Explaining predictive factors in patient pathways using autoencoders.
Ontology highlight
ABSTRACT:
SUBMITTER: De Oliveira H
PROVIDER: S-EPMC9648714 | biostudies-literature | 2022
REPOSITORIES: biostudies-literature
Publications
Explaining predictive factors in patient pathways using autoencoders.
De Oliveira Hugo H Martin Prodel P Ludovic Lamarsalle L Vincent Augusto A Xiaolan Xie X
PloS one 20221110 11
This paper introduces an end-to-end methodology to predict a pathway-related outcome and identifying predictive factors using autoencoders. A formal description of autoencoders for explainable binary predictions is presented, along with two objective functions that allows for filtering and inverting negative examples during training. A methodology to model and transform complex medical event logs is also proposed, which keeps the pathway information in terms of events and time, as well as the hi ...[more]
Similar Datasets
Project description:ObjectivesEvaluate whether hospital factors, including nurse resources, explain racial differences in Medicare black and white patient surgical outcomes and whether disparities changed over time.DesignRetrospective tapered-match.Setting571 hospitals at two time points (Early Era 2003-2005; Recent Era 2013-2015).Participants6752 black patients and three sets of 6752 white controls selected from 107 001 potential controls (Early Era). 4964 black patients and three sets of 4964 white controls selected from 74 108 potential controls (Recent Era).InterventionsBlack patients were matched to white controls on demographics (age, sex, state and year of procedure), procedure (demographics variables plus 136 International Classification of Diseases (ICD)-9 principal procedure codes) and presentation (demographics and procedure variables plus 34 comorbidities, a mortality risk score, a propensity score for being black, emergency admission, transfer status, predicted procedure time).Outcomes30-day and 1-year mortality.ResultsBefore matching, black patients had more comorbidities, higher risk of mortality despite being younger and underwent procedures at different percentages than white patients. Whites in the demographics match had lower mortality at 30 days (5.6% vs 6.7% Early Era; 5.4% vs 5.7% Recent Era) and 1-year (15.5% vs 21.5% Early Era; 12.3% vs 15.9% Recent Era). Black-white 1-year mortality differences were equivalent after matching patients with respect to presentation, procedure and demographic factors. Black-white 30-day mortality differences were equivalent after matching on procedure and demographic factors. Racial disparities in outcomes remained unchanged between the two time periods spanning 10 years. All patients in hospitals with better nurse resources had lower odds of 30-day (OR 0.60, 95% CI 0.46 to 0.78, p<0.010) and 1-year mortality (OR 0.77, 95% CI 0.65 to 0.92, p<0.010) even after accounting for other hospital factors.ConclusionsSurvival disparities among black and white patients are largely explained by differences in demographic, procedure and presentation factors. Better nurse resources (eg, staffing, work environment) were associated with lower mortality for all patients.
| S-EPMC10186454 | biostudies-literature
Project description:MotivationVariational autoencoders (VAEs) have rapidly increased in popularity in biological applications and have already successfully been used on many omic datasets. Their latent space provides a low-dimensional representation of input data, and VAEs have been applied, e.g. for clustering of single-cell transcriptomic data. However, due to their non-linear nature, the patterns that VAEs learn in the latent space remain obscure. Hence, the lower-dimensional data embedding cannot directly be related to input features.ResultsTo shed light on the inner workings of VAE and enable direct interpretability of the model through its structure, we designed a novel VAE, OntoVAE (Ontology guided VAE) that can incorporate any ontology in its latent space and decoder part and, thus, provide pathway or phenotype activities for the ontology terms. In this work, we demonstrate that OntoVAE can be applied in the context of predictive modeling and show its ability to predict the effects of genetic or drug-induced perturbations using different ontologies and both, bulk and single-cell transcriptomic datasets. Finally, we provide a flexible framework, which can be easily adapted to any ontology and dataset.Availability and implementationOntoVAE is available as a python package under https://github.com/hdsu-bioquant/onto-vae.
