Category: Nevin Manimala

Neural Netw. 2021 Feb 27;139:149-157. doi: 10.1016/j.neunet.2021.02.014. Online ahead of print.

ABSTRACT

Network initialization is the first and critical step for training neural networks. In this paper, we propose a novel network initialization scheme based on the celebrated Stein’s identity. By viewing multi-layer feedforward sigmoidal neural networks as cascades of multi-index models, the projection weights to the first hidden layer are initialized using eigenvectors of the cross-moment matrix between the input’s second-order score function and the response. The input data is then forward propagated to the next layer and such a procedure can be repeated until all the hidden layers are initialized. Finally, the weights for the output layer are initialized by generalized linear modeling. Such a proposed SteinGLM method is shown through extensive numerical results to be much faster and more accurate than other popular methods commonly used for training neural networks.

PMID:33706228 | DOI:10.1016/j.neunet.2021.02.014

Caries Res. 2021 Mar 11:1-7. doi: 10.1159/000514168. Online ahead of print.

ABSTRACT

The prevention of dental caries in preschool children is a priority for dental services. The aim of the study was to investigate the caries-preventive effect of 2 fluoride varnishes in caries-active preschool children, and then compare the outcome with a non-varnish control group. After screening, 180 preschool children aged 36-71 months, with at least 1 noncavitated lesion, were enrolled and randomly allocated into 3 parallel groups, namely A: 1.5% ammonium fluoride varnish (Fluor Protector S), B: 5% NaF varnish (Duraphat), and C: professional tooth-cleaning. All children were recalled every third month for intervention and their parents were instructed to have them brush their teeth with a 1,000-ppm fluoride toothpaste twice daily. Caries were recorded at baseline and after 12 months by a calibrated examiner and the incidence was scored on noncavitated (d2) and cavitated (d3) level. We tested differences between the groups with the χ2 and two-sided t tests. One hundred and seventy-two children (95.6%) completed the trial and 56 (32.6%) and 35 (19.2%) developed new d2 and d3 lesions, respectively. Both varnishes reduced the incidence of caries compared with the control group, but there was no significant difference between group A and group B. Compared with group C, the relative risk for developing cavitated lesions was 0.39 (95% CI 0.22-0.62) in group A and 0.26 (95% CI 0.14-0.50) in group B. The total prevented fraction (Δd2d3mft) for group A and group B was 19.9 and 22.5% (p < 0.05), respectively. No adverse effects were observed or reported during the study period. In conclusion, the 2 fluoride varnishes demonstrated an equal capacity to reduce the incidence of caries in caries-active preschool children over a 12-month period in comparison with a control group.

PMID:33706305 | DOI:10.1159/000514168

Cerebrovasc Dis. 2021 Mar 11:1-9. doi: 10.1159/000512604. Online ahead of print.

ABSTRACT

BACKGROUND: Comparison of symptomatic intracranial hemorrhage (SICH) rates between stroke patients treated with bridging therapy (BT) and primary mechanical thrombectomy (PMT) are scarce and difficult to interpret due to baseline differences between both populations.

METHODS: Retrospective analysis of patients with acute ischemic stroke treated with endovascular therapy (BT or PMT) was performed at our center between January 2010 and June 2017.

RESULTS: Six hundred twenty-three patients were included. Global SICH rate was 9% overall: 6.8% in the PMT group and 12.6% in the BT group. The following factors significantly associated with SICH after multivariate analysis: MCA occlusion (p: 0.047), stroke of unknown origin (p: 0.025), BT (p: 0.024), and procedural time over 65 min (p: 0.027). The following variables presented a statistically significant higher frequency in patients treated with PMT: atrial fibrillation (p: 0.005), anticoagulant medication (p < 0.001), wake-up strokes (p < 0.001), atherothrombotic etiology (p < 0.05), combined thrombectomy technique (p: 0.008), longer procedural times (p: 0.025), and favorable outcome at 3 months (p: 0.011). The following variables presented a statistically significant higher frequency in patients treated with BT: antiplatelet medication (p: 0.048), MCA occlusions (p: 0.017), cardioembolic etiology (p < 0.05), stent retriever/aspiration technique (p: 0.008), and SICH (p: 0.013). Patients with MCA occlusions had twice the risk of SICH after BT than after PMT (16.4 and 8.6%, p: 0.038).

