Category: Statistics

IEEE Trans Vis Comput Graph. 2022 Sep 27;PP. doi: 10.1109/TVCG.2022.3209373. Online ahead of print.

ABSTRACT

In a basketball play, players who are not in possession of the ball (i.e., off-ball players) can still effectively contribute to the team’s offense, such as making a sudden move to create scoring opportunities. Analyzing the movements of off-ball players can thus facilitate the development of effective strategies for coaches. However, common basketball statistics (e.g., points and assists) primarily focus on what happens around the ball and are mostly result-oriented, making it challenging to objectively assess and fully understand the contributions of off-ball movements. To address these challenges, we collaborate closely with domain experts and summarize the multi-level requirements for off-ball movement analysis in basketball. We first establish an assessment model to quantitatively evaluate the offensive contribution of an off-ball movement considering both the position of players and the team cooperation. Based on the model, we design and develop a visual analytics system called OBTracker to support the multifaceted analysis of off-ball movements. OBTracker enables users to identify the frequency and effectiveness of off-ball movement patterns and learn the performance of different off-ball players. A tailored visualization based on the Voronoi diagram is proposed to help users interpret the contribution of off-ball movements from a temporal perspective. We conduct two case studies based on the tracking data from NBA games and demonstrate the effectiveness and usability of OBTracker through expert feedback.

PMID:36166529 | DOI:10.1109/TVCG.2022.3209373

Alzheimers Dement. 2022 Sep 27. doi: 10.1002/alz.12792. Online ahead of print.

ABSTRACT

BACKGROUND: Mild cognitive impairment (MCI) is a heterogeneous condition with high individual variabilities in clinical outcomes driven by patient demographics, genetics, brain structure features, blood biomarkers, and comorbidities. Multi-modality data-driven approaches have been used to discover MCI subtypes; however, disease comorbidities have not been included as a modality though multiple diseases including hypertension are well-known risk factors for Alzheimer’s disease (AD). The aim of this study was to examine MCI heterogeneity in the context of AD-related comorbidities along with other AD-relevant features and biomarkers.

METHODS: A total of 325 MCI subjects with 32 AD-relevant comorbidities and features were considered. Mixed-data clustering is applied to discover and compare MCI subtypes with and without including AD-related comorbidities. Finally, the relevance of each comorbidity-driven subtype was determined by examining their MCI to AD disease prognosis, descriptive statistics, and conversion rates.

RESULTS: We identified four (five) MCI subtypes: poor-, average-, good-, and best-AD prognosis by including comorbidities (without including comorbidities). We demonstrated that comorbidity-driven MCI subtypes differed from those identified without comorbidity information. We further demonstrated the clinical relevance of comorbidity-driven MCI subtypes. Among the four comorbidity-driven MCI subtypes there were substantial differences in the proportions of participants who reverted to normal function, remained stable, or converted to AD. The groups showed different behaviors, having significantly different MCI to AD prognosis, significantly different means for cognitive test-related and plasma features, and by the proportion of comorbidities.

CONCLUSIONS: Our study indicates that AD comorbidities should be considered along with other diverse AD-relevant characteristics to better understand MCI heterogeneity.

PMID:36166485 | DOI:10.1002/alz.12792

PLoS Comput Biol. 2022 Sep 27;18(9):e1010512. doi: 10.1371/journal.pcbi.1010512. Online ahead of print.

ABSTRACT

Insect flight is a strongly nonlinear and actuated dynamical system. As such, strategies for understanding its control have typically relied on either model-based methods or linearizations thereof. Here we develop a framework that combines model predictive control on an established flight dynamics model and deep neural networks (DNN) to create an efficient method for solving the inverse problem of flight control. We turn to natural systems for inspiration since they inherently demonstrate network pruning with the consequence of yielding more efficient networks for a specific set of tasks. This bio-inspired approach allows us to leverage network pruning to optimally sparsify a DNN architecture in order to perform flight tasks with as few neural connections as possible, however, there are limits to sparsification. Specifically, as the number of connections falls below a critical threshold, flight performance drops considerably. We develop sparsification paradigms and explore their limits for control tasks. Monte Carlo simulations also quantify the statistical distribution of network weights during pruning given initial random weights of the DNNs. We demonstrate that on average, the network can be pruned to retain a small amount of original network weights and still perform comparably to its fully-connected counterpart. The relative number of remaining weights, however, is highly dependent on the initial architecture and size of the network. Overall, this work shows that sparsely connected DNNs are capable of predicting the forces required to follow flight trajectories. Additionally, sparsification has sharp performance limits.

