Dataset Information

Machine learning performance in a microbial molecular autopsy context: A cross-sectional postmortem human population study.

ABSTRACT: BACKGROUND:The postmortem microbiome can provide valuable information to a death investigation and to the human health of the once living. Microbiome sequencing produces, in general, large multi-dimensional datasets that can be difficult to analyze and interpret. Machine learning methods can be useful in overcoming this analytical challenge. However, different methods employ distinct strategies to handle complex datasets. It is unclear whether one method is more appropriate than others for modeling postmortem microbiomes and their ability to predict attributes of interest in death investigations, which require understanding of how the microbial communities change after death and may represent those of the once living host. METHODS AND FINDINGS:Postmortem microbiomes were collected by swabbing five anatomical areas during routine death investigation, sequenced and analyzed from 188 death cases. Three machine learning methods (boosted algorithms, random forests, and neural networks) were compared with respect to their abilities to predict case attributes: postmortem interval (PMI), location of death, and manner of death. Accuracy depended on the method used, the numbers of anatomical areas analyzed, and the predicted attribute of death. CONCLUSIONS:All algorithms performed well but with distinct features to their performance. Xgboost often produced the most accurate predictions but may also be more prone to overfitting. Random forest was the most stable across predictions that included more anatomic areas. Analysis of postmortem microbiota from more than three anatomic areas appears to yield limited returns on accuracy, with the eyes and rectum providing the most useful information correlating with circumstances of death in most cases for this dataset.

SUBMITTER: Zhang Y

PROVIDER: S-EPMC6464165 | biostudies-literature | 2019

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Machine learning performance in a microbial molecular autopsy context: A cross-sectional postmortem human population study.

Zhang Yu Y Pechal Jennifer L JL Schmidt Carl J CJ Jordan Heather R HR Wang Wesley W WW Benbow M Eric ME Sze Sing-Hoi SH Tarone Aaron M AM

PloS one 20190415 4

<h4>Background</h4>The postmortem microbiome can provide valuable information to a death investigation and to the human health of the once living. Microbiome sequencing produces, in general, large multi-dimensional datasets that can be difficult to analyze and interpret. Machine learning methods can be useful in overcoming this analytical challenge. However, different methods employ distinct strategies to handle complex datasets. It is unclear whether one method is more appropriate than others f ...[more]

PMID: 30986212

Similar Datasets

Project description:ObjectivesTo determine the diagnostic accuracy of postmortem magnetic resonance imaging (PM-MRI) and postmortem ultrasound (PM-US) for perinatal autopsy in the same patient cohort, and to determine whether PM-US can provide the same anatomical information as PM-MRI.MethodsIn this prospective, 5-year (July 2014-July 2019) single-center study, we performed 1.5-T PM-MRI and PM-US in an unselected cohort of perinatal deaths. The diagnostic accuracies of both modalities were calculated, using autopsy as the reference standard. As a secondary objective, the concordance rates between the two imaging modalities for the overall main diagnosis and for five anatomical regions (brain, spine, thorax, heart and abdomen) were calculated.ResultsDuring the study period, 136 cases underwent both PM-US and PM-MRI, of which 88 (64.7%) also underwent autopsy. There was no significant difference in the rates of concordance with autopsy between the two modalities for overall diagnosis (PM-US, 86.4% (95% CI, 77.7-92.0%) vs PM-MRI, 88.6% (95% CI, 80.3-93.7%)) or in the sensitivities and specificities for individual anatomical regions. There were more non-diagnostic PM-US than PM-MRI examinations for the brain (22.8% vs 3.7%) and heart (14.7% vs 5.1%). If an 'imaging-only' autopsy had been performed, PM-US would have achieved the same diagnosis as 1.5-T PM-MRI in 86.8% (95% CI, 80.0-91.5%) of cases, with the highest rates of agreement being for spine (99.3% (95% CI, 95.9-99.9%)) and cardiac (97.3% (95% CI, 92.4-99.1%)) findings and the lowest being for brain diagnoses (85.2% (95% CI, 76.9-90.8%)).ConclusionAlthough there were fewer non-diagnostic cases using PM-MRI than for PM-US, the high concordance rate for overall diagnosis suggests that PM-US could be used for triaging cases when PM-MRI access is limited or unavailable. © 2020 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.

