Cross-validation Sample Research Articles

Targeting Item–level Nuances Leads to Small but Robust Improvements in Personality Prediction from Digital Footprints

In the past decade, researchers have demonstrated that personality can be accurately predicted from digital footprint data, including Facebook likes, tweets, blog posts, pictures, and transaction records. Such computer–based predictions from digital footprints can complement—and in some circumstances even replace—traditional self–report measures, which suffer from well–known response biases and are difficult to scale. However, these previous studies have focused on the prediction of aggregate trait scores (i.e. a person's extroversion score), which may obscure prediction–relevant information at theoretical levels of the personality hierarchy beneath the Big 5 traits. Specifically, new research has demonstrated that personality may be better represented by so–called personality nuances—item–level representations of personality—and that utilizing these nuances can improve predictive performance. The present work examines the hypothesis that personality predictions from digital footprint data can be improved by first predicting personality nuances and subsequently aggregating to scores, rather than predicting trait scores outright. To examine this hypothesis, we employed least absolute shrinkage and selection operator regression and random forest models to predict both items and traits using out–of–sample cross–validation. In nine out of 10 cases across the two modelling approaches, nuance–based models improved the prediction of personality over the trait–based approaches to a small, but meaningful degree (4.25% or 1.69% on average, depending on method). Implications for personality prediction and personality nuances are discussed. © 2020 European Association of Personality Psychology

A Bifactor and Item Response Theory Analysis of the Eating Disorder Inventory-3

The Eating Disorder Inventory-3 (EDI-3; Garner, 2004) is a 91-item, self-report measure scored on 12 scales (three Eating Disorder Risk scales, nine Psychological scales) and six composites. A sample of 1206 female eating disorder patients was divided randomly into calibration (n = 607) and cross-validation (n = 599) samples for confirmatory factor analyses. A bifactor model best fit the data in both samples, but a model with second-order factors corresponding to the risk and psychological scales approached the fit of the bifactor model. Rasch analysis identified only two items whose level of misfit showed a lack of coherence with other scale items (the only items referring to drug and alcohol use), there were no items with reversed or “disordered” response categories, and only five items had sub-threshold estimated discrimination values. Overall, the results were supportive of the EDI-3’s psychometric properties and consistent with interpretive guidelines presented in the test manual.

Important Considerations when Using Models

Important Considerations when Using Models

A Method To Estimate Free-living Behaviors Using High-frequency Wrist Accelerometer Data

PURPOSE: To develop a novel method to estimate free-living sedentary behavior and activity intensity using high-frequency wrist accelerometer data. METHODS: Forty-nine participants (mean ± SD; age: 20.4±1.3 yrs, 45.8% male) completed four, 1-hour sessions of free-living behaviors in home, school, community, and exercise environments. Each session was video-recorded (DO) and participants wore an ActiGraph wGT3X-BT (AG) accelerometer on the non-dominant wrist. Videos were coded for whole body movement, contextual activity type, and activity intensity from the Compendium of Physical Activities (e.g. walking, shopping, 2.8 METs). The novel two-step method (SojWrist) first segments AG data into bouts, or “sojourns”, of inactivity (i.e. sedentary and standing behaviors) or activity using an acceleration standard deviation threshold and random forest model. The second step estimates the intensity of inactive (sedentary, light) and active (light, moderate, vigorous) sojourns. Separate inactive and active sojourn RF models were fit to estimate intensity using bout duration and time- and frequency-domain AG signal characteristics. A 90-10 sample split was used for SojWrist development (N = 44) and cross-validation (N=5). Percent agreement between DO and SojWrist was evaluated at each step using second-by-second data. RESULTS: In the cross-validation sample, 91.8% [95%CI: 87.0%, 96.5%] of inactive and active periods were classified correctly from step 1 of SojWrist. After step 2, overall percent agreement between DO and SojWrist was 86.9% [95%CI: 78.9%, 95.0%] across all intensity categories (Table). CONCLUSION: The new SojWrist performed well at estimating free-living activity intensity categories from a wrist worn accelerometer. Future work should strive to improve method performance for predicting activity intensity categories and test validity on a diverse, independent, free-living sample. Supported by NIH NIDDK 1R01DK110148

