Nevin Manimala

J Biomed Inform. 2021 Nov 23:103959. doi: 10.1016/j.jbi.2021.103959. Online ahead of print.

ABSTRACT

BACKGROUND: Scoring systems are highly interpretable and widely used to evaluate time-to-event outcomes in healthcare research. However, existing time-to-event scores are predominantly created ad-hoc using a few manually selected variables based on clinician’s knowledge, suggesting an unmet need for a robust and efficient generic score-generating method.

METHODS: AutoScore was previously developed as an interpretable machine learning score generator, integrated both machine learning and point-based scores in the strong discriminability and accessibility. We have further extended it to the time-to-event outcomes and developed AutoScore-Survival, for generating time-to-event scores with right-censored survival data. Random survival forest provided an efficient solution for selecting variables, and Cox regression was used for score weighting. We implemented our proposed method as an R package. We illustrated our method in a study of 90-day survival prediction for patients in intensive care units and compared its performance with other survival models, the random survival forest, and two traditional clinical scores.

RESULTS: The AutoScore-Survival-derived scoring system was more parsimonious than survival models built using traditional variable selection methods (e.g., penalized likelihood approach and stepwise variable selection), and its performance was comparable to survival models using the same set of variables. Although AutoScore-Survival achieved a comparable integrated area under the curve of 0.782 (95% CI: 0.767-0.794), the integer-valued time-to-event scores generated are favorable in clinical applications because they are easier to compute and interpret.

CONCLUSIONS: Our proposed AutoScore-Survival provides a robust and easy-to-use machine learning-based clinical score generator to studies of time-to-event outcomes. It gives a systematic guideline to facilitate the future development of time-to-event scores for clinical applications.

PMID:34826628 | DOI:10.1016/j.jbi.2021.103959

Behav Processes. 2021 Nov 23:104550. doi: 10.1016/j.beproc.2021.104550. Online ahead of print.

ABSTRACT

The number, duration and depth of social relationships that individuals maintain impact social cognition, but the connection between sociality and other aspects of cognition has hardly been explored. To date, the link between social living and intelligence has been mainly supported by studies on primates, and far fewer tests connecting sociality to cognitive abilities have used other taxa. Here, we present the first comparative study in fishes that examines whether complex social living is associated with better performance on a cognitively demanding spatial task. Using three cooperative, group-living cichlid fish species and three of their non-cooperative, more solitary close relatives, we studied maze learning and employed a new statistical extension for the ‘lme4’ and ‘glmmTMB’ packages in R that allows phylogeny to be included as a random effect term. Across trials, the three cooperative and the three non-cooperative species completed the maze faster, made fewer mistakes, and improved their inhibitory control. Although fish improved their performance, we did not detect any differences in the extent of improvement between cooperative and non-cooperative species. Both the cooperative species and the non-cooperative species took similar amounts of time to complete the maze, had comparable numbers of mistakes, and exhibited similar inhibitory control while in the maze. Our results suggest that living and breeding in complex social groups does not necessarily imply enhancement of other forms of cognition nor, more specifically, an enhanced spatial learning capacity.

PMID:34826584 | DOI:10.1016/j.beproc.2021.104550

Neurotoxicol Teratol. 2021 Nov 23:107052. doi: 10.1016/j.ntt.2021.107052. Online ahead of print.

ABSTRACT

This study was designed to assess whether prenatal cocaine exposure (PCE) is associated with sex differences in behavior problems in middle childhood and whether there are sex differences in the way in which parasympathetic functioning mediates the relations between PCE and behavior problems within a diverse low-income sample. Participants included 164 high risk mother-child dyads including 89 PC exposed children and 75 control children participating in an ongoing longitudinal study. Respiratory sinus arrhythmia (RSA) was measured to assess parasympathetic functioning at 13 months of age and maternal reports of child behavior problems were collected at 7 years of age. Results revealed no significant association between PCE and behavior problems for the full sample. A 2 × 2 Anova revealed a significant interaction between PCE and child sex on internalizing, externalizing, and total behavior problems (F (3, 160) = 13.45, p < .001) with cocaine exposed females averaging the highest behavior problem scores. Results also revealed a statistically significant indirect effect linking cocaine exposure to lower externalizing problems via lower baseline RSA among males. Findings indicate that cocaine exposed females may be more vulnerable to developing behavior problems than cocaine exposed males and that high baseline RSA may present a sex specific risk factor for externalizing problems among exposed males.

PMID:34826569 | DOI:10.1016/j.ntt.2021.107052

Int J Biostat. 2020 Dec 2;17(2):331-348. doi: 10.1515/ijb-2020-0022.