Project description:BackgroundDepression is a prevalent mental disorder that is undiagnosed and untreated in half of all cases. Wearable activity trackers collect fine-grained sensor data characterizing the behavior and physiology of users (ie, digital biomarkers), which could be used for timely, unobtrusive, and scalable depression screening.ObjectiveThe aim of this study was to examine the predictive ability of digital biomarkers, based on sensor data from consumer-grade wearables, to detect risk of depression in a working population.MethodsThis was a cross-sectional study of 290 healthy working adults. Participants wore Fitbit Charge 2 devices for 14 consecutive days and completed a health survey, including screening for depressive symptoms using the 9-item Patient Health Questionnaire (PHQ-9), at baseline and 2 weeks later. We extracted a range of known and novel digital biomarkers characterizing physical activity, sleep patterns, and circadian rhythms from wearables using steps, heart rate, energy expenditure, and sleep data. Associations between severity of depressive symptoms and digital biomarkers were examined with Spearman correlation and multiple regression analyses adjusted for potential confounders, including sociodemographic characteristics, alcohol consumption, smoking, self-rated health, subjective sleep characteristics, and loneliness. Supervised machine learning with statistically selected digital biomarkers was used to predict risk of depression (ie, symptom severity and screening status). We used varying cutoff scores from an acceptable PHQ-9 score range to define the depression group and different subsamples for classification, while the set of statistically selected digital biomarkers remained the same. For the performance evaluation, we used k-fold cross-validation and obtained accuracy measures from the holdout folds.ResultsA total of 267 participants were included in the analysis. The mean age of the participants was 33 (SD 8.6, range 21-64) years. Out of 267 participants, there was a mild female bias displayed (n=170, 63.7%). The majority of the participants were Chinese (n=211, 79.0%), single (n=163, 61.0%), and had a university degree (n=238, 89.1%). We found that a greater severity of depressive symptoms was robustly associated with greater variation of nighttime heart rate between 2 AM and 4 AM and between 4 AM and 6 AM; it was also associated with lower regularity of weekday circadian rhythms based on steps and estimated with nonparametric measures of interdaily stability and autocorrelation as well as fewer steps-based daily peaks. Despite several reliable associations, our evidence showed limited ability of digital biomarkers to detect depression in the whole sample of working adults. However, in balanced and contrasted subsamples comprised of depressed and healthy participants with no risk of depression (ie, no or minimal depressive symptoms), the model achieved an accuracy of 80%, a sensitivity of 82%, and a specificity of 78% in detecting subjects at high risk of depression.ConclusionsDigital biomarkers that have been discovered and are based on behavioral and physiological data from consumer wearables could detect increased risk of depression and have the potential to assist in depression screening, yet current evidence shows limited predictive ability. Machine learning models combining these digital biomarkers could discriminate between individuals with a high risk of depression and individuals with no risk.

Project description:OBJECTIVE:To describe parents' experience and views of the postmortem examination after the loss of a baby. DESIGN:Cross sectional survey. SETTING:Hospital with a dedicated bereavement counselling service, Newcastle upon Tyne. PARTICIPANTS:258 women who had attended a bereavement counselling service at the Royal Victoria Infirmary, Newcastle upon Tyne, on at least one occasion after losing a baby during pregnancy or infancy, between October 1996 and October 2000. METHOD:Self completion postal questionnaire incorporating fixed choice and open ended questions. MAIN OUTCOME MEASURES:Number of respondents who were asked if they would agree to a postmortem examination of their baby, and number who agreed to a postmortem examination; reasons for agreeing and not agreeing to a postmortem examination; quality of explanation received; number who regretted their decision to give or withhold consent for a postmortem examination. RESULTS:166 (64%) respondents completed the questionnaire. Of these, 148 (89%) had been asked to agree to a postmortem examination on their baby and 120/148 of these respondents (81%) agreed, most of whom recognised benefits resulting from the examination. 101/117 (86%) respondents believed the findings had been explained appropriately. Nine (7%) of the 120 respondents who had agreed to a postmortem examination regretted their decision. Of the respondents who refused an examination, four (14%) had regrets about their decision. DISCUSSION:Parents viewed the postmortem examination as a useful and necessary tool in helping to discover the reasons why their baby had died. Simplifying the language used to explain findings may further raise parents' understanding of the value of the postmortem examination and ensure that they are satisfied with it. Medical staff involved in consent for postmortem examinations should be fully trained in how to ask for parental consent, the postmortem examination procedure, and how to explain the findings.

Dataset Information

Machine learning performance in a microbial molecular autopsy context: A cross-sectional postmortem human population study.

Publications

Machine learning performance in a microbial molecular autopsy context: A cross-sectional postmortem human population study.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets