Positive Predictive Power Research Articles

Auditory stimulation and deep learning predict awakening from coma after cardiac arrest.

Assessing the integrity of neural functions in coma after cardiac arrest remains an open challenge. Prognostication of coma outcome relies mainly on visual expert scoring of physiological signals, which is prone to subjectivity and leaves a considerable number of patients in a 'grey zone', with uncertain prognosis. Quantitative analysis of EEG responses to auditory stimuli can provide a window into neural functions in coma and information about patients' chances of awakening. However, responses to standardized auditory stimulation are far from being used in a clinical routine due to heterogeneous and cumbersome protocols. Here, we hypothesize that convolutional neural networks can assist in extracting interpretable patterns of EEG responses to auditory stimuli during the first day of coma that are predictive of patients' chances of awakening and survival at 3 months. We used convolutional neural networks (CNNs) to model single-trial EEG responses to auditory stimuli in the first day of coma, under standardized sedation and targeted temperature management, in a multicentre and multiprotocol patient cohort and predict outcome at 3 months. The use of CNNs resulted in a positive predictive power for predicting awakening of 0.83 ± 0.04 and 0.81 ± 0.06 and an area under the curve in predicting outcome of 0.69 ± 0.05 and 0.70 ± 0.05, for patients undergoing therapeutic hypothermia and normothermia, respectively. These results also persisted in a subset of patients that were in a clinical 'grey zone'. The network's confidence in predicting outcome was based on interpretable features: it strongly correlated to the neural synchrony and complexity of EEG responses and was modulated by independent clinical evaluations, such as the EEG reactivity, background burst-suppression or motor responses. Our results highlight the strong potential of interpretable deep learning algorithms in combination with auditory stimulation to improve prognostication of coma outcome.

중년기 여성의 심리적 위기감과 심리적 웰빙, 속박감의 구조적 관계 : 여가경험은 심리적 완충효과가 있는가?

In order for humans to enjoy a healthy lifestyle, physical and mental health must be in harmony. Middle-aged women face conflict and difficulties. In this study, it is meaningful to confirm the relationship between psychological crisis, psychological well-being, and bondage and It was necessary to attempt a study on whether leisure experience would have a positive effect. 243 copies of data were collected for middle-aged women aged 40 to 60 and A structural equation model (SEM) was performed using IBM SPSS Amos 20.0 and χ² difference verification was performed to verify the moderating effect. First, the psychological sense of crisis in middle-aged women showed negative (-) significant predictive power in psychological well-being. Second, it was found that the psychological sense of crisis showed positive (+) significant predictive power in the sense of bondage. Third, it was found that leisure experience had a moderating effect in the relationship between psychological crisis and psychological well-being. Fourth, it was found that leisure experience had a positive buffering effect in the relationship between psychological crisis and bondage.

Exploring an early numeracy screening measure for English learners in primary grades

We investigated the technical characteristics of a brief early numeracy screening battery for both English learners (ELs) and English proficient students (EPs). Results indicated there were differences in performance of ELs and EPs. Further, we found reasonable overall accuracy of the screener predicting student outcomes. Similar overall accuracy results were found for ELs and EPs, as well as for predicting academic performance. We discuss study results related to sensitivity, specificity, and negative and positive predictive power as they relate to implications for practice, including screening for risk for ELs and the challenge of false positives in screening systems. We conclude by proposing future avenues of research.

Clinical efficacy of a method for identifying patients at high risk of carotid atherosclerosis in rheumatoid arthritis