| S-EPMC10301695 | biostudies-literature
Project description:Purpose: The intravitreal injection (IVI) of pharmacologic agents is the most commonly performed ocular procedure and is associated with a host of complications. Most IVI-related complications data are derived from randomized controlled clinical trials, which report a high adverse event rate. The nature of these protocol-driven trials limit their applicability to the diverse circumstances seen in routine clinical practice. The goal of this study was to determine the prevalence of patient-reported IVI-related complications, their risk factors, and the manner in which patients sought treatment at a tertiary eye care center.Design: Retrospective, institutional review board-approved study.Participants: Forty-four thousand seven hundred thirty-four injections in 5318 unique patients at the Cleveland Clinic Cole Eye Institute from 2012 through 2016.Methods: Intravitreal injection.Main outcome measures: Complication occurrence within 15 days of injection.Results: From 2012 through 2016, a total of 44734 injections were performed in 5318 unique patients. Overall, complication rates were low, representing 1.9% of all injections, with 1031 unique complications in 685 patients (12.9%). The most common minor complications, or those not requiring intervention, were irritation (n = 312) and subconjunctival hemorrhage (n = 284). The most common serious complications, or those requiring intervention, were corneal abrasion (n = 46) and iritis (n = 31). Most complications (66%) were managed adequately by a telephone or Epic (Epic Systems Corp., Verona, WI) electronic message encounter only. Importantly, no injection protocol parameter, such as type of anesthesia, preparation, or post-injection medication, increased the risk of a complication. However, a patient's gender, age, number of previous injections, and provider strongly influenced the risk of patient-reported complications.Conclusions: Overall, complication rates seen in routine clinical practice were low compared with clinical trial reporting. Providers should feel confident in the safety and administration of IVI during times when follow-up office visits and resources may be limited. When performing an IVI, factors such as a patient's gender, age, number of previous injections, and provider must be taken into account to ensure the best possible outcomes.
| S-EPMC7548755 | biostudies-literature
Project description:BackgroundPeople who present to hospital following self-harm are at high risk of suicide. Despite this, there are considerable variations in the management of this group across hospitals and the factors influencing such variations are not well understood.AimsThe aim of this study was to identify the specific hospital and individual factors associated with care pathways following hospital-presenting self-harm.MethodData on presentations to hospitals by those aged 18 years and over were obtained from the National Self-Harm Registry Ireland for 2017 and 2018. Factors associated with four common outcomes following self-harm (self-discharge, medical and psychiatric admission and psychosocial assessment before discharge) were examined using multilevel Poisson regression models.ResultsCare pathways following self-harm varied across hospitals and were influenced by both hospital and individual factors. Individual factors were primarily associated with self-discharge (including male gender, younger age and alcohol involvement), medical admission (older age, drug overdose as a sole method and ambulance presentations) and psychiatric admission (male gender, methods associated with greater lethality and older age). The hospital admission rate for self-harm was the only factor associated with all outcomes examined. The availability of psychiatric in-patient facilities and specialist mental health staff contributed to variation in psychiatric admissions and psychosocial assessments prior to discharge. Hospital factors explained the majority of observed variation in the provision of psychosocial assessments.ConclusionsCharacteristics of the presenting hospital and hospital admission rates influence the recommended care pathways following self-harm. Provision of onsite mental health facilities and specialist mental health staff has a strong impact on psychiatric care of these patients.
