (VR) environment. RESULTS: The accuracy of the fusions was determined qualitatively by comparing the transformed atlas overlaid on the appropriate CT. It was examined for where the transformed structure atlas was incorrectly overlaid (false positive) and where it was incorrectly not overlaid (false negative). According to this method, fusions 1 and 2 were correct roughly 50-75% of the time, while fusions 3 and 4 were correct roughly 75-100%. The CT dataset augmented with transformed dataset was viewed arbitrarily in user-centered perspective stereo taking advantage of features such as scaling, windowing and volumetric region of interest selection. CONCLUSIONS: This process of auto-coloring conserved structures in variable datasets is a step toward the goal of a broader, standardized automatic structure visualization method for radiological data. If successful it would permit identification, visualization or deletion of structures in radiological data by semi-automatically applying canonical structure information to the radiological data (not just processing and visualization of the data's intrinsic dynamic range). More sophisticated selection of control points and patterns of warping may allow for more accurate transforms, and thus advances in visualization, simulation, education, diagnostics, and treatment planning.

Algorithms↗

PubMed2000

Congenital Heart Surgery Nomenclature and Database Project: overview and minimum dataset.

The International Congenital Heart Surgery Nomenclature and Database Project was organized for the purpose of standardizing nomenclature and reporting strategies that would establish the foundations for an international database. Worldwide representatives met for a series of conferences, at which time, issues of nomenclature were discussed and debated. Authors were chosen to review the various congenital heart diagnoses and reflect the mediated debate that followed. Manuscripts were prepared that reviewed the appropriate extant nomenclature, made recommendations for an inclusive rather than an exclusive method of reporting, and determined a hierarchical database scheme that would allow several levels of reporting based on the data input. This manuscript outlines two datasets for an international congenital heart surgery database, a minimum dataset and a comprehensive dataset. The comprehensive dataset includes all the imagined variables, in a hierarchical scheme, which are detailed enough to generate risk stratification analyses. The minimum dataset will include data points that would create an essential dataset, which would be mandatory for data sharing and would lend itself to basic interpretation of trends. The minimum dataset has four drop-down menus for short lists of: (1) noncardiac abnormalities/general preoperative risk factors, (2) diagnoses, (3) procedures, and (4) complications, from which clinicians can choose for entry into the minimum dataset. There was universal agreement for these datasets and short lists by the assembled members of the Society of Thoracic Surgeons-Congenital Heart Surgery Database Committee and representatives from the European Association for Cardiothoracic Surgery. The datasets and short lists were also unanimously approved by the Congenital Heart Surgery Committee of The European Association for Cardiothoracic Surgery and adopted by the European Congenital Heart Surgeons Foundation.

Databases, Factual↗

PubMed2005

Managing the CT data explosion: initial experiences of archiving volumetric datasets in a mini-PACS.

Two image datasets (one thick section dataset and another volumetric dataset) were typically reconstructed from each single CT projection data. The volumetric dataset was stored in a mini-PACS with 271-Gb online and 680-Gb nearline storage and routed to radiologists' workstations, whereas the thick section dataset was stored in the main PACS. Over a 5-month sample period, 278 Gb of CT data (8976 examinations) was stored in the main PACS, and 738 Gb of volumetric datasets (6193 examinations) was stored in the mini-PACS. The volumetric datasets formed 32.8% of total data for all modalities (2.20 Tb) in the main PACS and mini-PACS combined. At the end of this period, the volumetric datasets of 1892 and 5162 examinations were kept online and nearline, respectively. Mini-PACS offers an effective method of archiving every volumetric dataset and delivering it to radiologists.

Database Management Systems↗

PubMed2006

Group testing for pathway analysis improves comparability of different microarray datasets.