PMID:36166481 | DOI:10.1371/journal.pcbi.1010512

PLoS One. 2022 Sep 27;17(9):e0275131. doi: 10.1371/journal.pone.0275131. eCollection 2022.

ABSTRACT

BACKGROUND: Severe acute respiratory syndrome coronavirus-2 is a global health care problem with high mortality. Despite early mortality seeming alarming, data regarding factors that lead to increased early mortality of COVID 19 patients is not well-documented yet. The objective of this study was to identify the risk factors of early mortality in patients with confirmed COVID-19 infections.

METHODOLOGY: A case-control study design was employed. With this, a total of 261 COVID-19 deceased recordings were reviewed. The cases of the study were recordings of patients deceased within three days of intensive care unit admission whereas, the rest 187 were recordings of patients who died after three days of admission. Data were collected using an extraction checklist, entered into Epi data version 4.4.2.2, and analyzed by SPSS version 25. After the description, binary logistic regression was run to conduct bivariate and multivariable analyses. Finally, statistical significance was declared at p-value <0.05, and an adjusted odds ratio with a 95% confidence interval was used to report the strength of association.

RESULT: The analysis was performed on 261 (87 cases and 174 controls) recordings. About 62.5% of the participants were aged above 65 years and two-thirds were males. The presence of cardiovascular disease (AOR = 4.79, with 95%CI: 1.73, 13.27) and bronchial-asthma (AOR = 6.57; 95% CI: 1.39, 31.13) were found to have a statistically significant association with early mortality. The existence of complications from COVID-19 (AOR = 0.22; 95% CI: 0.07, 0.74) and previous history of COVID-19 infection (AOR = 0.17, 95% CI: 0.04, 0.69) were associated with decreased risk of early mortality.

CONCLUSIONS: Having cardiovascular diseases and bronchial asthma was associated with an increased risk of early mortality. Conversely, the presence of intensive care unit complications and previous history of COVID-19 infection were associated with decreased risk of early mortality.

PMID:36166445 | DOI:10.1371/journal.pone.0275131

PLoS One. 2022 Sep 27;17(9):e0275299. doi: 10.1371/journal.pone.0275299. eCollection 2022.

ABSTRACT

OBJECTIVES: To investigate healthcare resource utilization (HRU) and associated costs by depression severity and year of diagnosis among patients with treatment-resistant depression (TRD) in Denmark.

METHODS: Including all adult patients with a first-time hospital contact for major depressive disorder (MDD) in 1996-2015, TRD patients were defined at the second shift in depression treatment (antidepressant medicine or electroconvulsive therapy) and matched 1:2 with non-TRD patients. The risk of utilization and amount of HRU and associated costs including medicine expenses 12 months after the TRD-defining date were reported, comparing TRD patients with non-TRD MDD patients.

RESULTS: Identifying 25,321 TRD-patients matched with 50,638 non-TRD patients, the risk of psychiatric hospitalization following TRD diagnosis was 138.4% (95%-confidence interval: 128.3-149.0) higher for TRD patients than for non-TRD MDD patients. The number of hospital bed days and emergency department (ED) visits were also higher among TRD patients, with no significant difference for somatic HRU. Among patients who incurred healthcare costs, the associated HRU costs for TRD patients were 101.9% (97.5-106.4) higher overall, and 55.2% (50.9-59.6) higher for psychiatric services than those of non-TRD patients. The relative differences in costs for TRD-patients vs non-TRD patients were greater for patients with mild depression and tended to increase over the study period (1996-2015), particularly for acute hospitalizations and ED visits.

LIMITATIONS: TRD was defined by prescription patterns besides ECT treatments.

CONCLUSION: TRD was associated with increased psychiatric-related HRU. Particularly the difference in acute hospitalizations and ED visits between TRD and non-TRD patients increased over the study period.

PMID:36166443 | DOI:10.1371/journal.pone.0275299

J Vet Med Educ. 2022 Sep 26:e20220036. doi: 10.3138/jvme-2022-0036. Online ahead of print.

ABSTRACT

The COVID-19 pandemic has exerted a huge adverse influence on global teaching activities and students’ psychological status. Veterinary microbiology is mainly concerned with bacterial and viral diseases, including coronavirus diseases. An innovative online-to-offline teaching approach for this course was established to stimulate students’ learning initiative and mitigate their anxiety about COVID-19. A well-established massive open online course (MOOC) was first adopted as preview material before class, followed by in-person teaching. Additionally, COVID-19-related scientific papers were also used as pre-class reading material in veterinary microbiology and were further explained in class. The effect of this innovative teaching mode was systematically evaluated by final examination scores and questionnaires. The average score (81.75) and excellence score rating (> 85 scores, 37.3%) resulting from this blended teaching mode were not statistically higher than those of the online-only (79.19, p = .115; 28.6%, p = .317) or offline-only (79.47, p = .151; 27.9%, p = .269) teaching modes. This may be due to the sample size investigated; however, the results indicate that the innovative teaching mode did not decrease teaching quality. Additionally, most subjects (72.9%) were satisfied with the blended mode and supported its future use. Intriguingly, the introduction of COVID-19-related scientific papers helped students understand virology, relieve their anxiety, and increase their professional identity. Collectively, the innovative approach to teaching veterinary microbiology in this study provides a beneficial reference for other teachers to maintain and improve teaching quality.