| S-EPMC7745229 | biostudies-literature
Project description:Lineage tracing involves the identification of all ancestors and descendants of a given cell, and is an important tool for studying biological processes such as development and disease progression. However, in many settings, controlled time-course experiments are not feasible, for example when working with tissue samples from patients. Here we present ImageAEOT, a computational pipeline based on autoencoders and optimal transport for predicting the lineages of cells using time-labeled datasets from different stages of a cellular process. Given a single-cell image from one of the stages, ImageAEOT generates an artificial lineage of this cell based on the population characteristics of the other stages. These lineages can be used to connect subpopulations of cells through the different stages and identify image-based features and biomarkers underlying the biological process. To validate our method, we apply ImageAEOT to a benchmark task based on nuclear and chromatin images during the activation of fibroblasts by tumor cells in engineered 3D tissues. We further validate ImageAEOT on chromatin images of various breast cancer cell lines and human tissue samples, thereby linking alterations in chromatin condensation patterns to different stages of tumor progression. Our results demonstrate the promise of computational methods based on autoencoding and optimal transport principles for lineage tracing in settings where existing experimental strategies cannot be used.
| S-EPMC7209334 | biostudies-literature
Project description:Cardiac anatomy and function vary considerably across the human population with important implications for clinical diagnosis and treatment planning. Consequently, many computer-based approaches have been developed to capture this variability for a wide range of applications, including explainable cardiac disease detection and prediction, dimensionality reduction, cardiac shape analysis, and the generation of virtual heart populations. In this work, we propose a variational mesh autoencoder (mesh VAE) as a novel geometric deep learning approach to model such population-wide variations in cardiac shapes. It embeds multi-scale graph convolutions and mesh pooling layers in a hierarchical VAE framework to enable direct processing of surface mesh representations of the cardiac anatomy in an efficient manner. The proposed mesh VAE achieves low reconstruction errors on a dataset of 3D cardiac meshes from over 1,000 patients with acute myocardial infarction, with mean surface distances between input and reconstructed meshes below the underlying image resolution. We also find that it outperforms a voxelgrid-based deep learning benchmark in terms of both mean surface distance and Hausdorff distance while requiring considerably less memory. Furthermore, we explore the quality and interpretability of the mesh VAE's latent space and showcase its ability to improve the prediction of major adverse cardiac events over a clinical benchmark. Finally, we investigate the method's ability to generate realistic virtual populations of cardiac anatomies and find good alignment between the synthesized and gold standard mesh populations in terms of multiple clinical metrics.
| S-EPMC9813669 | biostudies-literature
Project description:MotivationThe completion of the genome has paved the way for genome-wide association studies (GWAS), which explained certain proportions of heritability. GWAS are not optimally suited to detect non-linear effects in disease risk, possibly hidden in non-additive interactions (epistasis). Alternative methods for epistasis detection using, e.g. deep neural networks (DNNs) are currently under active development. However, DNNs are constrained by finite computational resources, which can be rapidly depleted due to increasing complexity with the sheer size of the genome. Besides, the curse of dimensionality complicates the task of capturing meaningful genetic patterns for DNNs; therefore necessitates dimensionality reduction.ResultsWe propose a method to compress single nucleotide polymorphism (SNP) data, while leveraging the linkage disequilibrium (LD) structure and preserving potential epistasis. This method involves clustering correlated SNPs into haplotype blocks and training per-block autoencoders to learn a compressed representation of the block's genetic content. We provide an adjustable autoencoder design to accommodate diverse blocks and bypass extensive hyperparameter tuning. We applied this method to genotyping data from Project MinE, and achieved 99% average test reconstruction accuracy-i.e. minimal information loss-while compressing the input to nearly 10% of the original size. We demonstrate that haplotype-block based autoencoders outperform linear Principal Component Analysis (PCA) by approximately 3% chromosome-wide accuracy of reconstructed variants. To the extent of our knowledge, our approach is the first to simultaneously leverage haplotype structure and DNNs for dimensionality reduction of genetic data.Availability and implementationData are available for academic use through Project MinE at https://www.projectmine.com/research/data-sharing/, contingent upon terms and requirements specified by the source studies. Code is available at https://github.com/gizem-tas/haploblock-autoencoders.