MOTIVATION: The wide use of DNA microarrays for the investigation of the cell transcriptome triggered the invention of numerous methods for the processing of microarray data and lead to a growing number of microarray studies that examine the same biological conditions. However, comparisons made on the level of gene lists obtained by different statistical methods or from different datasets hardly converge. We aimed at examining such discrepancies on the level of apparently affected biologically related groups of genes, e.g. metabolic or signalling pathways. This can be achieved by group testing procedures, e.g. over-representation analysis, functional class scoring (FCS), or global tests. RESULTS: Three public prostate cancer datasets obtained with the same microarray platform (HGU95A/HGU95Av2) were analyzed. Each dataset was subjected to normalization by either variance stabilizing normalization (vsn) or mixed model normalization (MMN). Then, statistical analysis of microarrays was applied to the vsn-normalized data and mixed model analysis to the data normalized by MMN. For multiple testing adjustment the false discovery rate was calculated and the threshold was set to 0.05. Gene lists from the same method applied to different datasets showed overlaps between 42 and 52%, while lists from different methods applied to the same dataset had between 63 and 85% of genes in common. A number of six gene lists obtained by the two statistical methods applied to the three datasets was then subjected to group testing by Fisher's exact test. Group testing by GSEA and global test was applied to the three datasets, as well. Fisher's exact test followed by global test showed more consistent results with respect to the concordance between analyses on gene lists obtained by different methods and different datasets than the GSEA. However, all group testing methods identified pathways that had already been described to be involved in the pathogenesis of prostate cancer. Moreover, pathways recurrently identified in these analyses are more likely to be reliable than those from a single analysis on a single dataset.

Biomarkers, Tumor↗

PubMed2004

Comparing phylogenetic signal in intraspecific and interspecific body size datasets.

Phylogenetic comparative methods have become a standard statistical approach for analysing interspecific data, under the assumption that traits of species are more similar than expected by chance (i.e. phylogenetic signal is present). Here I test for phylogenetic signal in intraspecific body size datasets to evaluate whether intraspecific datasets may require phylogenetic analysis. I also compare amounts of phylogenetic signal in intraspecific and interspecific body size datasets. Some intraspecific body size datasets contain significant phylogenetic signal. Detection of significant phylogenetic signal was dependant upon the number of populations (n) and the amount of phylogenetic signal (K) for a given dataset. Amounts of phylogenetic signal do not differ between intraspecific and interspecific datasets. Further, relationships between significance of phylogenetic signal and sample size and amount of phylogenetic signal are similar for intraspecific and interspecific datasets. Thus, intraspecific body size datasets are similar to interspecific body size datasets with respect to phylogenetic signal. Whether these results are general for all characters requires further study.

Animals↗

PubMed2003

The BrainIT group: concept and core dataset definition.

INTRODUCTION: An open collaborative international network has been established which aims to improve inter-centre standards for collection of high-resolution, neurointensive care data on patients with traumatic brain injury. The group is also working towards the creation of an open access, detailed and validated database that will be useful for post-hoc hypothesis testing. In Part A, the underlying concept, the group coordination structure, membership guidelines and database access and publication criteria are described. Secondly, in part B, we describe a set of meetings funded by the EEC that allowed us to define a "Core Dataset" and we present the results of a feasibility exercise for collection of this core dataset. METHODS: Four group meetings funded by the EEC have enabled definition of a "Core Dataset" to be collected from all centres regardless of specific project aim. A paper based pilot collection of data was conducted to determine the feasibility for collection of the core dataset. Specially designed forms to collect the core dataset demographic and clinical information as well as sample the time-series data elements were distributed by both email and standard mail to 22 BrainIT centres. A deadline of two months was set to receive completed forms back from centres. A pilot data collection of minute by minute physiological monitoring data was also performed. FINDINGS: A core-dataset was defined and can be downloaded from the BrainIT web-site (go to "Core dataset" link at: www.brainit.org). Eighteen centres (82%) returned completed forms by the set deadline. Overall the feasibility for collection of the core data elements was high with only 10 of the 64 questions (16%) showing missing data. Of those 10 fields with missing data, the average number of centres not responding was 12% and the median 6%. An SQL database to hold the data has been designed and is being tested. Software tools for collection of the core dataset have been developed. Ethics approval has been granted for collection of multi-centre data as part of a pilot data collection study. INTERPRETATION: The BrainIT network provides a more standardised and higher resolution data collection mechanism for research groups, organisations and the device industry to conduct multi-centre trials of new health care technology in patients with traumatic brain injury.