Development Of A Dxa-derived Body Volume Equation In Hispanic Adults For Administering In 4-compartment Models

PURPOSE: The purpose of the current study was two-fold: 1) to develop a new dual energy X-ray absorptiometry (DXA)-derived body volume (BV) equation with the GE-Lunar Prodigy while utilizing underwater weighing (UWW) as a criterion and 2) cross-validate 4-compartment(4C) models when utilizing the new DXA-derived BV equation (4C-DXANICKERSON), Wilson DXA-derived BV equation (4C-DXAWILSON), and air displacement plethysmography (ADP)-derived BV (4C-ADP) in Hispanic adults. METHODS: 191 Hispanic adults (18-45yrs) participated in the current study. The development sample consisted of 60 females and 60 males whereas the cross-validation sample comprised of 41 females and 30 males. Criterion body fat percentage (BF%) and fat-free mass (FFM) were determined using a 4C model with UWW as a criterion for BV (4C-UWW). The new DXA-derived BV equation (Nickerson) was developed by linearly regressing UWW-derived BV with DXA fat mass (FM), lean mass (LM), and bone mineral content (BMC). 4C-DXANICKERSON, 4C-DXAWILSON, and 4C-ADP were compared against 4C-UWW in the cross-validation sample. RESULTS: The new DXA-derived BV equation (L) was generated in the development sample as follows: (FM/0.91) + (LM/1.06) + (BMC/16.95) + 0.268. 4C-DXANICKERSON, 4C-DXAWILSON, and 4C-ADP all produced similar mean values (BF%=21.04±5.99, 22.23±6.93, and 20.62±6.26%, respectively) when compared to 4C-UWW (21.29±6.14%) in Hispanic males (all p>0.05). 4C-DXANICKERSON also yielded similar BF% and FFM values as 4C-UWW when evaluating the constant error (CE) in Hispanic females (CE=-0.79% and 0.38kg; p=0.060 and 0.174, respectively). However, 4C-DXAWILSON produced significantly different BF% and FFM values (CE=3.22% and -2.20kg, respectively; both p<0.001). Additionally, 4C-DXAWILSON yielded significant proportional bias when estimating BF% (coefficient=0.226; p<0.001) whereas 4C-ADP produced significant proportional bias for BF% and FFM (coefficient=0.188 and -0.084; both p<0.05) when evaluated in Hispanic females. CONCLUSIONS: Current study findings demonstrate that 4C-DXANICKERSON is a valid measure of BV in Hispanics and is recommended for use in clinics where DXA is the main body composition assessment technique.

Improving Prediction Accuracy Using Multi-allelic Haplotype Prediction and Training Population Optimization in Wheat.

The use of haplotypes may improve the accuracy of genomic prediction over single SNPs because haplotypes can better capture linkage disequilibrium and genomic similarity in different lines and may capture local high-order allelic interactions. Additionally, prediction accuracy could be improved by portraying population structure in the calibration set. A set of 383 advanced lines and cultivars that represent the diversity of the University of Minnesota wheat breeding program was phenotyped for yield, test weight, and protein content and genotyped using the Illumina 90K SNP Assay. Population structure was confirmed using single SNPs. Haplotype blocks of 5, 10, 15, and 20 adjacent markers were constructed for all chromosomes. A multi-allelic haplotype prediction algorithm was implemented and compared with single SNPs using both k-fold cross validation and stratified sampling optimization. After confirming population structure, the stratified sampling improved the predictive ability compared with k-fold cross validation for yield and protein content, but reduced the predictive ability for test weight. In all cases, haplotype predictions outperformed single SNPs. Haplotypes of 15 adjacent markers showed the best improvement in accuracy for all traits; however, this was more pronounced in yield and protein content. The combined use of haplotypes of 15 adjacent markers and training population optimization significantly improved the predictive ability for yield and protein content by 14.3 (four percentage points) and 16.8% (seven percentage points), respectively, compared with using single SNPs and k-fold cross validation. These results emphasize the effectiveness of using haplotypes in genomic selection to increase genetic gain in self-fertilized crops.