| S-EPMC11208726 | biostudies-literature
Project description:BACKGROUND:Next generation sequencing instruments are providing new opportunities for comprehensive analyses of cancer genomes. The increasing availability of tumor data allows to research the complexity of cancer disease with machine learning methods. The large available repositories of high dimensional tumor samples characterised with germline and somatic mutation data requires advance computational modelling for data interpretation. In this work, we propose to analyze this complex data with neural network learning, a methodology that made impressive advances in image and natural language processing. RESULTS:Here we present a tumor mutation profile analysis pipeline based on an autoencoder model, which is used to discover better representations of lower dimensionality from large somatic mutation data of 40 different tumor types and subtypes. Kernel learning with hierarchical cluster analysis are used to assess the quality of the learned somatic mutation embedding, on which support vector machine models are used to accurately classify tumor subtypes. CONCLUSIONS:The learned latent space maps the original samples in a much lower dimension while keeping the biological signals from the original tumor samples. This pipeline and the resulting embedding allows an easier exploration of the heterogeneity within and across tumor types and to perform an accurate classification of tumor samples in the pan-cancer somatic mutation landscape.
| S-EPMC6907172 | biostudies-literature
Project description:BackgroundAnthocyanins are important contributors to coloration across a wide phylogenetic range of plants. Biological functions of anthocyanins span from reproduction to protection against biotic and abiotic stressors. Owing to a clearly visible phenotype of mutants, the anthocyanin biosynthesis and its sophisticated regulation have been studied in numerous plant species. Genes encoding the anthocyanin biosynthesis enzymes are regulated by a transcription factor complex comprising MYB, bHLH and WD40 proteins.ResultsA systematic comparison of anthocyanin-pigmented vs. non-pigmented varieties was performed within numerous plant species covering the taxonomic diversity of flowering plants. The literature was screened for cases in which genetic factors causing anthocyanin loss were reported. Additionally, transcriptomic data sets from four previous studies were reanalyzed to determine the genes possibly responsible for color variation based on their expression pattern. The contribution of different structural and regulatory genes to the intraspecific pigmentation differences was quantified. Differences concerning transcription factors are by far the most frequent explanation for pigmentation differences observed between two varieties of the same species. Among the transcription factors in the analyzed cases, MYB genes are significantly more prone to account for pigmentation differences compared to bHLH or WD40 genes. Among the structural genes, DFR genes are most often associated with anthocyanin loss.ConclusionsThese findings support previous assumptions about the susceptibility of transcriptional regulation to evolutionary changes and its importance for the evolution of novel coloration phenotypes. Our findings underline the particular significance of MYBs and their apparent prevalent role in the specificity of the MBW complex.
| S-EPMC11221117 | biostudies-literature
Project description:affy_hypoctemp_medicago - The characterization of several genotypes of the model Legume Medicago truncatula showed a genetic variability for germination and hypocotyl heterotrophic growth at low temperature and optimal temperature. The most important contrast was between the accessions Jemalong A17 and F83005.5. In order to find genes differently expressed between temperatures and genotypes, the present work focuses on transcriptome profiling during hypocotyl heterotrophic growth under low (10°C) and optimal (20°C) temperature conditions for both genotypes. We used Jemalong A17 and F83005.5 seeds coming from the Medicago truncatula Biological Resource Center in Montpellier, produced in 2006. Experiments were performed in the dark to mimic pre-emergence growth. Pots (6.5 cm diameter, 10 cm high) were incubated in growth chambers either at 10°C or 20°C. They were filled with 500g sand and 100ml of a nutrient solution for young seedlings growth (Saglio and Pradet, 1980). Per pot, five scarified seeds were sown at 1.5 cm depth. Seedlings were harvested at three times: 35°Cd (degree day), 50°Cd and 100°Cd. At each timepoint, about 50 seedlings were harvested and hypocotyl lengths was measured. Hypocotyl were cut and immediately frozen in liquid nitrogen for RNA extractions. Time course of 24 arrays - Medicago
2010-09-10 | E-GEOD-18449 | biostudies-arrayexpress