ABSTRACT

We propose a nonparametric test of independence, termed optHSIC, between a covariate and a right-censored lifetime. Because the presence of censoring creates a challenge in applying the standard permutation-based testing approaches, we use optimal transport to transform the censored dataset into an uncensored one, while preserving the relevant dependencies. We then apply a permutation test using the kernel-based dependence measure as a statistic to the transformed dataset. The type 1 error is proven to be correct in the case where censoring is independent of the covariate. Experiments indicate that optHSIC has power against a much wider class of alternatives than Cox proportional hazards regression and that it has the correct type 1 control even in the challenging cases where censoring strongly depends on the covariate.

PMID:34826372 | DOI:10.1515/ijb-2020-0022

Int J Biostat. 2021 Jan 13;17(2):317-329. doi: 10.1515/ijb-2020-0136.

ABSTRACT

Gradient boosting from the field of statistical learning is widely known as a powerful framework for estimation and selection of predictor effects in various regression models by adapting concepts from classification theory. Current boosting approaches also offer methods accounting for random effects and thus enable prediction of mixed models for longitudinal and clustered data. However, these approaches include several flaws resulting in unbalanced effect selection with falsely induced shrinkage and a low convergence rate on the one hand and biased estimates of the random effects on the other hand. We therefore propose a new boosting algorithm which explicitly accounts for the random structure by excluding it from the selection procedure, properly correcting the random effects estimates and in addition providing likelihood-based estimation of the random effects variance structure. The new algorithm offers an organic and unbiased fitting approach, which is shown via simulations and data examples.

PMID:34826371 | DOI:10.1515/ijb-2020-0136

Cureus. 2021 Oct 20;13(10):e18918. doi: 10.7759/cureus.18918. eCollection 2021 Oct.

ABSTRACT

Aim and objective Diabetes mellitus is a chronic metabolic disorder that requires continuous self-management practices. The aim of our study is to assess the factors resulting in non-compliance with self-management practices in people with type 2 diabetes mellitus (T2DM). Methods This cross-sectional study was conducted at Baqai Institute of Diabetology and Endocrinology (BIDE), a tertiary care center in Karachi, Pakistan, from March 2019 to May 2019. People with T2DM diagnosed for at least six months were included. A predesigned questionnaire was used to assess various components of self-management such as the use of oral hypoglycemic agents (OHAs) and insulin, self-monitoring of blood glucose (SMBG), physical activity, and daily foot care. Certified diabetes educators conducted interviews on a one-to-one basis. Data were entered and analyzed by using SPSS (version 20; IBM Corp., Armonk, NY). Results Better glycated hemoglobin (HbA1c) levels were observed in compliant persons and a statistically significant difference was noted in those who were compliant with insulin use. Good compliance with self-management was observed in people who were given diabetes education previously. A total of 205 people with T2DM were included in the study, with a mean age of 52.66 ± 11.2 years and a mean duration of diabetes of 8.9 ± 7.5 years. There were 62.9% males and 37.1% females. Oral hypoglycemic agents (OHAs) were prescribed to 62.9% while 33.9 % were on both OHAs and insulin. Non-compliance with the intake of OHAs was 33.3%, insulin injection 21%, SMBG 25.7%, physical activity 69.5%, and foot care practice 34.3%. Various reasons identified for non-compliance included forgetfulness (negligence) (88%), fear of hypoglycemia (10.6%), time constraints (48%), and lack of foot care knowledge (84.8%). Conclusion Non-compliance with T2DM self-management is multifactorial and needs continuous reinforcement of structured diabetes education sessions. The study showed that the provision of diabetes education is directly proportional to self-management compliance levels.

PMID:34826318 | PMC:PMC8603089 | DOI:10.7759/cureus.18918

Br J Radiol. 2021 Nov 26:20210702. doi: 10.1259/bjr.20210702. Online ahead of print.

ABSTRACT

OBJECTIVES: The aim of this exploratory study was to evaluate whether three-dimensional texture analysis (3D-TA) features of non-contrast-enhanced T1-weighted MRI associate with traditional prognostic factors and disease-free survival (DFS) of breast cancer.

METHODS: 3D-T1-weighted images from 78 patients with 81 malignant histopathologically verified breast lesions were retrospectively analysed using standard-size volumes of interest. Grey-level co-occurrence matrix (GLCM) based features were selected for statistical analysis. In statistics the Mann-Whitney U and the Kruskal-Wallis tests, the Cox proportional hazards model and the Kaplan-Meier method were used.

RESULTS: Tumours with higher histological grade were significantly associated with higher contrast (1voxel: p = 0.033, two voxels: p = 0.036). All the entropy parameters showed significant correlation with tumour grade (p = 0.015-0.050) but there were no statistically significant associations between other TA parameters and tumour grade. The Nottingham Prognostic Index (NPI) was correlated with contrast and sum entropy parameters. A higher sum variance TA parameter was a significant predictor of shorter DFS.

CONCLUSION: Texture parameters, assessed by 3D-TA from non-enhanced T1-weighted images, indicate tumour heterogeneity but have limited independent prognostic value. However, they are associated with tumour grade, NPI, and DFS. These parameters could be used as an adjunct to contrast-enhanced TA parameters.

