Tag: statistics

IEEE J Biomed Health Inform. 2022 Aug 31;PP. doi: 10.1109/JBHI.2022.3203345. Online ahead of print.

ABSTRACT

Given that depression is one of the most prevalent mental illnesses, developing effective and unobtrusive diagnosis tools is of great importance. Recent work that screens for depression with text messages leverage models relying on lexical category features. Given the colloquial nature of text messages, the performance of these models may be limited by formal lexicons. We thus propose a strategy to automatically construct alternative lexicons that contain more relevant and colloquial terms. Specifically, we generate 36 lexicons from fiction, forum, and news corpuses. These lexicons are then used to extract lexical category features from the text messages. We utilize machine learning models to compare the depression screening capabilities of these lexical category features. Out of our 36 constructed lexicons, 14 achieved statistically significantly higher average F1 scores over the pre-existing formal lexicon and basic bag-of-words approach. In comparison to the pre-existing lexicon, our best performing lexicon increased the average F1 scores by 10%. We thus confirm our hypothesis that less formal lexicons can improve the performance of classification models that screen for depression with text messages. By providing our automatically constructed lexicons, we aid future machine learning research that leverages less formal text.

PMID:36044503 | DOI:10.1109/JBHI.2022.3203345

IEEE Trans Neural Netw Learn Syst. 2022 Aug 31;PP. doi: 10.1109/TNNLS.2022.3201160. Online ahead of print.

ABSTRACT

In this brief, the state estimation problem is investigated for a class of randomly delayed artificial neural networks (ANNs) subject to probabilistic saturation constraints (PSCs) and non-Gaussian noises under the redundant communication channels. A series of mutually independent Bernoulli distributed white sequences are introduced to govern the random occurrence of the time delays, the saturation constraints, and the transmission channel failures. A comprehensive redundant-channel-based communication mechanism is constructed to attenuate the phenomenon of packet dropouts so as to enhance the quality of data transmission. To compensate for the influence of randomly occurring time delays, the corresponding occurrence probability is exploited in the process of particle generation. In addition, an explicit expression of the likelihood function is established based on the statistical information to account for the impact of PSCs and redundant channels. By virtue of the modified operations of particle propagation and weight update, a particle-filter-based state estimation algorithm is proposed with mild restriction on the system type. Finally, an illustrative example with Monte Carlo simulations is provided to demonstrate the effectiveness of the developed state estimation scheme.

PMID:36044499 | DOI:10.1109/TNNLS.2022.3201160

IEEE Trans Med Imaging. 2022 Aug 31;PP. doi: 10.1109/TMI.2022.3203309. Online ahead of print.

ABSTRACT

Multi-class segmentation of cardiac magnetic resonance (CMR) images seeks a separation of data into anatomical components with known structure and configuration. The most popular CNN-based methods are optimised using pixel wise loss functions, ignorant of the spatially extended features that characterise anatomy. Therefore, whilst sharing a high spatial overlap with the ground truth, inferred CNN-based segmentations can lack coherence, including spurious connected components, holes and voids. Such results are implausible, violating anticipated anatomical topology. In response, (single-class) persistent homology-based loss functions have been proposed to capture global anatomical features. Our work extends these approaches to the task of multi-class segmentation. Building an enriched topological description of all class labels and class label pairs, our loss functions make predictable and statistically significant improvements in segmentation topology using a CNN-based post-processing framework. We also present (and make available) a highly efficient implementation based on cubical complexes and parallel execution, enabling practical application within high resolution 3D data for the first time. We demonstrate our approach on 2D short axis and 3D whole heart CMR segmentation, advancing a detailed and faithful analysis of performance on two publicly available datasets.

PMID:36044487 | DOI:10.1109/TMI.2022.3203309

PLoS One. 2022 Aug 31;17(8):e0270186. doi: 10.1371/journal.pone.0270186. eCollection 2022.

ABSTRACT

BACKGROUND: The burden of non-communicable diseases like diabetes and hypertension is increasing worldwide including low-and middle-income countries. Good knowledge of such diseases among young people will make them adopt a healthy lifestyle from an early age, which will, in turn, prevent them from developing such non-communicable diseases. This study aimed to assess the knowledge, attitude, and practice of rural and urban school students regarding diabetes and hypertension. We also aimed to see the differences in the knowledge, attitude, and practice of students from rural vs. urban communities.

METHODS: A school-based cross-sectional study was conducted from May 1 2021 to June 30, 2021, in four schools in Nepal (1 from a metropolitan city, 2 from an urban municipality, and 1 from a rural municipality). The study was conducted among the secondary-level students of classes 9 and 10 in each school. The data were collected from the participants via pre-tested questionnaires and analyzed in the Statistical Packages for Social Sciences (SPSS) version 20.0. Logistic regression analysis was carried out to determine the determinants of knowledge and attitude regarding diabetes and hypertension.