Objective: to test the patented “Method for determining a high risk of atherosclerosis in rheumatoid arthritis” on patients with rheumatoid arthritis.Materials and methods: 74 patients with RA who underwent inpatient treatment were examined. Gender composition of respondents: 77% women and 23% men. Mean age 54.1±9.1 years. The duration of RA is from 1 year to 26 years. 85.1% were diagnosed with a high degree of disease activity. Erosive arthritis was detected radiographically in 41 patients (55.4%). 83.7% (n=62) were seropositive for RF, 81% for ACCP (n=60). All received basic anti-inflammatory therapy. Systemic glucocorticoid therapy was received by 33.7%. All patients were calculated cardiovascular risk according to SCORЕ and mSCORЕ Also, all patients underwent duplex scanning of the BCA to detect atherosclerosis.Results: no significant increase in traditional factors of cardiovascular diseases was found among patients. When CVR was stratified according to SCORE, the frequency of low risk was 32.4%, moderate — 59.5%, high — 5.4%, very high — 2.7%. When using mSCORE, low risk was found in 27%, moderate — 56.7%, high — 13.5%, very high — 2.7%. Atherosclerotic plaques (ASP) in the carotid arteries in patients stratified according to SCORE for moderate, high and very high risk was detected in 68%, in persons with similar risks according to mSCORE — in 66.7%. With a score of 4 or more when using the tested method, ASP was found in 77.8%. Sensitivity for the detection of ASP in the carotid arteries using SCORE was 0.77, with mSCORE and the tested method 0.81 and 0.95, respectively. Specificity 0.47, 0.4 and 0.6 respectively. Positive predictive power for SCORE — 0.68, mSCORE — 0.67, tested method — 0.77. Negative predictive power for SCORE — 0.58, mSCORE — 0.6, tested method — 0.9.Conclusion: the study showed that our proposed method, which uses traditional clinical markers, significantly increases the sensitivity, specificity, and negative predictive power of calculation methods for identifying a group of individuals with a high risk of atherosclerotic lesions of extracranial arteries against the background of rheumatoid arthritis compared with using the SCORE and mSCORE parameters.

Diffusion-weighted MRI to determine response and long-term clinical outcomes in muscle-invasive bladder cancer following neoadjuvant chemotherapy

This study aims to determine local treatment response and long-term survival outcomes in patients with localised muscle-invasive bladder cancer (MIBC) patients receiving neoadjuvant chemotherapy (NAC) using diffusion-weighted MRI (DWI) and apparent diffusion coefficient (ADC) analysis. Patients with T2-T4aN0-3M0 bladder cancer suitable for NAC were recruited prospectively. DWI was performed prior to NAC and was repeated following NAC completion. Conventional response assessment was performed with cystoscopy and tumour site biopsy. Response was dichotomised into response (<T2) or poor response (≥T2). Patients proceeded to either radical cystectomy or chemo-radiotherapy as standard of care. Tumour ADC values were calculated for all b-values (ADCall) and high b-values (ADCb100). Mean ADC, percentiles, skew, kurtosis, and their change (ΔADC and %ΔADC) were determined. Threshold predictive of response with highest specificity was ascertained using receiver operating characteristic (ROC) analysis. Median overall survival (OS), bladder-cancer-specific survival (bCSS), progression-free survival (PFS), and time to cystectomy were estimated using Kaplan-Meier method. Significant area under the curve (AUC) cut points were used to determine relationship with long-term endpoints and were compared using log-rank test. Forty-eight patients (96 DWI) were evaluated. NAC response was associated with significant increase in mean ΔADC and %ΔADC compared to poor response (ΔADCall 0.32×10-3 versus 0.11×10-3 mm2/s; p=0.009, and %ΔADCall 21.70% versus 8.23%; p=0.013). Highest specificity predicting response was seen at 75th percentile ADC (AUC, 0.8; p=0.01). Sensitivity, specificity, positive predictive power, and negative predictive power of %ΔADCb100 75th percentile was 73.7%, 90.0%, 96.6%, and 52.9%, respectively. %ΔADCb100 75th percentile >15.5% was associated with significant improvement in OS (HR, 0.40; 95% CI, 0.19-0.86; p=0.0179), bCSS (HR, 0.26; 95% CI, 0.08-0.82; p=0.0214), PFS (HR, 0.16; 95% CI, 0.05-0.48; p=0.0012), and time to cystectomy (HR, 0.19; 95% CI, 0.07-0.47; p=0.0004). Quantitative ADC analysis can successfully identify NAC response and improved long-term clinical outcomes. Multi-centre validation to assess reproducibility and repeatability is required before testing within clinical trials to inform MIBC treatment decision making. We successfully demonstrated that measured change in DWI can successfully identify NAC response and improved long-term survival outcomes.