CONCLUSIONS: In this clinical series, the SICH rate was higher in patients treated with BT than in those treated with PMT. Relevant differences in baseline (related to IVT contraindications) were found between both groups. Randomized studies of BT versus PMT in populations with similar baseline characteristics might be of interest.

PMID:33706319 | DOI:10.1159/000512604

Arch Oral Biol. 2021 Mar 4;125:105102. doi: 10.1016/j.archoralbio.2021.105102. Online ahead of print.

ABSTRACT

OBJECTIVE: To test the applicability of “Gleiser and Hunt dental staging system modified by Kohler” (GHK) to assess third molar (3 M) development in a Russian population in order to determine the age of majority.

DESIGN: The sample consisted of 918 panoramic radiographs from Russian females (n = 551) and males (n = 367) within the age interval between 8 and 23 years. On each radiograph, 3 M development was classified based on the GHK technique. Statistics tested the data for normality. Mean age and standard deviation were described for each 3 M position. Ordinal logistic regression tested the performance of the technique to classify individuals below or above the 18-year threshold. Receiver Operating Characteristic (ROC) curves were used.

RESULTS: The mean ages and standard deviation (SD) for apex closure in females were 21,11 (SD = 1,47), 21,11 (SD = 1,43), 21,24 (SD = 1,39), and 21,29 (SD = 1,28) years for the teeth #18, 28, 38, and 48, respectively. Among males, the same teeth showed mean closure ages of 20,57 (SD = 1,69), 20,64 (SD = 1,76), 20,68 (SD = 1,68), and 20,81 (SD = 1,62) years, respectively. Area under the curve (AUC) reached 0.904 and 0.915 for classifying females and males below or above the 18-year threshold.

CONCLUSION: The GHK technique was able to describe 3 M development in a Russian population. The statistic model was able to classify individuals below or above the 18-year threshold. However, the outcomes must be carefully interpreted, especially in borderline cases (17-19-year spectrum).

PMID:33706152 | DOI:10.1016/j.archoralbio.2021.105102

J Surg Res. 2021 Mar 8;263:230-235. doi: 10.1016/j.jss.2020.12.038. Online ahead of print.

ABSTRACT

BACKGROUND: Frailty syndrome is an established predictor of adverse outcomes after surgical procedures. Our study aimed to compare the simplified National Surgical Quality Improvement Program 5-factor-modified frailty index (mFI-5) to its prior 11-factor-modified frailty index (mFI-11) with respect to the predictive ability for mortality, postoperative complications, and unplanned 30-d readmission in patients undergoing lower limb amputation.

METHODS: The National Surgical Quality Improvement Program (2005-2012) databank was queried for all geriatric patients (>65 y) who underwent above-knee and below-knee amputations. We calculated each mFI by dividing the number of factors present for a patient by the total number of available factors. To assess the correlation between the mFI-5 and mFI-11, we used Spearman’s rho rank coefficient. We then compared the two indices for each outcome (30-d complication, 30-d mortality, and 30-d readmission) and C-Statistic using predictive models.

RESULTS: A total of 8681 patients were included with mean age of 76 ± 9 y, complication rate 35.8%, mortality rate 10.2%, and readmission rate 15.9%. There was no difference in type of amputation in frail and nonfrail. Correlation between the mFI-5 and mFI-11 was above 0.9 for all outcome measures. Both mFI-5 and mFI-11 indexes had strong predictive ability for mortality, postoperative complications, and 30-d readmissions.

CONCLUSIONS: In patients undergoing major lower limb amputation, we found mFI-5 and the mFI-11 were equally effective in predicting postoperative outcomes. Frailty remained a strong predictor of postoperative complications, mortality, and 30-d readmission.

PMID:33706166 | DOI:10.1016/j.jss.2020.12.038

J Psychiatr Res. 2021 Mar 1;137:232-241. doi: 10.1016/j.jpsychires.2021.02.058. Online ahead of print.