Development and Evaluation of a Prediction Model for Ascertaining Rheumatic Heart Disease Status in Administrative Data.

BackgroundPrevious research has raised substantial concerns regarding the validity of the International Statistical Classification of Diseases and Related Health Problems (ICD) codes (ICD-10 I05–I09) for rheumatic heart disease (RHD) due to likely misclassification of non-rheumatic valvular disease (non-rheumatic VHD) as RHD. There is currently no validated, quantitative approach for reliable case ascertainment of RHD in administrative hospital data.MethodsA comprehensive dataset of validated Australian RHD cases was compiled and linked to inpatient hospital records with an RHD ICD code (2000–2018, n=7555). A prediction model was developed based on a generalized linear mixed model structure considering an extensive range of demographic and clinical variables. It was validated internally using randomly selected cross-validation samples and externally. Conditional optimal probability cutpoints were calculated, maximising discrimination separately for high-risk versus low-risk populations.ResultsThe proposed model reduced the false-positive rate (FPR) from acute rheumatic fever (ARF) cases misclassified as RHD from 0.59 to 0.27; similarly for non-rheumatic VHD from 0.77 to 0.22. Overall, the model achieved strong discriminant capacity (AUC: 0.93) and maintained a similar robust performance during external validation (AUC: 0.88). It can also be used when only basic demographic and diagnosis data are available.ConclusionThis paper is the first to show that not only misclassification of non-rheumatic VHD but also of ARF as RHD yields substantial FPRs. Both sources of bias can be successfully addressed with the proposed model which provides an effective solution for reliable RHD case ascertainment from hospital data for epidemiological disease monitoring and policy evaluation.

Integration of GF2 Optical, GF3 SAR, and UAV Data for Estimating Aboveground Biomass of China’s Largest Artificially Planted Mangroves

Accurate methods to estimate the aboveground biomass (AGB) of mangroves are required to monitor the subtle changes over time and assess their carbon sequestration. The AGB of forests is a function of canopy-related information (canopy density, vegetation status), structures, and tree heights. However, few studies have attended to integrating these factors to build models of the AGB of mangrove plantations. The objective of this study was to develop an accurate and robust biomass estimation of mangrove plantations using Chinese satellite optical, SAR, and Unmanned Aerial Vehicle (UAV) data based digital surface models (DSM). This paper chose Qi’ao Island, which forms the largest contiguous area of mangrove plantation in China, as the study area. Several field visits collected 127 AGB samples. The models for AGB estimation were developed using the random forest algorithm and integrating images from multiple sources: optical images from Gaofen-2 (GF-2), synthetic aperture radar (SAR) images from Gaofen-3 (GF-3), and UAV-based digital surface model (DSM) data. The performance of the models was assessed using the root-mean-square error (RMSE) and relative RMSE (RMSEr), based on five-fold cross-validation and stratified random sampling approach. The results showed that images from the GF-2 optical (RMSE = 33.49 t/ha, RMSEr = 21.55%) or GF-3 SAR (RMSE = 35.32 t/ha, RMSEr = 22.72%) can be used appropriately to monitor the AGB of the mangrove plantation. The AGB models derived from a combination of the GF-2 and GF-3 datasets yielded a higher accuracy (RMSE = 29.89 t/ha, RMSEr = 19.23%) than models that used only one of them. The model that used both datasets showed a reduction of 2.32% and 3.49% in RMSEr over the GF-2 and GF-3 models, respectively. On the DSM dataset, the proposed model yielded the highest accuracy of AGB (RMSE = 25.69 t/ha, RMSEr = 16.53%). The DSM data were identified as the most important variable, due to mitigating the saturation effect observed in the optical and SAR images for a dense AGB estimation of the mangroves. The resulting map, derived from the most accurate model, was consistent with the results of field investigations and the mangrove plantation sequences. Our results indicated that the AGB can be accurately measured by integrating images from the optical, SAR, and DSM datasets to adequately represent canopy-related information, forest structures, and tree heights.

