Tag: statistics

PLoS Comput Biol. 2023 May 5;19(5):e1011116. doi: 10.1371/journal.pcbi.1011116. Online ahead of print.

ABSTRACT

Our duration estimation flexibly adapts to the statistical properties of the temporal context. Humans and non-human species exhibit a perceptual bias towards the mean of durations previously observed as well as serial dependence, a perceptual bias towards the duration of recently processed events. Here we asked whether those two phenomena arise from a unitary mechanism or reflect the operation of two distinct systems that adapt separately to the global and local statistics of the environment. We employed a set of duration reproduction tasks in which the target duration was sampled from distributions with different variances and means. The central tendency and serial dependence biases were jointly modulated by the range and the variance of the prior, and these effects were well-captured by a unitary mechanism model in which temporal expectancies are updated after each trial based on perceptual observations. Alternative models that assume separate mechanisms for global and local contextual effects failed to capture the empirical results.

PMID:37146089 | DOI:10.1371/journal.pcbi.1011116

PLoS Comput Biol. 2023 May 5;19(5):e1011110. doi: 10.1371/journal.pcbi.1011110. Online ahead of print.

ABSTRACT

Convolutional neural networks (CNNs) can provide powerful and flexible models of neural sensory processing. However, the utility of CNNs in studying the auditory system has been limited by their requirement for large datasets and the complex response properties of single auditory neurons. To address these limitations, we developed a population encoding model: a CNN that simultaneously predicts activity of several hundred neurons recorded during presentation of a large set of natural sounds. This approach defines a shared spectro-temporal space and pools statistical power across neurons. Population models of varying architecture performed consistently and substantially better than traditional linear-nonlinear models on data from primary and non-primary auditory cortex. Moreover, population models were highly generalizable. The output layer of a model pre-trained on one population of neurons could be fit to data from novel single units, achieving performance equivalent to that of neurons in the original fit data. This ability to generalize suggests that population encoding models capture a complete representational space across neurons in an auditory cortical field.

PMID:37146065 | DOI:10.1371/journal.pcbi.1011110

Cardiovasc J Afr. 2023 May 5;34:1-7. doi: 10.5830/CVJA-2023-011. Online ahead of print.

ABSTRACT

OBJECTIVE: Inflammatory mechanisms play an important role in the pathogenesis of atherosclerosis and myocardial infarction. The clinical and prognostic importance of inflammatory parameters, such as neutrophil-lymphocyte (NLR) and platelet-lymphocyte ratios (PLR) in complete blood counts in acute myocardial infarction and other cardiovascular diseases has been demonstrated. However, systemic immune-inflammation index (SII) calculated from neutrophils, lymphocytes and platelets in the complete blood cell count has not been studied sufficiently and is thought to provide a better prediction. This study investigated whether haematological parameters such as SII, NLR and PLR were associated with clinical outcomes in acute coronary syndrome (ACS) patients.

METHODS: We included 1 103 patients who underwent coronary angiography for ACS between January 2017 and December 2021. The association between major adverse cardiac events (MACE) that developed in hospital and at 50 months of follow up and SII, NLR and PLR was compared. Long-term MACE were defined as mortality, re-infarction and target-vessel revascularisation. SII was calculated using the formula: NLR × total platelet count in the peripheral blood (per mm³).

RESULTS: Of the 1 103 patients, 403 were diagnosed with ST-elevation myocardial infarction and 700 with non-STelevation myocardial infarction. The patients were divided into a MACE and a non-MACE group. In hospital and during the 50-month follow up, 195 MACE were observed. SII, PLR and NLR were found to be statistically significantly higher in the MACE group (p < 0.001). SII, C-reactive protein level, age and white blood cell count were independent predictors of MACE in ACS patients.

CONCLUSIONS: SII was found to be a strong independent predictor of poor outcomes in ACS patients. This predictive power was greater than that of PLR and NLR.

PMID:37145864 | DOI:10.5830/CVJA-2023-011

Comput Inform Nurs. 2023 May 1;41(5):300-308. doi: 10.1097/CIN.0000000000000920.