ABSTRACT

The majority of antidepressant medication trials have focused on adult populations (ages 18-65), with much less research in older and younger populations. Moreover, key differences in the efficacy and safety of antidepressants have been identified between these age groups. Ketamine has emerged as a promising new treatment for treatment resistant depression (TRD). The objective of this review is to summarize and synthesize the extant literature on the effectiveness, safety and tolerability of ketamine for depression in special age populations (age ≤18 and ≥ 60). Following PRISMA guidelines, a systematic review was performed, searching EMBASE, PsycInfo, and PubMed from inception through July 2020. Studies reporting the use of any ketamine formulation with variable routes of administration to treat clinically diagnosed depression in adolescents or older adults were included. Thirteen studies were included in the analysis and ten observed rapid (≤2 week latency) antidepressant effects following ketamine treatments, with better outcomes following larger, repeated doses, and in open-label rather than blinded settings. Two case reports in adolescents assessed measures of suicidal ideation and both found ketamine to effectuate rapid anti-suicidal effects. Ketamine appears to be safe and well-tolerated in adolescents and older adults. The small quantity, high heterogeneity, and generally low quality of available studies precludes statistical syntheses and significantly limits the strength of our conclusions. Preliminary proof-of-concept studies are promising, however, rigorously designed randomized controlled trials (RCTs) are still required to ascertain effectiveness, safety and tolerability in these groups.

PMID:33706168 | DOI:10.1016/j.jpsychires.2021.02.058

Phys Med. 2021 Mar 8;83:38-45. doi: 10.1016/j.ejmp.2021.02.023. Online ahead of print.

ABSTRACT

Lung ultrasound (LUS) imaging as a point-of-care diagnostic tool for lung pathologies has been proven superior to X-ray and comparable to CT, enabling earlier and more accurate diagnosis in real-time at the patient’s bedside. The main limitation to widespread use is its dependence on the operator training and experience. COVID-19 lung ultrasound findings predominantly reflect a pneumonitis pattern, with pleural effusion being infrequent. However, pleural effusion is easy to detect and to quantify, therefore it was selected as the subject of this study, which aims to develop an automated system for the interpretation of LUS of pleural effusion. A LUS dataset was collected at the Royal Melbourne Hospital which consisted of 623 videos containing 99,209 2D ultrasound images of 70 patients using a phased array transducer. A standardized protocol was followed that involved scanning six anatomical regions providing complete coverage of the lungs for diagnosis of respiratory pathology. This protocol combined with a deep learning algorithm using a Spatial Transformer Network provides a basis for automatic pathology classification on an image-based level. In this work, the deep learning model was trained using supervised and weakly supervised approaches which used frame- and video-based ground truth labels respectively. The reference was expert clinician image interpretation. Both approaches show comparable accuracy scores on the test set of 92.4% and 91.1%, respectively, not statistically significantly different. However, the video-based labelling approach requires significantly less effort from clinical experts for ground truth labelling.

PMID:33706149 | DOI:10.1016/j.ejmp.2021.02.023

J Basic Clin Physiol Pharmacol. 2021 Mar 11. doi: 10.1515/jbcpp-2020-0378. Online ahead of print.

ABSTRACT

OBJECTIVES: The novel corona virus disease, which was initially reported in China in late 2019, has become a global pandemic affecting 330 million cases. COVID-19 affects predominantly the respiratory system, in addition to other organ systems, mainly the cardiovascular system. One of the hypotheses is that virus entering the target cells by binding to angiotensin converting enzyme 2 affecting hypothalamic pituitary axis could lead to dysautonomia which is measured by heart rate variability (HRV). HRV is a non-invasive measure of autonomic function that facilitates identification of COVID-19 patients at the risk of developing cardiovascular complications. So, we aimed to assess HRV in COVID patients and compare between COVID patients and normal controls.

METHODS: In a case control design, we compared 63 COVID-19 infected patients with 43 healthy controls matched for age and gender. Along with clinical characterization, heart rate variability was evaluated using ambulatory 5 min ECG in lead II and expressed in frequency and time domain measures. Statistical analysis was performed using SPSS 17.0.