RESULTS: Of 380 respondents, 35.5% were residents of metropolitan city, 37.4% were from the urban municipality and 27.1% were from the rural municipality. The mean age of respondents was 15.61±0.99 years and 51.1% were male. Respondents having a family history of diabetes and hypertension were 21.1% and 37.9% respectively. Respondents from the metropolitan city had significantly higher mean knowledge scores than the respondents from the urban and rural municipality (p<0.001) while there was no significant difference in mean attitude scores. There was significantly higher daily consumption of fruits and vegetables among the participants from rural municipality (p<0.01) while no significant difference was seen in salt consumption and time spent on physical activity. In univariate regression analysis, place of residence, family occupation, parental education, and family history of diabetes and hypertension were significantly associated with good knowledge level. In multivariate analysis, only a higher grade of study (grade 10 in comparison to grade 9) was an independent predictor of a student’s good attitude level.

CONCLUSION: In general, there was a good attitude towards diabetes and hypertension despite poor knowledge. The mean knowledge scores were lower in urban municipality and rural municipality compared to metropolitan city. Low knowledge scores on diabetes and hypertension among the students show an urgent need for school-based interventional programs focusing on non-communicable diseases and lifestyle modification with more emphasis on rural communities.

PMID:36044457 | DOI:10.1371/journal.pone.0270186

PLoS One. 2022 Aug 31;17(8):e0273691. doi: 10.1371/journal.pone.0273691. eCollection 2022.

ABSTRACT

BACKGROUND: COVID-19 is spreading rapidly worldwide, and the population is generally susceptible to SARS-CoV-2, especially those with cancer. Hence, our study aims to design a protocol for a systematic review and meta-analysis of the clinical characteristics and prognoses of lung cancer patients with COVID-19.

METHODS: The protocol is prepared following the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines. The literature will be searched in Embase, Pubmed, the Cochrane Library, LitCovid, and CNKI for potentially eligible articles. The quality of the articles will be used in the Newcastle-Ottawa Quality Assessment Scale (NOS) and Cochrane Handbook for Systematic Reviews of Interventions. Statistical analysis will be performed through RevMan 5 software. This review protocol has been registered in PROSPERO (CRD42022306866).

DISCUSSION: To clarify whether COVID-19 affects the clinical symptoms and prognoses of lung cancer patients. Further study is needed to establish the best evidence-based for the management of lung cancer patients with COVID-19.

CONCLUSION: The definitive conclusion will be important to physicians effectively manage lung cancer patients with COVID-19.

PMID:36044455 | DOI:10.1371/journal.pone.0273691

PLoS One. 2022 Aug 31;17(8):e0272460. doi: 10.1371/journal.pone.0272460. eCollection 2022.

ABSTRACT

Recovery activities during short breaks taken between work tasks are solutions for preventing the impairing effects of accumulated strain. No wonder then that a growing body of scientific literature from various perspectives emerged on this topic. The present meta-analysis is aimed at estimating the efficacy of micro-breaks in enhancing well-being (vigor and fatigue) and performance, as well as in which conditions and for whom are the micro-breaks most effective. We searched the existent literature on this topic and aggregated the existing data from experimental and quasi-experimental studies. The systematic search revealed 19 records, which resulted in 22 independent study samples (N = 2335). Random-effects meta-analyses shown statistically significant but small effects of micro-breaks in boosting vigor (d = .36, p < .001; k = 9, n = 913), reducing fatigue (d = .35, p < .001; k = 9, n = 803), and a non-significant effect on increasing overall performance (d = .16, p = .116; k = 15, n = 1132). Sub-groups analyses on performance types revealed significant effects only for tasks with less cognitive demands. A meta-regression showed that the longer the break, the greater the boost was on performance. Overall, the data support the role of micro-breaks for well-being, while for performance, recovering from highly depleting tasks may need more than 10-minute breaks. Therefore, future studies should focus on this issue.

PMID:36044424 | DOI:10.1371/journal.pone.0272460

Alzheimers Dement. 2022 Aug 31. doi: 10.1002/alz.12764. Online ahead of print.

ABSTRACT

BACKGROUND: An ample literature documents the growing prevalence of dementia and associated costs. Less attention has been paid to decreased dementia incidence in some countries.

METHODS: We analyzed trends in age-standardized dementia, stroke, and ischemic heart disease (the triple threat) incidence rates and population attributable fraction of death and disability attributable to 12 risk factors in 204 countries and territories and 51 regions using Global Burden of Disease 2019 data.

RESULTS: During 1990 to 2019, dementia incidence declined in 71 countries; 18 showed statistically significant declines, ranging from -12.1% (95% uncertainty intervals -16.9 to -6.8) to -2.4% (-4.6 to -0.5). During 2010 to 2019, 16 countries showed non-significant declines. Globally, the burden of the triple threat attributable to air pollution, dietary risks, non-optimal temperature, lead exposure, and tobacco use decreased from 1990 to 2019.

CONCLUSION: The declining incidence of dementia in some countries, despite growing prevalence, is encouraging and urges further investigation.