PMID:36166201 | DOI:10.3138/jvme-2022-0036

JAMA Netw Open. 2022 Sep 1;5(9):e2233441. doi: 10.1001/jamanetworkopen.2022.33441.

ABSTRACT

IMPORTANCE: A genetic diagnosis can help elucidate the prognosis of hearing loss, thus significantly affecting management. Previous studies on diagnostic yield of hearing loss genetic tests have been based on largely homogenous study populations.

OBJECTIVES: To examine the diagnostic yield of genetic testing in a diverse population of children, accounting for sociodemographic and patient characteristics, and assess whether these diagnoses are associated with subsequent changes in clinical management.

DESIGN, SETTING, AND PARTICIPANTS: This retrospective cohort study included 2075 patients seen at the Children’s Communications Clinic, of whom 517 completed hearing loss gene panel testing between January 1, 2015, and November 1, 2021, at the University of California, San Francisco Benioff Children’s Hospital system. From those 517 patients, 426 children with at least 2 audiograms were identified and analyzed. Data were gathered from November 2021 to January 2022 and analyzed from January to February 2022.

MAIN OUTCOMES AND MEASURES: The measures of interest were sociodemographic characteristics (age at testing, gender, race and ethnicity, primary language, and insurance type), hearing loss characteristics, and medical variables. The outcome was genetic testing results. Variables were compared with univariate and multivariable logistic regression.

RESULTS: Of the 2075 patients seen at the Children’s Communications Clinic, 517 (median [range] age, 8 [0-31] years; 264 [51.1%] male; 351 [67.9%] from an underrepresented minority [URM] group) underwent a hearing loss panel genetic test between January 1, 2015, and November 1, 2021. Among those 517 patients, 426 children (median [range] age, 8 [0-18] years; 221 [51.9%] male; 304 [71.4%] from an URM group) with 2 or more audiograms were included in a subsequent analysis. On multivariable logistic regression, age at testing (odds ratio [OR], 0.87; 95% CI, 0.78-0.97), URM group status (OR, 0.29; 95% CI, 0.13-0.66), comorbidities (OR, 0.27; 95% CI, 0.14-0.53), late-identified hearing loss (passed newborn hearing screen; OR, 0.27; 95% CI, 0.08-0.86), and unilateral hearing loss (OR, 0.04; 95% CI, 0.005-0.33) were the only factors associated with genetic diagnosis. No association was found between genetic diagnosis yield and other sociodemographic variables or hearing loss characteristics. Patients in URM and non-URM groups had statistically similar clinical features. A total of 32 of 109 children (29.4%) who received a genetic diagnosis received diagnoses that significantly affected prognosis because of identification of syndromic or progressive sensorineural hearing loss or auditory neuropathy spectrum disorder relating to otoferlin.

CONCLUSIONS AND RELEVANCE: This cohort study’s findings suggest that genetic testing may be broadly useful in improving clinical management of children with hearing loss. More research is warranted to discover and characterize diagnostic genes for those who have been historically underrepresented in research and medicine.

PMID:36166228 | DOI:10.1001/jamanetworkopen.2022.33441

Ir J Med Sci. 2022 Sep 27. doi: 10.1007/s11845-022-03168-7. Online ahead of print.

ABSTRACT

BACKGROUND: Platelet-rich plasma(PRP) and prolotherapy(PRL) are regenerative treatment approaches in the knee osteoarthritis (KOA).

AIM: To see how efficient PRP and PRL are in treating KOA.

METHODS: A total of 108 patients with a diagnosis of KOA who received either PRL, PRP, or exercise therapy and whose 3-month follow-up data were available were included in this retrospective study (PRL n = 35 or PRP n = 35, exercise n = 38). Visual Analogue Scale(VAS) and The Western Ontario McMaster University Osteoarthritis Index(WOMAC) were used as outcome measures at baseline, 1 month, and 3 months.