Application of a Spectroscopic Analysis-Based Portable Sensor for Phosphate Quantitation in Hydroponic Solutions

Hydroponic plant culturing requires the concentration of nutrients in the supplied solution to be maintained at a constant level. For example, phosphate ion concentration directly affects crop growth, which necessitates the development of convenient and rapid techniques for on-site phosphate quantitation. Herein, we developed a new low-cost colorimetric method of quick on-site phosphate quantitation based on a modification of the conventional molybdenum colorimetric method. Specifically, the nutrient solution treated with ascorbic acid and molybdate was analyzed by colorimetric method after 10 min incubation, and a phosphate quantitation protocol was proposed. To verify this protocol, 50 nutrient solution samples with concentrations of 0–200 ppm were used to develop a model and perform a validation experiment, and the PLSR (Partial Least Squares Regression) and PCR (Principal Component Regression) models were developed and validated using a crossvalidation method and sample transmission spectra. The PLSR model, employing smoothing preprocessing at a 5 nm wavelength spacing, exhibited the best prediction performance and showed an error of ~10% within the measurement range during verification. In addition, an artificial neural network-based model achieved R2=0.93 for the training set and R2=0.86 for the validation set. Finally, we developed convenient-to-use software for phosphate ion quantitation by the presented method and performed a demonstration test.

T157. THE COURSE OF SCHIZOPHRENIA-RELATED NEURAL FINGERPRINTS OVER NINE YEARS - A LONGITUDINAL POPULATION-BASED MACHINE LEARNING STUDY

BackgroundPrevious machine learning studies using structural MRI (sMRI) have been able to separate schizophrenia from controls with relatively high (about 80%) sensitivity and specificity (Kambeitz et al. Neuropsychopharmacology 2015). Interestingly, prediction accuracy in first-episode psychosis is lower compared to older and probably more chronic patients. One possibility is that the appearance of the neurodiagnostic fingerprints (NF) originated from the schizophrenia vs. controls classifier become more visible over time in schizophrenia due to the progressive nature of the disorder.MethodsUsing the Cobre sample (70 schizophrenia and 74 controls), we trained support vector machine (SVM) to differentiate schizophrenia from controls using sMRI. Next, we utilized the Northern Finland Birth Cohort 1966 (NFBC 1966) sample of 29 schizophrenia and 61 non-psychotic controls who participated in the nine-year follow-up. We applied the Cobre-trained SVM models at the baseline (participants 34 years old) and the follow-up (participants 43 years old) using out of sample cross-validation without any in-between retraining. Two independent schizophrenia datasets (the Neuromorphometry by Computer Algorithm Chicago [NMorphCH] and the Consortium for Neuropsychiatric Phenomics [CNP]) were utilized for replication analyses of the SVM generalizability. To address the possibility that the NF mainly capture some general psychopathology, we tested whether the NF generalize to depression using two independent MDD samples from Munich and Münster, Germany.ResultsUsing the Cobre-trained SVM models for schizophrenia vs. controls differentiation in the NFBC 1966, we found balanced accuracy (i.e. mean of sensitivity and specificity, [BAC]) of 72.8% (sensitivity=58.6%, specificity=86.9%) at the baseline and BAC of 79.7% (sensitivity=75.9%, specificity=83.6%) at the follow-up. In the NFBC 1966 schizophrenia patients, we found that SVM decision scores varied as a function of timepoint into the direction of more schizophrenia-likeness at the follow-up (paired T-test, Cohen’s d=0.58, P=0.004). The same was not true in controls (Cohen’s d=0.09, P=0.49). The SVM decision score difference*timepoint interaction related to the decrease of hippocampus and medial prefrontal cortex. The SVM models’ performance was also validated at the two replication samples (BAC of 77.5% in the CNP and BAC of 69.1% in the NMorphCH). In the NFBC 1966 the strongest clinical variable correlating with the trajectory of SVM decision scores over the follow-up was poor performance in the California Verbal Learning Test. This finding was also replicated in the CNP dataset. Further, in the NFBC 1966, those schizophrenia patients with a low degree of SVM decision scores had a higher probability of being in remission, being able to work, and being without antipsychotic medication at the follow-up. The generalization of the SVM models to MDD was worse compared to schizophrenia classification (DeLong’s tests for the two ROC curves: P<0.001).DiscussionThe degree of schizophrenia-related neurodiagnostic fingerprints appear to magnify over time in schizophrenia. By contrast, the discernibility of these fingerprints in controls does not change over time. This indicates that the NF captures some schizophrenia-related progressive neural changes, and not, e.g., normal aging-related brain volume loss. The fingerprints were also generalizable to other schizophrenia samples. Further, the fingerprints seem to have some disorder specificity as the SVM models do not generalize to depression. Lastly, it appears that a low degree of schizophrenia-related NF in schizophrenia might possess some value in predicting patients’ future remission and recovery-related factors.

Calibration and Cross-Validation of Accelerometery for Estimating Movement Skills in Children Aged 8-12 Years.

(1) Background: This study sought to calibrate triaxial accelerometery, worn on both wrists, waist and both ankles, during children’s physical activity (PA), with particular attention to object control motor skills performed at a fast and slow cadence, and to cross-validate the accelerometer cut-points derived from the calibration using an independent dataset. (2) Methods: Twenty boys (10.1 ±1.5 years) undertook seven, five-minute bouts of activity lying supine, standing, running (4.5kmph−1) instep passing a football (fast and slow cadence), dribbling a football (fast and slow cadence), whilst wearing five GENEActiv accelerometers on their non-dominant and dominant wrists and ankles and waist. VO2 was assessed concurrently using indirect calorimetry. ROC curve analysis was used to generate cut-points representing sedentary, light and moderate PA. The cut-points were then cross-validated using independent data from 30 children (9.4 ± 1.4 years), who had undertaken similar activities whilst wearing accelerometers and being assessed for VO2. (3) Results: GENEActiv monitors were able to discriminate sedentary activity to an excellent level irrespective of wear location. For moderate PA, discrimination of activity was considered good for monitors placed on the dominant wrist, waist, non-dominant and dominant ankles but fair for the non-dominant wrist. Applying the cut-points to the cross-validation sample indicated that cut-points validated in the calibration were able to successfully discriminate sedentary behaviour and moderate PA to an excellent standard and light PA to a fair standard. (4) Conclusions: Cut-points derived from this calibration demonstrate an excellent ability to discriminate children’s sedentary behaviour and moderate intensity PA comprising motor skill activity.

Development of a Body Mass Index-based Body Fat Equation: Effect of Handgrip Strength.

The purpose of the current study was twofold: 1) to develop a new BMI-based body fat equation that includes the measurement of RHG (BMINICKERSON) and 2) to cross-validate BMINICKERSON, BMIWOMERSLEY, BMIJACKSON, BMIDEURENBERG, and BMIGALLAGHER against a four-compartment criterion. The development and cross-validation samples consisted of 230 and 110 participants, respectively. Criterion body fat percent was determined with a four-compartment model. RHG was calculated by summing the max of each handgrip strength measurement and dividing by body mass. BMI (kg·m), RHG (kg·kg), age (yr), ethnicity (Hispanic or non-Hispanic White), and sex (male or female) were entered into a stepwise regression to calculate BMINICKERSON. BMINICKERSON was calculated as follows: body fat percent = 21.504 - (12.484 × RHG) - (7.998 × sex) + (0.722 × BMI). In the cross-validation sample, BMINICKERSON produced lower constant error (CE) and total error (TE) values (CE = -0.11%, TE = 4.28%) than all other BMI-based body fat equations (CE = 0.89%-1.90%, TE = 5.71%-6.87%). Furthermore, the 95% limits of agreement were lower for BMINICKERSON ± 8.47% than previous BMI-based body fat equations (95% limits of agreement = ±11.14% to 13.33%). Current study results confirm that previous BMI-based body fat equations produce large error in Hispanics and non-Hispanic Whites but can be improved by accounting for RHG. Allied health professionals are encouraged to use BMINICKERSON in clinical and field settings for adiposity assessments.

Sexually Harassing Behaviors from Patients or Clients and Care Workers' Mental Health: Development and Validation of a Measure.

Although evidence reveals severe effects of sexual harassment on care workers’ mental health, there is a scarcity of studies that investigate care workers’ experiences of sexually inappropriate behavior from patients or clients. One reason for this lack of research is that validated measures that assess different types of sexual harassment experienced by employees working with patients or clients are lacking. In this study, we seek to establish a conceptual framework for investigating extraorganizational sexual harassment in healthcare work. Based on this theoretical framework, we developed and validated a measure for assessing sexually harassing behaviors from patients or clients. Data were gathered from heterogeneous samples of employees working in a variety of settings in healthcare. To evaluate the factorial structure of the measure, we conducted exploratory factor analysis (EFA) using a calibration sample (N = 179) and confirmatory factor analysis (CFA) using a cross-validation sample (N = 305). The construct validity of the measure was demonstrated by investigating relationships with indicators of care workers’ mental health. EFA revealed three factors, namely, nonverbal, verbal, and physical acts of sexual harassment. Examination of the measure comprising 14 items revealed acceptable internal consistencies and substantial correlations with indicators of care workers’ mental health. This study provides a useful and sound measure for assessing sexual harassment from patients or clients and paves the way for the development of a comprehensive theoretical framework for the assessment of sexual harassment. Furthermore, it facilitates future investigations of risk factors for sexual harassment and protective factors helping healthcare workers cope with sexual harassment from patients or clients.

Accurate Genomic Predictions for Chronic Wasting Disease in U.S. White-Tailed Deer.

The geographic expansion of chronic wasting disease (CWD) in U.S. white-tailed deer (Odocoileus virginianus) has been largely unabated by best management practices, diagnostic surveillance, and depopulation of positive herds. Using a custom Affymetrix Axiom single nucleotide polymorphism (SNP) array, we demonstrate that both differential susceptibility to CWD, and natural variation in disease progression, are moderately to highly heritable ( among farmed U.S. white-tailed deer, and that loci other than PRNP are involved. Genome-wide association analyses using 123,987 quality filtered SNPs for a geographically diverse cohort of 807 farmed U.S. white-tailed deer (n = 284 CWD positive; n = 523 CWD non-detect) confirmed the prion gene (PRNP; G96S) as a large-effect risk locus (P-value < 6.3E-11), as evidenced by the estimated proportion of phenotypic variance explained (PVE ≥ 0.05), but also demonstrated that more phenotypic variance was collectively explained by loci other than PRNP. Genomic best linear unbiased prediction (GBLUP; n = 123,987 SNPs) with k-fold cross validation (k = 3; k = 5) and random sampling (n = 50 iterations) for the same cohort of 807 farmed U.S. white-tailed deer produced mean genomic prediction accuracies ≥ 0.81; thereby providing the necessary foundation for exploring a genomically-estimated CWD eradication program.

Outlier Guided Optimization of Abdominal Segmentation.

Abdominal multi-organ segmentation of computed tomography (CT) images has been the subject of extensive research interest. It presents a substantial challenge in medical image processing, as the shape and distribution of abdominal organs can vary greatly among the population and within an individual over time. While continuous integration of novel datasets into the training set provides potential for better segmentation performance, collection of data at scale is not only costly, but also impractical in some contexts. Moreover, it remains unclear what marginal value additional data have to offer. Herein, we propose a single-pass active learning method through human quality assurance (QA). We built on a pre-trained 3D U-Net model for abdominal multi-organ segmentation and augmented the dataset either with outlier data (e.g., exemplars for which the baseline algorithm failed) or inliers (e.g., exemplars for which the baseline algorithm worked). The new models were trained using the augmented datasets with 5-fold cross-validation (for outlier data) and withheld outlier samples (for inlier data). Manual labeling of outliers increased Dice scores with outliers by 0.130, compared to an increase of 0.067 with inliers (p<0.001, two-tailed paired t-test). By adding 5 to 37 inliers or outliers to training, we find that the marginal value of adding outliers is higher than that of adding inliers. In summary, improvement on single-organ performance was obtained without diminishing multi-organ performance or significantly increasing training time. Hence, identification and correction of baseline failure cases present an effective and efficient method of selecting training data to improve algorithm performance.

A Generalizable Multivariate Brain Pattern for Interpersonal Guilt

Feeling guilty when we have wronged another is a crucial aspect of prosociality, but its neurobiological bases are elusive. Although multivariate patterns of brain activity show promise for developing brain measures linked to specific emotions, it is less clear whether brain activity can be trained to detect more complex social emotional states such as guilt. Here, we identified a distributed guilt-related brain signature (GRBS) across two independent neuroimaging datasets that used interpersonal interactions to evoke guilt. This signature discriminated conditions associated with interpersonal guilt from closely matched control conditions in a cross-validated training sample (N = 24; Chinese population) and in an independent test sample (N = 19; Swiss population). However, it did not respond to observed or experienced pain, or recalled guilt. Moreover, the GRBS only exhibited weak spatial similarity with other brain signatures of social-affective processes, further indicating the specificity of the brain state it represents. These findings provide a step toward developing biological markers of social emotions, which could serve as important tools to investigate guilt-related brain processes in both healthy and clinical populations.

Development of a dual-energy X-ray absorptiometry-derived body volume equation in Hispanic adults for administering a four-compartment model.

The purpose of the present study was: (1) to develop a new dual-energy X-ray absorptiometry (DXA)-derived body volume (BV) equation with the GE-Lunar prodigy while utilising underwater weighing (UWW) as a criterion and (2) to cross-validate the novel DXA-derived BV equation (4C-DXANickerson), Wilson DXA-derived BV equation (4C-DXAWilson) and air displacement plethysmography (ADP)-derived BV (4C-ADP) in Hispanic adults. A total of 191 Hispanic adults (18-45 years) participated in the present study. The development sample consisted of 120 females and males (50 % females), whereas the cross-validation sample comprised of forty-one females and thirty males (n 71). Criterion body fat percentage (BF %) and fat-free mass (FFM) were determined using a four-compartment (4C) model with UWW as a criterion for BV (4C-UWW). 4C-DXANickerson, 4C-DXAWilson and 4C-ADP were compared against 4C-UWW in the cross-validation sample. 4C-DXANickerson, 4C-DXAWilson and 4C-ADP all produced similar validity statistics when compared with 4C-UWW in Hispanic males (all P > 0·05). 4C-DXANickerson also yielded similar BF % and FFM values as 4C-UWW when evaluating the mean differences (constant error (CE)) in Hispanic females (CE = -0·79 % and 0·38 kg; P = 0·060 and 0·174, respectively). However, 4C-DXAWilson produced significantly different BF % and FFM values (CE = 3·22 % and -2·20 kg, respectively; both P < 0·001). Additionally, 4C-DXAWilson yielded significant proportional bias when estimating BF % (P < 0·001), whereas 4C-ADP produced significant proportional bias for BF % and FFM (both P < 0·05) when evaluated in Hispanic females. The present study findings demonstrate that 4C-DXANickerson is a valid measure of BV in Hispanics and is recommended for use in clinics, where DXA is the main body composition assessment technique.

Does Consumer‟s Retail Format Choice Decisions Distinguish among 4Cs of Marketing? Evidence from a Research on Jeans wear Retailing

PURPOSE: To understand consumer’s perceptions on 4 Cs of marketing (Customer solution, Customer Cost, Convenience, and Communication) and their impact on retail format choice decisions in buying jeans. This study found evidence on rarely explored research question i.e. ‘Which 4 Cs of Marketing differs among customer’s retail format choice decisions in buying jeans. DESIGN / METHODOLOGY / APPROACH: The data was collected through structured questionnaire by random mall intercept method using simple random sampling technique from 505 customers in Hyderabad, Andhra Pradesh. A valid Discriminant model was developed to test the predictors of 4 Cs of retail attributes across two retail formats (Departmental Stores &amp;Shopping malls). The data was analyzed through descriptive and inferential statistics. FINDINGS: Findings indicated that the classification accuracy in analysis, holdout and cross-validated sample is able to correctly classify the 4 Cs of marketing across the two retail formats.Out of the 4Cs of retail marketing attributes studied, ‘Convenience and Cost’ related attributes, differ significantly across the two retail formats (Departmental Store and Shopping malls), while for ‘Communication and Customer solution’ no significant impact was found on customer’s retail format choice decisions in buying jeans. RESEARCH IMPLICATIONS: Findings from this study,aid retailers, marketers and manufacturers in understanding customer paradigm shifts in buying patterns across emerging retail formats and direct them in altering 4 Cs of marketing to serve customers better, hence increasing the overall marketing efficiency. LIMITATIONS: Despite efforts taken to represent all sections, it is appropriate to conduct such studies geographical segment-wise to be more accurate, secondly retailers’ perceptions were not considered and thirdly unorganized retail was not taken in this context. ORIGINALITY / VALUE –The researcher could not come across any studies on retail marketing attributes in specific category like jeans wear retail, especially in Indian context and hence this study would add more value in understanding customer’s buying behavior across emerging retail formats.

Evaluation of NIR and Raman spectroscopies for the quality analytical control of a solid pharmaceutical formulation with three active ingredients.

The manufacture of pharmaceutical products is one of the most regulated industrial sectors due to the importance of its products on human health. These regulations require a high number of physical and chemical determinations. These last are commonly performed by using chromatographic techniques coupled to different detectors, even though the use of spectroscopic techniques is growing.This study was aimed to evaluate the performance of a NIR and Raman spectroscopy methodologies for the quantitative determination of three active ingredients (APIs) with different concentrations (6.8%w/w, 26.3%w/w and 26.3%w/w) in a solid pharmaceutical product. Calibration models and prediction results were compared in terms of precision and prediction accuracy with those obtained through HPLC reference method. Different spectral pretreatments have been used to reduce the spectral variability associated with the physical characteristics of samples. Standard Normal Variate (SNV) and 2nd derivative (2D) were used as spectral treatments to correct scattering effect in NIR spectra. The best pretreatments for Raman were a baseline correction and a normalization. Each model was developed and evaluated with full subsets cross-validation and external validation sample set. For each technique, the best model was selected based on conventional criteria (R2, RMSEPCV).Results show highly predictive models based on NIR and Raman spectroscopy with root mean square error of prediction values between 0.26%w/w and 2.60%w/w depending on the active ingredient and the analytical technique. The methodologies have been applied for an adequate determination of the uniformity of the dosing units of two commercial lots. In addition, Raman spectroscopy application allowed the detection of an inappropriate distribution of active ingredients in pharmaceutical products.

Scoring algorithms for a computer-based cognitive screening tool: An illustrative example of overfitting machine learning approaches and the impact on estimates of classification accuracy.

Computerized cognitive screening tools, such as the self-administered Computerized Assessment of Memory Cognitive Impairment (CAMCI), require little training and ensure standardized administration and could be an ideal test for primary care settings. We conducted a secondary analysis of a data set including 887 older adults (M age = 72.7 years, SD = 7.1 years; 32.1% male; M years education = 13.4, SD = 2.7 years) with CAMCI scores and independent diagnoses of mild cognitive impairment (MCI). A study by the CAMCI developers used a portion of this data set with a machine learning decision tree model and suggested that the CAMCI had high classification accuracy for MCI (sensitivity = 0.86, specificity = 0.94). We found similar support for accuracy (sensitivity = 0.94, specificity = 0.94) by overfitting a decision tree model, but we found evidence of lower accuracy in a cross-validation sample (sensitivity = 0.62, specificity = 0.66). A logistic regression model, however, discriminated modestly in both training (sensitivity = 0.72, specificity = 0.80) and cross-validation data sets (sensitivity = 0.69, specificity = 0.74). Evidence for strong accuracy when overfitting a decision tree model and substantially reduced accuracy in cross-validation samples was replicated across 500 bootstrapped samples. In contrast, the evidence for accuracy of the logistic regression model was similar in the training and cross-validation samples. The logistic regression model produced accuracy estimates consistent with other published CAMCI studies, suggesting evidence for classification accuracy of the CAMCI for MCI is likely modest. This case study illustrates the general need for cross-validation and careful evaluation of the generalizability of machine learning models. (PsycINFO Database Record (c) 2019 APA, all rights reserved).