Tag: pubmed

Front Artif Intell. 2021 May 18;4:589632. doi: 10.3389/frai.2021.589632. eCollection 2021.

ABSTRACT

Dataset shift refers to the problem where the input data distribution may change over time (e.g., between training and test stages). Since this can be a critical bottleneck in several safety-critical applications such as healthcare, drug-discovery, etc., dataset shift detection has become an important research issue in machine learning. Though several existing efforts have focused on image/video data, applications with graph-structured data have not received sufficient attention. Therefore, in this paper, we investigate the problem of detecting shifts in graph structured data through the lens of statistical hypothesis testing. Specifically, we propose a practical two-sample test based approach for shift detection in large-scale graph structured data. Our approach is very flexible in that it is suitable for both undirected and directed graphs, and eliminates the need for equal sample sizes. Using empirical studies, we demonstrate the effectiveness of the proposed test in detecting dataset shifts. We also corroborate these findings using real-world datasets, characterized by directed graphs and a large number of nodes.

PMID:34179767 | PMC:PMC8223254 | DOI:10.3389/frai.2021.589632

Arch Rehabil Res Clin Transl. 2021 Mar 9;3(2):100119. doi: 10.1016/j.arrct.2021.100119. eCollection 2021 Jun.

ABSTRACT

OBJECTIVE: To compare the effects of physiological saline interfascial and lidocaine trigger point injections in the treatment of myofascial pain syndrome (MPS).

DESIGN: Double-blind randomized controlled study.

SETTING: Department of Rehabilitation Medicine.

PARTICIPANTS: Eighty patients (N=80; 15 men, 65 women; mean age, 40.4±10.9y) with MPS in the upper trapezius muscle.

INTERVENTIONS: Patients were randomly assigned to 1 of 2 groups: group 1 (n=40) received ultrasound-guided interfascial injection with physiological saline, and group 2 (n=40) underwent ultrasound-guided trigger point injection with lidocaine.

MAIN OUTCOME MEASURES: The primary outcome was pain scores measured by the visual analog scale (VAS). Secondary outcomes included cervical range of motion (ROM) and adverse effects of the procedure. The outcomes were assessed before treatment and 10 minutes, 2 weeks, and 4 weeks after treatment.

RESULTS: Lidocaine trigger point injection showed higher VAS score improvement at 10 minutes after the procedure (P=.037). However, there was no statistically significant difference at other follow-up points. Pain scores significantly decreased at 10 minutes, 2 weeks, and 4 weeks from baseline in both groups. Cervical ROM increased significantly over time in some directions without significant differences between the groups. There were no serious adverse effects in this study.

CONCLUSION: Physiologic saline interfascial and lidocaine trigger point injections effectively decreased MPS pain in the upper trapezius muscle at 2 and 4 weeks after treatment. However, lidocaine trigger point injection demonstrated better pain improvement at 10 minutes after treatment.

PMID:34179755 | PMC:PMC8211995 | DOI:10.1016/j.arrct.2021.100119

Arch Rehabil Res Clin Transl. 2021 Mar 9;3(2):100118. doi: 10.1016/j.arrct.2021.100118. eCollection 2021 Jun.

ABSTRACT

OBJECTIVE: To investigate the feasibility and efficacy of short-term functional power training and further examine whether the addition of cognitive training targeting sustained attention and inhibitory control would augment the effect on the outcomes.

DESIGN: Randomized pilot study.

SETTING: Clinical research facility.

PARTICIPANTS: Community-dwelling primary care patients (N=25) aged >65 years with mobility limitation within the VA Boston Healthcare System.

INTERVENTIONS: Participants were randomly assigned to either functional power training (n=14) or functional power+cognitive training (n=11), offered 3 times a week for 6 weeks. Session durations were either 70 minutes (functional power+cognitive training) or 40 minutes (functional power training).