RESULTS: There were no statistically significant differences between the three groups in terms of demographic parameters, baseline assessments of pain intensity, or WOMAC scores. At the first and third months, all groups showed a substantial improvement in the VAS activity, resting and WOMAC values as compared to before treatment(p < 0.05). When the groups were compared, the VAS activity, resting, and WOMAC values in PRP and PRL improved significantly in the first and third months compared to the exercise group. At one month, there was a statistically significant improvement in VAS activity and WOMAC pain and total scores compared to PRP and PRL, but this improvement was not significant at 3 months.

CONCLUSION: Pain and disability were significantly improved with PRL and PRP compared with exercise therapy. Although PRP is more effective than PRL in the first month after treatment, PRL may be preferred due to its low cost, long-term efficacy, and low complication rates due to the periarticular application.

PMID:36166187 | DOI:10.1007/s11845-022-03168-7

Hormones (Athens). 2022 Sep 27. doi: 10.1007/s42000-022-00398-3. Online ahead of print.

ABSTRACT

PURPOSE: The purpose of this study is to explore the association between adiposity indices and blood lipid indices and prediabetes. We compare the predictive value of new adiposity indices and traditional adiposity indices and blood lipid indices in the diagnosis of prediabetes.

METHODS: This is a prospective cohort study of 7953 participants. The follow-up time was 3 years. The eight adiposity indices included the following: body mass index (BMI), waist circumference (WC), body roundness index (BRI), A Body Shape Index (ABSI), visceral adiposity index (VAI), lipid accumulation product (LAP), fatty liver index (FLI), and triglyceride-to-glucose fasting index (TyG), as well as four blood lipid indices as follows: total cholesterol (TC), triglycerides (TG), high-density lipoprotein (HDL-C), and low-density lipoprotein (LDL-C).The association between adiposity indices and blood lipid indices for diagnosis of prediabetes was estimated using a logistic regression model to obtain the odds ratio (OR) and its 95% confidence interval (CI). We calculated the area under the curve (AUC) of receiver operating characteristic (ROC) curve analysis to measure the predictive value of adiposity indices and blood lipid indicators for the diagnosis of prediabetes in the general population stratified by gender.

RESULTS: The median age of the participants was 56 years old, men accounting for 35.3% of the final group. After adjusting for confounding factors, association of BMI, BRI, VAI, LAP, TyG, TC, TG, and LDL-C with prediabetes status was assessed at both baseline and follow-up. TyG (AUC, overall: 0.677 (95% CI, 0.665, 0.689), male: 0.645 (95% CI, 0.624-0.667), and female: 0.693 (95% CI, 0.678-0.708)) have better diagnostic value for prediabetes than VAI, LAP, FLI, TC, TG, HDL-C, and LDL-C. The predictive value of the combination of TyG, BRI, VAI, and TG significantly improves the power of any single index in the diagnosis of prediabetes. The AUC and corresponding 95% CI of TyG, BRI, VAI, and TG and the combination of these four indicators to diagnose prediabetes were 0.677 (0.665, 0.689), 0.630 (0.617, 0.643), 0.618 (0.606, 0.631), 0.622 (0.609, 0.635), and 0.728 (0.716, 0.739), respectively.

CONCLUSIONS: Among the eight adiposity indices and four blood lipid indices evaluated in the study, TyG had the highest diagnostic value for prediabetes in isolated indexes, and the combination of TyG, BRI, VAI, and TG significantly improved the diagnostic value for prediabetes of any single indicator.

PMID:36166170 | DOI:10.1007/s42000-022-00398-3

Community Ment Health J. 2022 Sep 27. doi: 10.1007/s10597-022-01025-6. Online ahead of print.

ABSTRACT

The COVID-19 pandemic and associated sequelae have disproportionately exacerbated refugee mental health due to health disparities, poverty, and unique risk factors. In response to the pandemic, most mental health providers have shifted to virtual platforms. Given the high need for services in this population, it is essential to understand the effectiveness and potential barriers to serving refugees via telehealth. This study is one of the first to examine the extent that socio-cultural and structural barriers impact telemental health services received by resettled refugees during the COVID-19 pandemic. This study also addresses the potential benefits of telemental health service delivery to refugees. We surveyed 85 providers serving refugee and non-refugee clients in the United States. Statistical analyses revealed that more significant socio-cultural and structural barriers, including access to technology, linguistic challenges, and privacy limitations, exist for refugees compared to non-refugee clients. Potential benefits of telemental health for refugees during the pandemic included fewer cancellations, fewer transportation concerns, and better access to childcare. These results highlight the need to address the disparity in telemental health service delivery to refugees to limit inequities for this population.

PMID:36166148 | DOI:10.1007/s10597-022-01025-6