ABSTRACT

Although simulation laboratories are widely used in today’s nursing education programs, it is becoming more and more difficult to find adequate physical space, equipment, and instructors for laboratory practices in schools. With increasing access to quality technology, schools prefer Web-based education or virtual games as another way of learning through simulation laboratories. This study was carried out to evaluate the effect of digital game-supported teaching activity given to nursing students for the developmental care of infants in neonatal ICU on learning. This study is a quasi-experimental study with a control group. Within the scope of the study, the researchers, together with the technical team, developed a digital game in accordance with the purpose of the study. The study was conducted in a health sciences faculty, nursing department, between September 2019 and March 2020. Sixty-two students were included in the study, which were divided into two groups, with 31 students in the experimental group and 31 students in the control group. The study data were collected using a personal information tool and a developmental care information tool. Whereas the digital game learning method was used for the students in the experimental group, the traditional teaching method was used for the students in the control group. No significant difference has been found between the pretest knowledge scores of the students in the experimental and control groups (P > .05). A statistically significant difference has been found in the rates of giving correct answers between the groups in the posttest and retention test (P < .05). The students in the experimental group gave more correct answers in the posttest and retention test than the students in the control group. In line with these results, the learning method with digital games is effective in increasing the knowledge level of nursing undergraduate students. Therefore, it is recommended to integrate digital games into education.

PMID:37145853 | DOI:10.1097/CIN.0000000000000920

Eur J Dent Educ. 2023 May 5. doi: 10.1111/eje.12916. Online ahead of print.

ABSTRACT

INTRODUCTION: The development of early self-assessment skills and determining its correlation with academic performance could provide evidence to alter dental curricula. The aim of this retrospective study was to examine the relationships between students’ early self-assessment ability in waxing and three different evaluation methods (waxing assessment, written examinations, and tooth identification examination) in a dental anatomy course.

MATERIALS AND METHODS: The dental anatomy scores during the academic years of 2018-2019 and 2019-2020 from two cohorts of second-year pre-doctoral dental students at the Harvard School of Dental Medicine were analyzed. Regression analyses were completed to assess the relationship between all evaluation methods.

RESULTS: There was a statistically significant correlation between self-assessment ability and the waxing assessment but no significant correlation between self-assessment ability and the other evaluation methods.

CONCLUSION: Our results showed that the introduction of self-assessments in dental anatomy waxing was correlated with successful waxing skills. Furthermore, a relevant finding is that students who received higher classifications were also capable of doing better self-assessments. These findings provide evidence that impacts dental curricula.

PMID:37145843 | DOI:10.1111/eje.12916

Urol Pract. 2022 Nov;9(6):589-595. doi: 10.1097/UPJ.0000000000000345. Epub 2022 Nov 1.

ABSTRACT

INTRODUCTION: Patient and clinical factors are the most commonly identified variables associated with hospital readmission after radical cystectomy, but other factors may be important drivers of outcomes, such as hospital and physician characteristics. This study investigates the contribution of patient, physician, and hospital factors in hospital readmission after radical cystectomy.

METHODS: This was a retrospective review of the Surveillance, Epidemiology, and End Results-Medicare database focusing on bladder cancer patients who underwent radical cystectomy between 2007 and 2016. Medicare claims were identified using International Statistical Classification of Diseases-9/-10 or Healthcare Common Procedure Coding System codes from Medicare Provider Analysis and Review or National Claims History claims, from which the annual hospital/physician volumes were calculated and classified as low, medium, and high. A multivariable analysis was done for 90-day readmission as the outcome using a multilevel model to explore the association between readmission and characteristics of patient, hospital, and physician. Models with random intercepts were constructed to consider the variation from hospital and physician.

RESULTS: Of 3,530 patients, 1,291 (36.6%) were readmitted within 90 days of the index surgery. On multilevel multivariable analysis, factors significantly associated with readmission included continent urinary diversion (OR 1.55, 95% CI 1.21, 2.00), greater National Cancer Institute comorbidity index (2<4 vs 0-<2, OR 1.35, 95% CI 1.05, 1.75; 4+ vs 0-<2, OR 1.76, 95% CI 1.20, 2.58), American Joint Committee on Cancer stage (P = .04), and hospital region (P = .05). Neither hospital volume, physician volume, teaching hospital status, nor National Cancer Institute center designation was associated with hospital readmission. The main source of variation was determined to be the patient factors (95.89%), followed by the physician (1.43%), and then hospital (2.68%) factors.

CONCLUSIONS: Patient-specific factors are the most important in impacting the odds of readmission after radical cystectomy, while hospital and physician factors contribute minimally to this outcome.

PMID:37145810 | DOI:10.1097/UPJ.0000000000000345

Urol Pract. 2022 Nov;9(6):581-586. doi: 10.1097/UPJ.0000000000000336. Epub 2022 Nov 1.

ABSTRACT

INTRODUCTION: While urological complaints increase in aging populations and conditions commonly require management by multiple physician specialty types, exposure to formal urological education in United States medical schools is limited and has been decreasing over time. We aim to update the current status of urological education in the United States curriculum and delve further into the subject matter being taught and the type and timing of this education.

METHODS: An 11-question survey was developed to describe the current status of urological education. The survey was distributed using Survey Monkey to the American Urological Association’s medical student listserv in November 2021. Descriptive statistics were used to summarize survey findings.

RESULTS: Of 879 invitations sent, 173 responded (20%). Most (112/173, 65%) of respondents were in their fourth year. Only 4 (2%) reported that their school had a required clinical urology rotation. Kidney stones (98%) and urinary tract infections (100%) were the most frequent topics taught. The least exposure included infertility (20%), urological emergencies (19%), bladder drainage (17%), and erectile dysfunction (13%). Videos and case vignettes were the preferred learning modalities and the majority (84%) of respondents were familiar with the American Urological Association’s medical student curriculum material.

CONCLUSIONS: The majority of United States medical schools do not have a required clinical urology rotation and some core urological topics are not taught at all. Future incorporation of urological educational material through video and case vignette learning may be the best opportunity to provide exposure to clinical topics that will commonly be encountered regardless of chosen medical discipline.

PMID:37145808 | DOI:10.1097/UPJ.0000000000000336

Urol Pract. 2022 Nov;9(6):561-566. doi: 10.1097/UPJ.0000000000000339. Epub 2022 Nov 1.

ABSTRACT

INTRODUCTION: Clinical research can be expensive and time consuming due to high associated costs and/or duration of the study. We hypothesized that urine sample collection using online recruitment and engagement of research participants via social medial has the potential to reach a large population in a small timeframe, at a reasonable cost.

METHODS: We performed a retrospective cost analysis of a cohort study comparing cost per sample and time per sample for both online and clinically recruited participants for urine sample collection. During this time, cost data were collected based on study associated costs from invoices and budget spreadsheets. The data were subsequently analyzed using descriptive statistics.

RESULTS: Each sample collection kit contained 3 urine cups, 1 for the disease sample and 2 for control samples. Out of the 3,576 (1,192 disease + 2,384 control) total sample cups mailed, 1,254 (695 control) samples were returned. Comparatively, the 2 clinical sites collected 305 samples. Although the initial startup cost of online recruitment was higher, cost per sample for online recruited was found to be $81.45 compared to $398.14 for clinic sample.

CONCLUSIONS: We conducted a nationwide, contactless, urine sample collection through online recruitment in the midst of the COVID-19 pandemic. Results were compared with the samples collected in the clinical setting. Online recruitment can be utilized to collect urine samples rapidly, efficiently, and at a cost per sample that was 20% of an in-person clinic, and without risk of COVID-19 exposure.

PMID:37145804 | DOI:10.1097/UPJ.0000000000000339

ASAIO J. 2023 May 5. doi: 10.1097/MAT.0000000000001950. Online ahead of print.

ABSTRACT

Refractory right ventricular failure has significant morbidity and mortality. Extracorporeal membrane oxygenation is indicated when medical interventions are deemed ineffective. However, it is still being determined if one configuration is better. We conducted a retrospective analysis of our institutional experience comparing the peripheral veno-pulmonary artery (V-PA) configuration versus the dual-lumen cannula with the tip in the pulmonary artery (C-PA). The analysis of a cohort of 24 patients (12 patients in each group). There was no difference in survival after hospital discharge (58.3% in the C-PA group compared to 41.7% in the V-PA group, p = 0.4). Among the C-PA group, there was a statistically significant shorter ICU length of stay (23.5 days [interquartile range {IQR} = 19-38.5] vs. 43 days [IQR = 30-50], p = 0.043) and duration of mechanical ventilation (7.5 days [IQR = 4.5-9.5] compared to (16.5 days [IQR = 9.5-22.5], p = 0.006) in the V-PA group. In the C-PA group, there were lower incidents of bleeding (33.33% vs. 83.33%, p =0.036) and combined ischemic events (0 vs. 41.67%, p = 0.037). In our single-center experience, the C-PA configuration might have a better outcome than the V-PA one. Further studies are needed to confirm our findings.

PMID:37145800 | DOI:10.1097/MAT.0000000000001950

Urol Pract. 2022 Sep;9(5):490. doi: 10.1097/UPJ.0000000000000311.02. Epub 2022 Jun 1.

NO ABSTRACT

PMID:37145766 | DOI:10.1097/UPJ.0000000000000311.02