MAIN OUTCOME MEASURES: We evaluated feasibility (dropouts, attendance), mobility performance (Short Physical Performance Battery [SPPB]), leg power [stair climb test]), dynamic balance [figure-of-8], and gait characteristics [gait speed, stance time, step width, swing time, step length, variabilities under single-task and dual-task conditions]). Nonparametric analyses were used to compare overall pre-post changes and between-group differences.

RESULTS: Of the 39 veterans screened, 25 were randomized and enrolled. Twenty-one men with a mean age 76±7 years completed the study; 86% were white. Participants had a mean SPPB score of 8.3±1.6 out of 12. For those completing the study, overall attendance was 79%. Among all participants, clinically relevant and/or statistically significant median change in mobility performance (∆1 point), leg power (∆25.0W), dynamic balance (∆-1.1s), and gait characteristics (gait speed [∆0.08s, ∆0.09s], step length [∆1.9cm, ∆3.8cm], and stance time [∆-0.02s, ∆-0.05s] under single- and dual-task, respectively) were observed after 6 weeks of training. There were no statistically significant group differences in dropouts, attendance rate, or any of the outcomes based on cognitive training status.

CONCLUSIONS: Short-term functional power training with or without a cognitive training led to clinically meaningful improvements in mobility performance, leg power, dynamic balance, and gait characteristics. These findings add to the body of evidence supporting the benefits of functional power training on clinically relevant outcomes. Additional cognitive training did not have an added effect on the study outcomes from our study. Further research is needed.

PMID:34179754 | PMC:PMC8211979 | DOI:10.1016/j.arrct.2021.100118

EClinicalMedicine. 2021 Jul;37:100967. doi: 10.1016/j.eclinm.2021.100967. Epub 2021 Jun 18.

ABSTRACT

BACKGROUND: The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is the etiological agent of coronavirus disease 19 (COVID-19), a respiratory infection that, starting from December 2019, has spread around the world in a few months, becoming a pandemic. The lack of initial knowledge on its management has led to a great effort in developing vaccines and in finding therapeutic weapons capable of improving the clinical outcome of the affected patients. In particular, the possible role of vitamin D status in the management of COVID-19 has been widely analysed, resulting in a great amount of data. This systematic review and meta-analysis aimed to assess whether hypovitaminosis D is a risk factor for developing SARS-CoV-2 infection and whether it affects the worsening of the clinical course of COVID-19.

METHODS: Data were extracted through extensive searches in the Pubmed, MEDLINE, Cochrane, Academic One Files, Google Scholar, and Scopus databases from December 2019 to January 2021, using the keywords: “Vitamin D”, “25 hydroxy Vitamin D”, “25 hydroxycholecalciferol”, “cholecalciferol”, “COVID 19″, “SARS-CoV-2″. We included observational cohort, cross-sectional, and case-control studies that evaluated differences in serum levels of 25‑hydroxy-cholecalciferol [25(OH)D] in patients who were positive or negative for SARS-CoV-2, in patients with mild or severe forms of COVID-19, and in patients who died or were discharged from the hospital. Finally, studies that evaluated the risk of developing severe illness or death in patients with vitamin D deficiency (VDD), defined as levels of 25(OH)D <20 ng/ml, were also included. We calculated the mean difference (MD) and the 95% confidence intervals (CI) for quantitative variables such as 25(OH)D levels in patients with or without SARS-CoV-2 infection, in those with mild vs. severe COVID-19, or those who have died vs. those who have been discharged. Instead, we calculated odds ratios and 95% CI for qualitative ones, such as the number of patients with severe illness/death in the presence of VDD vs. those with normal serum 25(OH)D levels. A p-value lower than 0.05 was considered statistically significant. The study was registered on PROSPERO (CRD42021241473).