ADVANCES IN KNOWLEDGE: 3D texture analysis of non-contrast enhanced T1-weighted breast MRI associates with tumour grade, NPI, and DFS. The use of non-contrast 3D TA parameters in adjunct with contrast-enhanced 3D TA parameters warrants further research.

PMID:34826254 | DOI:10.1259/bjr.20210702

Disabil Rehabil. 2021 Nov 26:1-11. doi: 10.1080/09638288.2021.2003453. Online ahead of print.

ABSTRACT

PURPOSE: To study psychological distress and fatigue among persons with spina bifida (SB) 50 years or older and living in Norway.

METHODS: In 2017, cross-sectional data were collected (n = 30). The Hospital Anxiety and Depression Scale (HADS) and the Fatigue Severity Scale (FSS) were used. Descriptive statistics, non-parametric tests, and Spearman correlations were performed. Relevant information from previous studies on adults with chronic spinal cord injury (SCI) and the general population, were collected for comparison.

RESULTS: Participants were 18 women and 12 men, mean age 57.5 (SD 5.6), 26 with myelomeningocele, and six with hydrocephalus. Thirty percent scored above the HADS-A- and 20% above the HADS-D thresholds, thus in the same range as previous studies of SB, but higher compared to persons with SCI and norms. HADS-D correlated with pain and FSS scores. Forty percent reported fatigue symptoms (9/15 without hydrocephalus, 3/6 with hydrocephalus).

CONCLUSIONS: The study revealed a high prevalence of fatigue symptomatology among middle-aged and older adults with SB. Symptoms of anxiety and depression were more common than among persons with chronic SCI and norms. SB follow-ups should include awareness of psychological distress and fatigue, and investigate pain and medication side effects among possible influencing factors.IMPLICATIONS FOR REHABILITATIONClinicians treating adult persons with SB should be aware of possible psychological distress and fatigue symptomatology among these patients.We suggest an initial screening for psychological distress and fatigue in persons with SB during follow-up visits and rehabilitation.Interventions to reduce pain may influence levels of psychological distress and/or fatigue in patients with SB.Clinicians should enquire about the effects of medication on fatigue when assessing and prescribing new medications; a thorough medication review helps to assess the benefits and risks.

PMID:34826231 | DOI:10.1080/09638288.2021.2003453

Mol Genet Genomic Med. 2021 Nov 26:e1844. doi: 10.1002/mgg3.1844. Online ahead of print.

ABSTRACT

BACKGROUND: Variants in the ATP binding cassette protein subfamily D member 1 (ABCD1) gene are known to cause X-linked adrenoleukodystrophy (X-ALD). This study focused on the characteristics of ABCD1 variants in Chinese X-ALD families and elucidated the value of genetic approaches for X-ALD.

METHODS: 68 male probands diagnosed as X-ALD were screened for ABCD1 variants by the Sanger sequencing of polymerase chain reaction (PCR) products and multiplex ligation-dependent probe amplification (MLPA) combined with long-range PCR. Prenatal diagnosis was performed in 20 foetuses of 17 probands’ mothers. Descriptive statistics were used to summarise the gene variants and prenatal diagnosis characteristics and outcomes.

RESULTS: This study allowed the identification of 61 variants occurring in 68 families, including 58 single nucleotide variants or small deletion/insertion variants and 3 large deletions. Three probands with no variants detected by next-generation sequencing were found to have variants by PCR-sequencing. Prenatal diagnosis found that 10 of the 20 foetuses had no variants in ABCD1.

CONCLUSION: PCR primers that do not amplify the pseudogenes must be used for PCR-sequencing. MLPA combined with long-range PCR can detect large deletions and insertions, which are usually undetectable by PCR-sequencing. Prenatal diagnosis could help to prevent the birth of infants with X-ALD.

PMID:34826210 | DOI:10.1002/mgg3.1844

Integr Environ Assess Manag. 2021 Nov 26. doi: 10.1002/ieam.4560. Online ahead of print.

ABSTRACT

The pharmaceutical manufacturing industry, via the AMR Industry Alliance, has developed and implemented steps to help minimize the potential impact of pharmaceutical manufacturing on the spread of AMR. One of these steps was to publish Predicted No-Effect Concentrations (PNECs) to serve as targets for antibiotic manufacturing wastewater effluent risk assessments aimed to help protect environmental receptors and to mitigate against the spread of antibiotic resistance. Concentrations below which adverse effects in the environment are not expected to occur (predicted no effect concentrations, PNECs) were first published in 2018 and are updated annually. The current list now stands at 125 antibiotics; however, it is recognized that this list does not encompass all manufactured antibiotics. Therefore, a statistical evaluation of currently available data was conducted and a default PNEC of 0.05 µg/L for antibiotics in the absence of other data was derived. This article is protected by copyright. All rights reserved.

PMID:34826209 | DOI:10.1002/ieam.4560