Modeling English teachers’ behavioral intention to use artificial intelligence in middle schools

Artificial intelligence (AI) provides new opportunities for K-12 English as foreign language (EFL) teachers to improve their teaching. To address the emerging trend of integrating AI into teaching, this study investigated EFL teachers’ perceptions, knowledge, and behavioral intention to use AI to support teaching and learning of English in middle schools. This study combined relevant aspects of the Unified Theory of Acceptance and Use of Technology (UTAUT) and Technological Pedagogical and Content Knowledge (TPACK) as the theoretical basis. A survey was conducted in an AI education demonstration district in China. This survey adopted a 5-point Likert scale which was developed from previous research and the interview records of EFL teachers. A total of 470 valid responses were collected. The reliability and validity of the scale were satisfied with eight constructs: Performance Expectancy (PE), Effort Expectancy (EE), Social Influence (SI), Facilitating Conditions (FC), AI language technological knowledge (AIL-TK), AI technological pedagogical knowledge (AI-TPK), AI-TPACK, and Behavioral Intention (BI). The results showed that the EFL teachers were positive with regard to the measured factors. PE, SI, AIL-TK, and AI-TPACK had significant positive predictive power on BI; and EE, FC, AI-TPK had indirect effects on BI. The complex interrelations were mapped out to provide educators and policymakers with a theoretically grounded scheme to foster teachers’ BI to use AI in teaching.

Predicting efficacy of viloxazine extended-release treatment in adults with ADHD using an early change in ADHD symptoms: Machine learning Post Hoc analysis of a phase 3 clinical trial

Early response to viloxazine extended-release (viloxazine ER, Qelbree®) treatment predicted efficacy outcome in pediatric subjects with attention-deficit/hyperactivity disorder (ADHD). This study sought to determine whether the machine learning lasso model used in the pediatric study would predict response to viloxazine ER in an adult population based on early improvements in ADHD symptoms. We used data from a double-blind, placebo-controlled, flexible-dose (200–600 mg) study of viloxazine ER (N = 354; 18 to 60 years old). Area under the Receiver Operating Characteristic Curve (ROC AUC) statistics were computed using the lasso model from pediatric data to predict responder status in adults. Response was defined as ≥50% reduction from baseline in the Adult ADHD Investigator Symptoms Rating Scale (AISRS) Total score at Week 6. The adult study sample included 127 viloxazine ER-treated subjects with Week 6 data. Fifty-one subjects (40.2%) were categorized as responders. The ROC curves indicated that data collected up to Week 2 were sufficient to accurately predict treatment response at Week 6 with 68% positive predictive power, 80% sensitivity, and 74% specificity. This analysis demonstrated that the predictive model estimated from the child data generalizes to adults with ADHD, further supporting the consistency of viloxazine ER treatment across age groups.

Effects of nondiscretionary trading on futures prices

AbstractThis paper examines the effects of the nondiscretionary trading demands of volatility index (VIX) exchange‐traded products (ETPs) issuers on the prices and volumes in the VIX futures. We find that the ETPs' informationless, mechanical rebalancing of futures positions to maintain the constant maturity of the index and the promised leverage ratios of the VIX ETPs have significantly positive predictive power for end‐of‐day futures returns. We also show that the impact on price has diminished through time from increased liquidity provided by hedge funds, and the “natural” hedging of the issuers' inverse products.

A-311 WMS-IV Visual Reproduction Recognition as a Performance Validity Test (PVT) in Clinical and Forensic Settings

Abstract Objective: This study examined WMS-IV Visual Reproduction Recognition (VRRec) as an embedded PVT in clinical and forensic samples. Methods: VRRec was analyzed as a PVT in a forensic sample and in a hospital-based, outpatient clinical sample. Neither sample included any major neurocognitive disorders and well-established PVTs determined valid and invalid categories. T-tests (parametric and nonparametric) and ROCs were used to assess how accurately VRRec classified valid and invalid group-membership. Results: Forensic and clinical (n=265, 84) samples differed in context (incentivized, non-incentivized), age (m=49.1, 60.6), race (Caucasian=17%, 89%), and gender (males=100%, 29%), respectively. VRRec scores were significantly different between valid and invalid groups in both the forensic sample (U=2404.0, p&amp;lt;.0001, d=-1.26; n=196, 69; m=5.57, 3.33; sd=1.29, 1.91) and the clinical sample (t=-2.39, p=.019, d=-.93; n=76, 8; m=4.99, 3.50; sd=1.68, 1.51). A cutoff score of &amp;lt; 3 yielded the following: Forensic: AUC=.822, sensitivity=.551, specificity=.944, 95%CI=.76-.88, p&amp;lt;.0001; and Clinical: AUC=.752, sensitivity=.500, specificity=.816, 95%CI=.61-.89, p=.019. For the clinical sample, &amp;lt; 2 improved specificity (.908) but decreased sensitivity (.250). When using a 15% estimated invalidity base rate in the clinical sample for &amp;lt; 3 and &amp;lt; 2 cutoffs, positive predictive power equaled .324 and .324 and negative predictive power equaled .902 and .873, respectively. Conclusion: VRRec adequately differentiated validity groups in a forensic sample but was less valuable in a clinical sample. Its use as a PVT should only be considered within a larger group of multiple PVTs. Primary limitations included differences in demographics and specific criterion-PVTs used between samples. Future studies should assess VRRec as a PVT in similar populations.

The Sensitivity and Specificity of WHO clinical Staging in Predicting Cd4 Cell Counts in HIV Infected Patients at the Police Hospital in Accra, Ghana

Introduction : Initiating an HIV positive client on ART is an important step in the care of PLHIV. Due to a shortage of CD4 machines and reagents, the WHO clinical staging guideline is widely used to determine when to initiate ART in developing countries.Objective : To determine whether WHO clinical staging is predictive of CD4 of 350 in initiation of ART in patients presenting at the Police Hospital ART clinic.Methods : We reviewed the records of clients on ART, from 2010 to 2012 and compared the WHO clinical staging and CD4 counts at their first visit to the clinic. Pregnant women, hepatitis B infected patients and children below fifteen were excluded from the study.Results : Two hundred and five PLHIVs were eligible for inclusion in the study. Sensitivity of the WHO Clinical staging to predict initiation of ART was 44% and specificity 87%. Positive predictive power was 83% and Negative predictive power was 51%.Conclusion : Almost 50% of those needing ART as per CD4 count were classified as WHO stage 1 or 2. This means that about half of those deemed not qualified to start ART by WHO clinical staging actually needed to be initiated on ART. Access to CD4 machines andreagents must be increased to minimize delay in start of treatment for patients.

Historical Earnings, Accrual Accounting, and Future Cash Flows: A Malaysian Perspective

This study investigates the influence of historical earnings and historical accrual accounting on projecting future cash flows. The sample consisted of 159 construction, energy, and property development companies from the main capital market of Malaysia. The compiled data of these companies spanned from 2015 to 2019 and was collected from the financial statements of the companies, including operating cash flows and earnings. The profit or loss statement and cash flow statement were used for calculating historic accrual accounting. Earnings, cash flow, and accrual accounting models were used as the basis for the regression model construction. The accounting data demonstrated that the prediction performance of the models was improved by the three-year lag. The previous two-year earnings and accrual accounting had positive and significant predictive power for forecasting future cash flows. Moreover, the past one- and two-year cash flows from operations significantly predicted future cash flows. These results are important for a sound understanding among academics and practitioners of the crux of historical earnings, operating cash flow, and accrual accounting. The findings can assist corporate leaders and management executives in tracking the sustainability and financial growth of an organization.

3D Ultrasound-Guided Photoacoustic Imaging to Monitor the Effects of Suboptimal Tyrosine Kinase Inhibitor Therapy in Pancreatic Tumors.

Pancreatic cancer is a disease with an incredibly poor survival rate. As only about 20% of patients are eligible for surgical resection, neoadjuvant treatments that can relieve symptoms and shrink tumors for surgical resection become critical. Many forms of treatments rely on increased vulnerability of cancerous cells, but tumors or regions within the tumors that may be hypoxic could be drug resistant. Particularly for neoadjuvant therapies such as the tyrosine kinase inhibitors utilized to shrink tumors, it is critical to monitor changes in vascular function and hypoxia to predict treatment efficacy. Current clinical imaging modalities used to obtain structural and functional information regarding hypoxia or oxygen saturation (StO2) do not provide sufficient depth penetration or require the use of exogenous contrast agents. Recently, ultrasound-guided photoacoustic imaging (US-PAI) has garnered significant popularity, as it can noninvasively provide multiparametric information on tumor vasculature and function without the need for contrast agents. Here, we built upon existing literature on US-PAI and demonstrate the importance of changes in StO2 values to predict treatment response, particularly tumor growth rate, when the outcomes are suboptimal. Specifically, we image xenograft mouse models of pancreatic adenocarcinoma treated with suboptimal doses of a tyrosine kinase inhibitor cabozantinib. We utilize the US-PAI data to develop a multivariate regression model that demonstrates that a therapy-induced reduction in tumor growth rate can be predicted with 100% positive predictive power and a moderate (58.33%) negative predictive power when a combination of pretreatment tumor volume and changes in StO2 values pretreatment and immediately posttreatment was employed. Overall, our study indicates that US-PAI has the potential to provide label-free surrogate imaging biomarkers that can predict tumor growth rate in suboptimal therapy.

Serum Levels of Glial Fibrillary Acidic Protein Association with Cognitive Impairment and Type 2 Diabetes

Peripheral biomarkers associated with neurocognitive disorders (NCD) have been evaluated in an attempt to improve diagnosis and early detection and potentially even prevent them. Along with increasing age, type 2 diabetes (T2D) increases the risk of central nervous system disorders and cognitive impairment due to the loss of synaptic function. Central damage triggers an astroglial response, increasing the expression of glial fibrillary acidic protein (GFAP), which can be found peripherally when the blood-brain barrier is compromised. To evaluate the value of GFAP as a peripheral biomarker of central dysfunction. Serum levels of GFAP were compared between cases of NCD (n=69) and age-matched controls (n=69), analyzing the influence of diabetes as contributing factor. We found higher levels of serum GFAP in subjects with NCD compared with the control group (p <0.0001). The receiver operating characteristic (ROC) curve using the GFAP levels showed 65.22% sensitivity and 71.01% specificity (AUC=0.7608), indicating good performance in the classification of controls and NCD patients. Logistic regression indicated a positive predictive power of 67.50% considering T2D status; adding GFAP levels, the predictive power rises to 71.93%. GFAP levels and T2D could be considered good predictors of NCD risk. Our findings open the possibility that peripheral GFAP could be used as an objective measurement related, under certain conditions, to central damage; thereby serving as a follow-up marker to refer diabetic patients for appropriate neurological evaluation, which could offer a low cost, minimally invasive strategy to improve the assessment of cognitive affectation and subsequent treatment.

The role of F-18 FDG PET/CT in the prognosis of patients with hypoxic/anoxic brain injury after cardio-circulatory arrest.

Advances in resuscitation techniques have resulted in more patients surviving cardio-circulatory arrest (CA) and consequently developing hypoxic/anoxic brain damage. The aim of this study is to evaluate the role of PET/CT (Positron Emission Tomography/Computerized Tomography scan) with F-18 FDG (F-18 fluorodeoxyglucose) during the early rehabilitative hospitalization phase in determining the V/C (Vermis/Cerebellar) ratio as a prognostic index to predict patient outcome, as defined by clinical evaluation scales. This is a single-center retrospective study of 37 consecutive adult patients admitted to the neurorehabilitation center between January 2011 and June 2019. Functional status was measured by the following clinical scales: FIM (Functional Indipendence Measure), LCFS (Levels of Cognitive Functioning Scale), GOS (Glasgow Outcome Scale) and CRS-R (Coma Recovery Scale-Revised). PET/CT with F-18 FDG as a functional imaging technique was used to calculate the V/C ratio as a ratio between the metabolism of the vermis and of the Cerebellar Hemisphere. A statistically significant correlation was observed between the V/C ratio and the delta values (difference between discharge and admission value) for each clinical evaluation scale (Delta FIM: P=0.0014; Delta LCFS P=0.0003). A statistically significant difference was observed between the V/C ratio of patients with LCFS ≥4 that showed an improved outcome (defined as an improvement of at least two points in LCFS), and that of patients with LCFS <4 that did not improve (P=0.0011). A V/C ratio cut-off of 1.5 corresponded with a positive predictive power of 80% and a negative predictive power of 82%; a value <1.5 predicted a better outcome. Clinical evaluation scales when associated with F-18 FDG PET/CT measurement of metabolism, provide a more reliable prognosis. This allows for more focused rehabilitation treatment and better management of family members' expectation.

Screening screeners: calculating classification indices using correlations and cut-points.

Given the recent push for universal screening, it is important to take into account how well a screener identifies children at risk for reading problems as well as how screener and sample information contribute to this classification. Picking the best cut-point for a particular sample and screening goal can be challenging given that test manuals often report classification information for a specific cut-point and sample base rate which may not generalize to other samples. By assuming a bivariate normal distribution, it is possible to calculate all of the classification information for a screener based on the correlation between the screener and outcome, the cut-point on the outcome (i.e., the base rate in the sample), and the cut-point on the screener. We provide an example with empirical data to validate these estimation procedures. This information is the basis for a free online tool that provides classification information for a given correlation between screener and outcome and cut-points on each. Results show that the correlation between screener and outcome needs to be greater than .9 (higher than observed in practice) to obtain good classification. These findings are important for researchers, administrators, and practitioners because current screeners do not meet these requirements. Since a correlation is dependent on the reliability of the measures involved, we need screeners with better reliability and/or multiple measures to increase reliability. Additionally, we demonstrate the impact of base rate on positive predictive power and discuss how gated screening can be useful in samples with low base rates.

Beyond screening in primary practice settings: Time to stop fiddling while Rome is burning.

This forum presents the current state of research in the screening and identification of people with eating disorders in community and primary care, taking a longer-term perspective that highlights the slow rate of progression in development of instruments, and impact on polices and practice. An historical overview is presented, followed by a critique of contemporary instruments and practice, and barriers to case detection and appropriate referral pathways. There are now many instruments but all lack high levels of positive predictive power. However, some do have high sensitivity. Barriers contributing to poor detection and the treatment gap include need for improved education and support for primary care professionals and lack of confidence of individuals with eating disorders to initiate a discussion with health professionals. The best screening instrument would not overcome either of these barriers. We purport there is an urgent need to improve current screening instruments (not to develop more), particularly those with high sensitivity. These should be being employed alongside programs to both improve primary care professionals' skills in assessment and management of people with eating disorders, and to empower consumers to navigate care pathways. We argue that further screening instruments for eating disorders are not needed. Rather, it is more urgent to have a greater research focus on how to encourage primary care workers to ask about eating and body image and how to best translate that to more individuals with eating disorders being offered treatment. This work needs to be linked with tools that empower consumers to navigate care pathways.

Decision Making Model for Municipal Wastewater Conventional Secondary Treatment with Bayesian Networks

Technical, economic, regulatory, environmental, and social and political interests make the process of selecting an appropriate wastewater treatment technology complex. Although this problem has already been addressed from the dimensioning approach, our proposal in this research, a model of decision making for conventional secondary treatment of municipal wastewater through continuous-discrete, non-parametric Bayesian networks was developed. The most suitable network was structured in unit processes, independent of each other. Validation, with data in a mostly Mexican context, provided a positive predictive power of 83.5%, an excellent kappa (0.77 &gt; 0.75), and the criterion line was surpassed with the location of the model in a receiver operating characteristic (ROC) graph, so the model can be implemented in this region. The final configuration of the Bayesian network allows the methodology to be easily extended to other types of treatments, wastewater, and to other regions.

Utility of CVLT–3 response bias as a measure of performance validity after traumatic brain injury

Objective We sought to determine the utility of a new performance validity index that was recently proposed. In particular, we wanted to determine if this index would be associated with a specificity of at least .90, a sensitivity of at least .40, and an Area Under the Curve of at least .70 in a traumatic brain injury (TBI) sample. Method We used logistic regression to investigate how well this new index could distinguish persons with TBI (n = 148) who were evaluated within 1–36 months after injury. All participants had been classified on the basis of at least two independent performance validity tests as having provided valid performance (n = 128) or invalid performance (n = 20). Results The new performance validity index had acceptable specificity (.96) but had suboptimal sensitivity (.35) and Area Under the Curve (.66). It was concerning that almost half (5/12) of the cases that were identified by this index as providing invalid effort were false positives. Although a slightly more liberal cut-off improved sensitivity, the problem with poor positive predictive power remained. The conventional Forced Choice index had relatively better classification accuracy. Conclusion Differences in base rates between the original sample of Martin et al. and the current one most likely affected positive predictive power of the new index. Although their performance validity has excellent specificity, the current results do not support the application of this index in the clinical evaluation of patients with traumatic brain injury when base rates of invalid performance differ markedly from those in the original study.

Classification Accuracy of Early Literacy Assessments: Linking Preschool and Kindergarten Performance

A strong foundation in early literacy supports children’s academic pursuits and impacts personal, social, and economic outcomes. Therefore, examining the adequacy of early literacy assessments as predictors of future performance on important outcomes is critical for identifying students at risk of reading problems. This study explored the predictive validity of preschoolers’ literacy skills measured in the spring with the Individual Growth and Development Indicators 2.0 (IGDIs 2.0) to performance in the fall and winter of kindergarten as assessed by the Dynamic Indicators of Basic Early Literacy Skills Next Edition (DIBELS Next) using Pearson product-moment correlations. In addition, the classification accuracy of student performance on the IGDIs 2.0 measures to the publisher-identified benchmark scores on DIBELS Next assessment in kindergarten was examined by calculating the sensitivity, specificity, positive and negative predictive power, overall correct classification, and kappa. Participants were 537 children from ethnically diverse backgrounds enrolled in an urban school district in the U.S. Northeast region. Results indicated small to moderate relationships between the individual IGDIs 2.0 tasks and DIBELS Next measures. Classification accuracy of student performance on the IGDIs 2.0 measures to the publisher-identified benchmark score on DIBELS Next composite in the fall and winter of kindergarten revealed inadequate levels of sensitivity; however, locally derived cut-scores improved sensitivity and specificity.

Prediction of Online Psychological Help-Seeking Behavior During the COVID-19 Pandemic: An Interpretable Machine Learning Method.

Online mental health service (OMHS) has been named as the best psychological assistance measure during the COVID-19 pandemic. An interpretable, accurate, and early prediction for the demand of OMHS is crucial to local governments and organizations which need to allocate and make the decision in mental health resources. The present study aimed to investigate the influence of the COVID-19 pandemic on the online psychological help-seeking (OPHS) behavior in the OMHS, then propose a machine learning model to predict and interpret the OPHS number in advance. The data was crawled from two Chinese OMHS platforms. Linguistic inquiry and word count (LIWC), neural embedding-based topic modeling, and time series analysis were utilized to build time series feature sets with lagging one, three, seven, and 14 days. Correlation analysis was used to examine the impact of COVID-19 on OPHS behaviors across different OMHS platforms. Machine learning algorithms and Shapley additive explanation (SHAP) were used to build the prediction. The result showed that the massive growth of OPHS behavior during the COVID-19 pandemic was a common phenomenon. The predictive model based on random forest (RF) and feature sets containing temporal features of the OPHS number, mental health topics, LIWC, and COVID-19 cases achieved the best performance. Temporal features of the OPHS number showed the biggest positive and negative predictive power. The topic features had incremental effects on performance of the prediction across different lag days and were more suitable for OPHS prediction compared to the LIWC features. The interpretable model showed that the increase in the OPHS behaviors was impacted by the cumulative confirmed cases and cumulative deaths, while it was not sensitive in the new confirmed cases or new deaths. The present study was the first to predict the demand for OMHS using machine learning during the COVID-19 pandemic. This study suggests an interpretable machine learning method that can facilitate quick, early, and interpretable prediction of the OPHS behavior and to support the operational decision-making; it also demonstrated the power of utilizing the OMHS platforms as an always-on data source to obtain a high-resolution timeline and real-time prediction of the psychological response of the online public.