FINDINGS: Out of 662 records, 30 articles met inclusion criteria and, therefore, were included in the meta-analysis. We found that the serum levels of 25(OH)D were significantly lower in patients with SARS-CoV-2 infection than in negative ones [MD -3.99 (-5.34, -2.64); p <0.00001; I²= 95%]. Furthermore, its levels were significantly lower in patients with severe disease [MD -6.88 (-9.74, -4.03); p <0.00001; I²=98%] and in those who died of COVID-19 [MD -8.01 (-12.50, -3.51); p = 0.0005; I²=86%]. Finally, patients with VDD had an increased risk of developing severe disease [OR 4.58 (2.24, 9.35); p <0.0001; I²=84%] but not a fatal outcome [OR 4.92 (0.83, 29.31); p = 0.08; I²=94%].

INTERPRETATION: This meta-analysis revealed a large heterogeneity of the studies included due to the different enrolment criteria of patient samples (age, body mass index, ethnicity, comorbidities), the country where they live, all factors influencing serum 25(OH)D levels, and the different criteria used to define the severity of COVID-19. Furthermore, the observational nature of these studies does not allow to establish a cause-effect relationship, even taking into account that 25(OH)D represents a marker of acute inflammation. Treatment with vitamin D might be considered for the primary prevention of SARS-CoV-2 infection and the management of patients with COVID-19. However, further intervention studies are needed to prove this hypothesis.

PMID:34179737 | PMC:PMC8215557 | DOI:10.1016/j.eclinm.2021.100967

Forensic Sci Int Synerg. 2021 Jun 12;3:100157. doi: 10.1016/j.fsisyn.2021.100157. eCollection 2021.

ABSTRACT

Forensic-evaluation systems should output likelihood-ratio values that are well calibrated. If they do not, their output will be misleading. Unless a forensic-evaluation system is intrinsically well-calibrated, it should be calibrated using a parsimonious parametric model that is trained using calibration data. The system should then be tested using validation data. Metrics of degree of calibration that are based on the pool-adjacent-violators (PAV) algorithm recalibrate the likelihood-ratio values calculated from the validation data. The PAV algorithm overfits on the validation data because it is both trained and tested on the validation data, and because it is a non-parametric model with weak constraints. For already-calibrated systems, PAV-based ostensive metrics of degree of calibration do not actually measure degree of calibration; they measure sampling variability between the calibration data and the validation data, and overfitting on the validation data. Monte Carlo simulations are used to demonstrate that this is the case. We therefore argue that, in the context of casework, PAV-based metrics are not meaningful metrics of degree of calibration; however, we also argue that, in the context of casework, a metric of degree of calibration is not required.

PMID:34179740 | PMC:PMC8212664 | DOI:10.1016/j.fsisyn.2021.100157

Arch Rehabil Res Clin Transl. 2021 Feb 23;3(2):100116. doi: 10.1016/j.arrct.2021.100116. eCollection 2021 Jun.

ABSTRACT

OBJECTIVE: To demonstrate the effects of a newly designed postural alignment relearning system on postural control dysfunction in a typical patient with cerebral palsy (CP).

DESIGN: Evaluation before and after 8 weeks of Constraint Standing Training 3-dimensional postural alignment relearning system.

SETTING: Department of Rehabilitation Medicine.

PARTICIPANT: A 6-year-old girl with CP and postural misalignment on Gross Motor Function Classification System level I.

INTERVENTIONS: Constraint Standing Training for 8 weeks to correct postural misalignment.

MAIN OUTCOME MEASURES: Parameters of lateral plain radiographs in static standing, posturography measurements in standing and walking, motor ability (Gross Motor Function Measure-88 [GMFM-88] scores, manual muscle testing [MMT] scores, muscle architecture), and gait kinematic parameters (40 3-dimensional parameters of arms, trunk, waist, and lower limbs).