Brain Injuries↗

PubMed2002

Evaluation of land use/land cover datasets for urban watershed modeling.

Land use/land cover (LULC) data are a vital component for nonpoint source pollution modeling. Most watershed hydrology and pollutant loading models use, in some capacity, LULC information to generate runoff and pollutant loading estimates. Simple equation methods predict runoff and pollutant loads using runoff coefficients or pollutant export coefficients that are often correlated to LULC type. Complex models use input variables and parameters to represent watershed characteristics and pollutant buildup and washoff rates as a function of LULC type. Whether using simple or complex models an accurate LULC dataset with an appropriate spatial resolution and level of detail is paramount for reliable predictions. The study presented in this paper compared and evaluated several LULC dataset sources for application in urban environmental modeling. The commonly used USGS LULC datasets have coarser spatial resolution and lower levels of classification than other LULC datasets. In addition, the USGS datasets do not accurately represent the land use in areas that have undergone significant land use change during the past two decades. We performed a watershed modeling analysis of three urban catchments in Los Angeles, California, USA to investigate the relative difference in average annual runoff volumes and total suspended solids (TSS) loads when using the USGS LULC dataset versus using a more detailed and current LULC dataset. When the two LULC datasets were aggregated to the same land use categories, the relative differences in predicted average annual runoff volumes and TSS loads from the three catchments were 8 to 14% and 13 to 40%, respectively. The relative differences did not have a predictable relationship with catchment size.

Cities↗

PubMed2026

Dataset Readiness Assessment With Large Language Model (DRAFT-LLM): A Multi-Axis Audit Guided by LLM.

This article details the Dataset Readiness Assessment for Training (DRAFT), a systematic method for determining whether a high-dimensional biological dataset is suitable for developing reliable, equitable (i.e., the extent to which model performance, error patterns, and potential benefits or harms are evaluated and found to be acceptably distributed across relevant demographic, biological, clinical, and contextual subgroups), and scientifically meaningful machine-learning models, and DRAFT Large Language Model (DRAFT-LLM), its optional human-in-the-loop extension for calibrating study-specific audits through structured, critically reviewed LLM guidance. Standard model validation often fails to detect when apparent performance is driven by spurious correlations, technical artifacts, or hidden stratification, leading to irreproducible and inequitable findings. DRAFT-LLM addresses this gap by shifting the focus from model tuning to structured dataset auditing, organized around Support Protocols 1 to 4 that capture the scientific intent, data structure, and governance constraints of a given study. These Support Protocols: (1) elicit and formalize investigator input into a study intake and dataset card; (2) compute standardized dataset statistics and structural summaries suitable for downstream analysis and LLM context; (3) configure the language model using form-based responses, safety guardrails, and governance rules; and (4) generate personalized instructions, prompts, and code templates for running DRAFT audits. Basic Protocols 1 to 3 are instantiated from this support layer for generalization, equity, and stability: they are reusable execution patterns whose concrete behavior is determined by the cards, statistics, and configurations defined in the Support Protocols. DRAFT-LLM and DRAFT are demonstrated in this article through an end-to-end case study on The Cancer Genome Atlas (TCGA). © 2026 Wiley Periodicals LLC. Support Protocol 1: Study intake and dataset card construction Support Protocol 2: Dataset structure and advanced summary statistics for LLM context Support Protocol 3: LLM configuration using structured form responses Support Protocol 4: Generation of personalized instructions for DRAFT audits Basic Protocol 1: Generalization audit Basic Protocol 2: Equity audit Basic Protocol 3: Stability audit.

Large Language Models↗

PubMed2026

Freely available genomic datasets for atrial fibrillation research: current resources and analytical pipeline.

Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia, characterized by clinical and genetic heterogeneity. Increasing use of genomics and other omics approaches has driven reliance on publicly available AF datasets to advance biological discovery. Thus, this systematic review aimed to identify freely available genomic AF datasets through Mendeley Data and its interconnected repositories, and to characterize the most common analyses performed on these data. The search was conducted in adherence to the PRISMA 2020 guideline. Nineteen freely available genomic AF datasets were identified: Summary statistics for 'Biobank-driven genomic discovery yields new insight into atrial fibrillation biology', hum0014.v8.58qt.v1, AF GWAS in UK Biobank, UK Biobank (Publication 9659), GWAS summary statistics from a 2025 multi-ancestry AF meta-analysis, GSE115574, GSE128188, GSE14975, GSE2240, GSE238242, GSE254133, GSE261170, GSE271748, GSE271839, GSE293813, GSE294456, GSE31821, GSE41177, and GSE79768. The GEO datasets were further examined using differential gene expression, functional enrichment, protein-protein interaction networks, hub gene analysis, microRNA target prediction, and gene clustering, as well as, for the more recently deposited datasets, eQTL colocalization, single-cell/single-nucleus clustering, cell-cell communication analysis, and gene-dosage-dependent transcriptional and electrophysiological profiling. These analyses show some consistency but also considerable heterogeneity in initial conditions, data normalization, and analytical methodological settings. In conclusion, only a limited number of datasets are freely available, so additional, well-characterized and standardized datasets are needed to provide a complete picture of the AF pathology.

Mendeley Data↗

PubMed2000

Towards a minimum dataset in urology.

OBJECTIVE: To establish and prospectively collect data for a minimum dataset in urology by agreeing a dataset amongst a group of urologists, designing structured forms around the data items and implementing them in clinical practice. METHODS: Consultant urologists decided the initial dataset. Structured forms, which incorporated the dataset items, were designed. These forms were piloted before implementation in two urology departments. They were used to document clinical information instead of writing this in the traditional medical record. After use forms were 'scanned' before being filed in the medical record, thus storing the data in an electronic format. RESULTS: Nine forms were designed; inpatient and outpatient activity was captured separately. There were six outpatient forms; a generic new patient form and five specific follow-up forms (lower urinary tract symptoms, oncology, erectile dysfunction, female urology and stones). Three generic inpatient forms were designed. The forms were used by nine registrars, five senior house officers and six clinical nurse practitioners, who required minimal training. CONCLUSION: It is possible to agree a minimum dataset amongst a relatively small group of urologists. The information can be acquired prospectively using structured forms instead of the traditional medical record. This locally agreed dataset could form the basis for a national consensus on a minimum dataset in urology.

Data Collection↗

PubMed2006

Frequency estimates from prescription drug datasets (revision of #04-11-066A).

PURPOSE: Accurate information about the number of times a drug is prescribed or dispensed annually is important to marketers, pharmacoepidemiologists, and patient safety researchers. Yet there is no standard reference for prescribing frequency data. The multiple sources that do exist vary in their sampling methods, target populations, nomenclature, and methods of tallying individual medications prescribed or dispensed. These differences are likely to create ambiguity and contradictions in the scientific literature, but they are not well understood. METHODS: We conducted a descriptive study to examine the similarities and differences between five well-known sources of prescribing frequency data: the National Ambulatory Medical Care Survey (NAMCS), the National Hospital Ambulatory Medical Care Survey (NHAMCS, emergency department and outpatient department), the IMS National Prescription Drug Audit, the Solucient outpatient dataset, and the Solucient inpatient dataset. We compared survey methods, costs, overall frequencies, number of unique names in each database, correlations between frequency estimates from different databases, the extent of overlap in the databases, and nomenclature differences between and within datasets. RESULTS: All the correlations between frequency estimates derived from different datasets were significant, but the frequency estimates differed considerably. The lowest correlation (0.17) was found between the IMS and emergency department of the NHAMCS, and the highest correlation (0.93) was between IMS and Solucient outpatient data. CONCLUSIONS: Although there were significant correlations between frequency estimates for comparable datasets, sampling methods and nomenclature choices resulted in important differences both for individual drug products and for overall frequency statistics. Researchers need to be aware of the differences when deriving drug frequency with these datasets.

Databases as Topic↗

PubMed2005

The use of axial reconstructed images from three-dimensional MRI datasets for morphological diagnosis of meniscal tears of the knee.