PMID:36044376 | DOI:10.1002/alz.12764

Elife. 2022 Aug 31;11:e78012. doi: 10.7554/eLife.78012.

ABSTRACT

Omics-based technologies are driving major advances in precision medicine, but efforts are still required to consolidate their use in drug discovery. In this work, we exemplify the use of multi-omics to support the development of 3-chloropiperidines, a new class of candidate anticancer agents. Combined analyses of transcriptome and chromatin accessibility elucidated the mechanisms underlying sensitivity to test agents. Furthermore, we implemented a new versatile strategy for the integration of RNA- and ATAC-seq (Assay for Transposase-Accessible Chromatin) data, able to accelerate and extend the standalone analyses of distinct omic layers. This platform guided the construction of a perturbation-informed basal signature predicting cancer cell lines’ sensitivity and to further direct compound development against specific tumor types. Overall, this approach offers a scalable pipeline to support the early phases of drug discovery, understanding of mechanisms, and potentially inform the positioning of therapeutics in the clinic.

PMID:36043458 | DOI:10.7554/eLife.78012

Nephrology (Carlton). 2022 Aug 31. doi: 10.1111/nep.14108. Online ahead of print.

ABSTRACT

INTRODUCTION: Predicting outcome after transplantation of marginal kidneys is a challenging task. Donor creatinine or estimated glomerular filtration rate (eGFR) are integral components of the respective risk scores. However, there is uncertainty on which of their values obtained successively during procurement is the most suitable.

MATERIAL AND METHODS: This is a retrospective study of 221 adult brain death donors with marginal kidneys, transplanted in 223 recipients. We applied logistic regression analysis to investigate the association between initial (at hospital admission), nadir (lowest), zenith (highest) and terminal (at recovery) donor eGFR with primary non-function (PNF), delayed graft function (DGF), 3- and 12-months graft function and 1- and 3-years patient- and death censored graft survival.

RESULTS: In the multivariate analysis, admission, terminal, and lowest donor eGFR could most accurately predict DGF. The respective ORs [95% CI] were: 0.875 [0.771-0.993], 0.818 [95% CI: 0.726-0.922] and 0.793 [0.689-0.900]. Although not being significant for DGF (OR 0.931 [95% CI: 0.817-1.106]), the highest eGFR was the best predictor of 3-month graft function (adjusted b coefficient 1.161 [95% CI: 0.355-1.968]). Analysis of primary nonfunction showed that determination of initial and highest eGFR proved to be the best predictors. The respective ORs [95% CI] were: 0.804 [0.667-0.968] and 0.750 [0.611-0.919]. There were no differences in the risk associations of each of the four eGFR recordings with patient- and graft survival.

CONCLUSION: The various eGFR recordings determined during the procurement process of marginal donors can predict PNF, DGF and 3- and 12-months graft function. Regarding short term patient- and graft survival, there appears to be impacted by recipient factors rather than donor kidney function. This article is protected by copyright. All rights reserved.

PMID:36043436 | DOI:10.1111/nep.14108

Database (Oxford). 2022 Aug 31;2022:baac069. doi: 10.1093/database/baac069.

ABSTRACT

The coronavirus disease 2019 (COVID-19) pandemic has been severely impacting global society since December 2019. The related findings such as vaccine and drug development have been reported in biomedical literature-at a rate of about 10 000 articles on COVID-19 per month. Such rapid growth significantly challenges manual curation and interpretation. For instance, LitCovid is a literature database of COVID-19-related articles in PubMed, which has accumulated more than 200 000 articles with millions of accesses each month by users worldwide. One primary curation task is to assign up to eight topics (e.g. Diagnosis and Treatment) to the articles in LitCovid. The annotated topics have been widely used for navigating the COVID literature, rapidly locating articles of interest and other downstream studies. However, annotating the topics has been the bottleneck of manual curation. Despite the continuing advances in biomedical text-mining methods, few have been dedicated to topic annotations in COVID-19 literature. To close the gap, we organized the BioCreative LitCovid track to call for a community effort to tackle automated topic annotation for COVID-19 literature. The BioCreative LitCovid dataset-consisting of over 30 000 articles with manually reviewed topics-was created for training and testing. It is one of the largest multi-label classification datasets in biomedical scientific literature. Nineteen teams worldwide participated and made 80 submissions in total. Most teams used hybrid systems based on transformers. The highest performing submissions achieved 0.8875, 0.9181 and 0.9394 for macro-F1-score, micro-F1-score and instance-based F1-score, respectively. Notably, these scores are substantially higher (e.g. 12%, higher for macro F1-score) than the corresponding scores of the state-of-art multi-label classification method. The level of participation and results demonstrate a successful track and help close the gap between dataset curation and method development. The dataset is publicly available via https://ftp.ncbi.nlm.nih.gov/pub/lu/LitCovid/biocreative/ for benchmarking and further development. Database URL https://ftp.ncbi.nlm.nih.gov/pub/lu/LitCovid/biocreative/.

PMID:36043400 | DOI:10.1093/database/baac069