RESULTS: Knee hyperextension angle in static standing; peaks of knee flexion angle (KFA) when walking, hip flexion angle and ankle flexion angle in dynamic standing; and the KFA at initial contact in gait cycle all decreased significantly (P<.01). Scores of GMFM-88 sections D and E and MMT of 5 core stability muscles improved (P<.01). The velocities and range of motion of the arms, the 3-dimensinoal range of motion of the trunk and waist, and most of the parameters of the lower limbs showed statistically significant change (P<.01). Bilateral muscle thickness did not change significantly after the treatment (P=.738 left, P=.978 right), but the gluteus maximus morphology was changed: the muscle fibers became rounder, the interfiber space decreased, and the border lines of the muscle fibers got clearer.

CONCLUSIONS: Postural alignment, motor ability, and gait may be homologous external manifestations of more fundamental core abilities, referring to correct standing posture cognition, muscle activation, and postural unconsciousness. Constraint Standing Training 3-dimensional postural alignment relearning system aimed to improve the static and dynamic standing control ability, may fix postural misalignment and improve motor ability and flexed-knee gait. Future work should use Constraint Standing Training with patients with different kinds of misalignment, choose sensitive indicators, observe the duration of each step, and reveal the mechanism causes postural misalignment.

PMID:34179752 | PMC:PMC8212001 | DOI:10.1016/j.arrct.2021.100116

SN Compr Clin Med. 2021 Jun 18:1-7. doi: 10.1007/s42399-021-00997-y. Online ahead of print.

ABSTRACT

COVID-19 represented an important challenge to the Italian healthcare system (IHCS). Our main aim was to obtain evidence to support the use of modified national early warning score (m-NEWS) as an interdisciplinary, common, and universal scoring scale to quickly recognize patients with a risk of clinical deterioration before admission and during hospitalization. As a secondary goal, we tried to find a score threshold that can trigger patients’ immediate medical review as a part of an optimal triaging protocol for an emergency setting where healthcare resources are overloaded. We performed a retrospective observational study. We included in our study all patients treated for COVID-19 infection in surgical departments between 01 March 2020 and 16 April 2020. Patients with negative test results for SARS-COV-2 were excluded. m-NEWS was obtained twice a day. Patients’ m-NEWS were analyzed in order to verify the correlation between m-NEWS (at admission and m-NEWS variation 24 h after admission) and outcome (positive outcome-survival, negative outcome-death, or intensive care unit (ICU) transfer). We included a population-based sample of 225 SARS-COV-2-infected patients. Overall, the average age at hospitalization was 71 (ranging from 40 to 95). 144 (64%) patients were males and 81 (36%) females. m-NEWS values lower or equal to 7 were associated with the majority of the “recovered” population (100/132 75.75%) and at the same time with the minority of the “non-recovered” population (25/93 26.88%). For our sample, age is statistically correlated to the outcome but a triage protocol based solely on this variable is less effective than m-NEWS, which showed to be a reliable and easy-to-use score for first patient evaluation. Our observations pave the way towards further studies aiming at optimizing territorial and community healthcare management protocols.

PMID:34179692 | PMC:PMC8211943 | DOI:10.1007/s42399-021-00997-y

Pediatr Investig. 2021 Jan 25;5(2):140-147. doi: 10.1002/ped4.12233. eCollection 2021 Jun.

ABSTRACT

IMPORTANCE: The prevalence and characteristics of short stature (SS) among children in China should be assessed to provide guidance for planning and implementation of nationwide public health policies. Thus far, there have been no accurate estimates of the prevalence of SS in China.

OBJECTIVE: To analyze the prevalence of SS among children in China and to explore the influences of sex, area, age, study year, and study site on prevalence rates.

METHODS: Relevant literature was identified by searching the following databases: PubMed, Embase, The Cochrane Library, Chinese Biomedical Literature, China Knowledge Resource Integrated, WeiPu, and WanFang databases. Meta-analysis was carried out using STATA 11.2.

RESULTS: This meta-analysis included 39 studies with 348 326 Chinese participants; the studies covered 20 provinces, municipalities, and autonomous regions. The pooled prevalence of SS was 3.2% (95% confidence interval [CI], 2.6%-3.7%; I ² = 99.8%). The prevalence of SS in boys and girls were 3.1% (95% CI, 2.5%-3.7%) and 3.2% (95% CI, 2.6%-3.9%), respectively. The sex difference was not statistically significant (P > 0.05). The prevalence of SS was higher in rural areas than in urban areas (4.7% [95% CI, 3.6%-5.8%] vs. 2.8% [95% CI, 2.2%-3.4%]; P < 0.001). The prevalence of SS was higher in West China (5.2%; 95% CI, 4.4%-6.0%) than in Northeast China (0.6%; 95% CI, 0.3%-0.8%), East China (2.3%; 95% CI, 1.9%-2.8%), or Central China (2.9%; 95% CI, 1.9%-3.9%).

INTERPRETATION: The prevalence of SS among children was higher in western and rural areas of China. Close attention to children’s growth and development is needed to prevent the occurrence of SS.

PMID:34179712 | PMC:PMC8212717 | DOI:10.1002/ped4.12233

Pediatr Qual Saf. 2021 Jun 23;6(4):e424. doi: 10.1097/pq9.0000000000000424. eCollection 2021 Jul-Aug.

ABSTRACT

INTRODUCTION: Live video visits for ambulatory encounters offer potential benefits, including access to remote subspecialty services, care coordination between providers, and improved convenience for patients. We aimed to increase the utilization of video visits for pediatric patients at our medical center using an iterative quality improvement process.

METHODS: A multispecialty improvement team identified opportunities to increase video visit utilization and prioritized interventions using benefit-effort analyses. Interventions focused on 6 key drivers. The outcome measure was the percentage of ambulatory encounters conducted by video. The process measure was the percentage of ambulatory pediatricians conducting video visits. The balancing measure was the percentage of no-shows among scheduled video visits. All measures were analyzed using statistical process control.

RESULTS: Interventions were associated with increases in our outcome and process measures from 0.1% to 1.2% and 0.6% to 6.3%, respectively, during the first 8 months. Subsequently, the novel coronavirus (COVID-19) pandemic was associated with further increases in these measures to 41.8% and 73.5%, respectively, over 3 months. The balancing measure increased from 0% at baseline to 14.7% with no special cause variation during the intervention period. The most impactful interventions included clinician training outreach, providing equipment, and streamlining MyChart patient enrollment.

CONCLUSIONS: This improvement project effectively increased pediatric ambulatory video visit utilization, although the most significant driver of utilization was the COVID-19 pandemic. Project interventions implemented before COVID-19 facilitated rapid video visit adoption during the pandemic. A similar improvement process may be beneficial for other medical centers aiming to improve video visit utilization.

PMID:34179675 | PMC:PMC8225361 | DOI:10.1097/pq9.0000000000000424

Nat Mach Intell. 2021 Jun;3(6):536-544. doi: 10.1038/s42256-021-00333-y. Epub 2021 May 10.

ABSTRACT

Recent advances in single-cell technologies, including single-cell ATAC-seq (scATAC-seq), have enabled large-scale profiling of the chromatin accessibility landscape at the single cell level. However, the characteristics of scATAC-seq data, including high sparsity and high dimensionality, have greatly complicated the computational analysis. Here, we proposed scDEC, a computational tool for single cell ATAC-seq analysis with deep generative neural networks. scDEC is built on a pair of generative adversarial networks (GANs), and is capable of learning the latent representation and inferring the cell labels, simultaneously. In a series of experiments, scDEC demonstrates superior performance over other tools in scATAC-seq analysis across multiple datasets and experimental settings. In downstream applications, we demonstrated that the generative power of scDEC helps to infer the trajectory and intermediate state of cells during differentiation and the latent features learned by scDEC can potentially reveal both biological cell types and within-cell-type variations. We also showed that it is possible to extend scDEC for the integrative analysis of multi-modal single cell data.

PMID:34179690 | PMC:PMC8223760 | DOI:10.1038/s42256-021-00333-y