INTRODUCTION: In recent years, three-dimensional (3D) MRI has been utilized to detect meniscal tears and has displayed several advantages over 2D MRI. The diagnostic performance to discriminate the type of meniscal tears by 2D MRI (sagittal and coronal images) and axial images from 3D MRI datasets has not been reported yet. The aim of the present study is to evaluate the efficacy of the axial reconstructed images from 3D MRI datasets for the morphological diagnosis of meniscal tears of the knee. MATERIALS AND METHODS: Seventy-four menisci in 37 patients who underwent arthroscopic surgery of the knee were studied. All patients were examined by 2D sagittal and coronal MRI and by axial images from 3D MRI datasets prior to surgery. First, we compared the diagnostic performance for meniscal tears by 2D sagittal and coronal MRI to that by axial images from 3D MRI datasets as judged by arthroscopic findings. Second, we compared the correlation of tear types presumed from 2D sagittal and coronal MRI to arthroscopy with those presumed from axial images from 3D MRI datasets to arthroscopy. RESULTS: For the diagnostic performance of meniscal tears on 2D MRI, the sensitivity, specificity and accuracy were 93.5%, 88.4% and 90.5%, respectively. On axial images from 3D MRI, the sensitivity, specificity and accuracy were 96.8%, 79.1% and 86.5%, respectively. Nine cases were false-positive on axial images, while 5 cases were false-positive on 2D images. Correlations to arthroscopy were 100% by both scans for longitudinal tears and discoid meniscus tears. Only 29% were correctly diagnosed by 2D sagittal and coronal images for radial tears, however, all radial tears could be correctly diagnosed by axial images. All horizontal tears were correctly diagnosed on 2D sagittal and coronal images, while none of the horizontal tears could be correctly diagnosed from axial images. CONCLUSION: Axial images from 3D MRI datasets were useful in the diagnosis of radial tears, but two limitations are noted concerning the use of axial images. First, medial menisci should be carefully read on axial slices because of the relatively high rate of false-positives on medial menisci. Second, axial images cannot discriminate horizontal tears from other types of meniscus tears.

Adolescent↗

PubMed2006

Mammographic masses characterization based on localized texture and dataset fractal analysis using linear, neural and support vector machine classifiers.

OBJECTIVE: Localized texture analysis of breast tissue on mammograms is an issue of major importance in mass characterization. However, in contrast to other mammographic diagnostic approaches, it has not been investigated in depth, due to its inherent difficulty and fuzziness. This work aims to the establishment of a quantitative approach of mammographic masses texture classification, based on advanced classifier architectures and supported by fractal analysis of the dataset of the extracted textural features. Additionally, a comparison of the information content of the proposed feature set with that of the qualitative characteristics used in clinical practice by expert radiologists is presented. METHODS AND MATERIAL: An extensive set of textural feature functions was applied to a set of 130 digitized mammograms, in multiple configurations and scales, constructing compact datasets of textural "signatures" for benign and malignant cases of tumors. These quantitative textural datasets were subsequently studied against a set of a thorough and compact list of qualitative texture descriptions of breast mass tissue, normally considered under a typical clinical assessment, in order to investigate the discriminating value and the statistical correlation between the two sets. Fractal analysis was employed to compare the information content and dimensionality of the textural features datasets with the qualitative information provided through medical diagnosis. A wide range of linear and non-linear classification architectures was employed, including linear discriminant analysis (LDA), least-squares minimum distance (LSMD), K-nearest-neighbors (K-nn), radial basis function (RBF) and multi-layer perceptron (MLP) artificial neural network (ANN), as well as support vector machine (SVM) classifiers. The classification process was used as the means to evaluate the inherent quality and informational content of each of the datasets, as well as the objective performance of each of the classifiers themselves in real classification of mammographic breast tumors against verified diagnosis. RESULTS: Textural features extracted at larger scales and sampling box sizes proved to be more content-rich than their equivalents at smaller scales and sizes. Fractal analysis on the dimensionality of the textural datasets verified that reduced subsets of optimal feature combinations can describe the original feature space adequately for classification purposes and at least the same detail and quality as the list of qualitative texture descriptions provided by a human expert. Non-linear classifiers, especially SVMs, have been proven superior to any linear equivalent. Breast mass classification of mammograms, based only on textural features, achieved an optimal score of 83.9%, through SVM classifiers.

Artificial Intelligence↗

PubMed2000

Validation of a decision support system for the cytodiagnosis of fine needle aspirates of the breast using a prospectively collected dataset from multiple observers in a working clinical environment.

We have used a 692 case dataset, collected retrospectively by a single observer, to develop decision support systems for the cytodiagnosis of fine needle aspirates of breast lesions. In this study, we use a 322 case dataset that was prospectively collected by multiple observers in a working clinical environment to test two predictive systems, using logistic regression and the multilayer perceptron (MLP) type of neural network. Ten observed features and the patient age were used as input features. The systems were developed using a training set and test set from the single observer dataset and then applied to the multiple observer dataset. For the independent test cases from the single observer dataset, with a threshold set for no false positives on the training set, logistic regression produced a sensitivity of 82% (95% confidence interval 73-91) and a predictive value of a positive result (PV +) of 98% (95-99), the values for the MLP were 79% (69-89) and 100%, respectively. However the performance on the prospective multiple observer dataset was much worse, with a sensitivity of 72% (65-80), and PV + of 97% (94-99) for logistic regression and 67% (60-75) and 91% (85-97) for the MLP. These results suggest that there is considerable interobserver variability for the defined features and that this system is unsuitable for further development in the clinical environment unless this problem can be overcome.

Adolescent↗

PubMed2025

Negative dataset selection impacts machine learning-based predictors for multiple bacterial species promoters.

MOTIVATION: Advances in bacterial promoter predictors based on machine learning have greatly improved identification metrics. However, existing models overlooked the impact of negative datasets, previously identified in GC-content discrepancies between positive and negative datasets in single-species models. This study aims to investigate whether multiple-species models for promoter classification are inherently biased due to the selection criteria of negative datasets. We further explore whether the generation of synthetic random sequences (SRS) that mimic GC-content distribution of promoters can partly reduce this bias. RESULTS: Multiple-species predictors exhibited GC-content bias when using CDS as a negative dataset, suggested by specificity and sensibility metrics in a species-specific manner, and investigated by dimensionality reduction. We demonstrated a reduction in this bias by using the SRS dataset, with less detection of background noise in real genomic data. In both scenarios DNABERT showed the best metrics. These findings suggest that GC-balanced datasets can enhance the generalizability of promoter predictors across Bacteria. AVAILABILITY AND IMPLEMENTATION: The source code of the experiments is freely available at https://github.com/maigonzalezh/MultispeciesPromoterClassifier.

Machine Learning↗

PubMed2001

Identifying 'non-medical' datasets to monitor community health and well-being.

BACKGROUND: The aim of the study was to identify 'non-medical' datasets holding routinely collected information that might be used to measure and monitor the wider determinants of community health and well-being. METHODS: An expert panel discussion, involving public health and environmental health academics and professionals with expertise in a variety of backgrounds (including environmental health, housing, transport, community safety, public health, primary and secondary care), and interrogation of the Office for National Statistics database were carried out for the West Midlands region. The aim was to identify routinely collected 'non-medical' datasets containing information on the following factors: physical environment, crime, housing and homelessness, social services, socio-economic environment including employment, lifestyles, education, leisure and culture, transport and accidents. RESULTS: Fifty-six datasets were identified. Although 43 (77 per cent) were collected at least annually, few (17; 30 per cent) held data that were disaggregated and routinely available at the sub-local authority level. CONCLUSIONS: This study has identified a number of datasets that hold information relevant to health. However, no single dataset is likely to provide information on all dimensions of health and the determinants of health, and local agencies should consider carefully the strengths and weaknesses of each. Through the development of inter-sectoral working and multi-agency involvement at the local level there is now considerable scope to improve the quality of many of these datasets and to promote their use in the measurement and monitoring of community health.

Accidents↗