RESULTS: Mean age of the study population was 49.1 ± 14.2 years and 71 (66.9%) were males. Frequency domain measures high (HF) and low (LF) frequency powers were significantly decreased in COVID-19 patients compared to controls. HF/LF and LF/HF ratios were not different between groups. Time domain measures rMSSD (root mean square of successive RR interval differences) and SDNN (standard deviation of NN intervals) were significantly increased among COVID-19 subjects. COVID-19 infection was associated with increased parasympathetic activity as defined by rMSSD>40 {adjusted odds ratio 7.609 (95% CI 1.61-35.94); p=0.01} and SDNN>60 {adjusted odds ratio 2.620 (95% CI 1.070-6.44); p=0.035} after adjusting for age, gender and comorbidities.

CONCLUSIONS: Our study results showed increased parasympathetic tone in COVID patients. Early diagnosis of autonomic imbalance in COVID patients is needed to plan management and limit progression of disease.

PMID:33705614 | DOI:10.1515/jbcpp-2020-0378

Ecology. 2021 Mar 11:e03332. doi: 10.1002/ecy.3332. Online ahead of print.

ABSTRACT

Our planet is facing significant changes of biodiversity across spatial scales. Although the negative effects of local biodiversity (α diversity) loss on ecosystem stability are well documented, the consequences of biodiversity changes at larger spatial scales, in particular biotic homogenization, i.e. reduced species turnover across space (β diversity), remain poorly known. Using data from 39 grassland biodiversity experiments, we examine the effects of β diversity on the stability of simulated landscapes while controlling for potentially confounding biotic and abiotic factors. Our results show that higher β diversity generates more asynchronous dynamics among local communities and thereby contributes to the stability of ecosystem productivity at larger spatial scales. We further quantify the relative contributions of α and β diversity to ecosystem stability and find a relatively stronger effect of α diversity, possibly due to the limited spatial scale of our experiments. The stabilizing effects of both α and β diversity lead to a positive diversity-stability relationship at the landscape scale. Our findings demonstrate the destabilizing effect of biotic homogenization and suggest that biodiversity should be conserved at multiple spatial scales to maintain the stability of ecosystem functions and services.

PMID:33705570 | DOI:10.1002/ecy.3332

Epilepsia. 2021 Mar 11. doi: 10.1111/epi.16864. Online ahead of print.

ABSTRACT

OBJECTIVE: Although temporal lobe epilepsy (TLE) is recognized as a system-level disorder, little work has investigated pathoconnectomics from a dynamic perspective. By leveraging computational simulations that quantify patterns of information flow across the connectome, we tested the hypothesis that network communication is abnormal in this condition, studied the interplay between hippocampal- and network-level disease effects, and assessed associations with cognition.

METHODS: We simulated signal spreading via a linear threshold model that temporally evolves on a structural graph derived from diffusion-weighted magnetic resonance imaging (MRI), comparing a homogeneous group of 31 patients with histologically proven hippocampal sclerosis to 31 age- and sex-matched healthy controls. We evaluated the modulatory effects of structural alterations of the neocortex and hippocampus on network dynamics. Furthermore, multivariate statistics addressed the relationship with cognitive parameters.

RESULTS: We observed a slowing of in- and out-spreading times across multiple areas bilaterally, indexing delayed information flow, with the strongest effects in ipsilateral frontotemporal regions, thalamus, and hippocampus. Effects were markedly reduced when controlling for hippocampal volume but not cortical thickness, underscoring the central role of the hippocampus in whole-brain disease expression. Multivariate analysis associated slower spreading time in frontoparietal, limbic, default mode, and subcortical networks with impairment across tasks tapping into sensorimotor, executive, memory, and verbal abilities.

SIGNIFICANCE: Moving beyond descriptions of static topology toward the formulation of brain dynamics, our work provides novel insight into structurally mediated network dysfunction and demonstrates that altered whole-brain communication dynamics contribute to common cognitive difficulties in TLE.

PMID:33705572 | DOI:10.1111/epi.16864