Correction of Temperature Variation with Independent Water Samples to Predict Soluble Solids Content of Kiwifruit Juice Using NIR Spectroscopy

Using the framework of aquaphotomics, we have sought to understand the changes within the water structure of kiwifruit juice occurring with changes in temperature. The study focuses on the first (1300–1600 nm) and second (870–1100 nm) overtone regions of the OH stretch of water and examines temperature differences between 20, 25, and 30 °C. Spectral data were collected using a Fourier transform–near-infrared spectrometer with 1 mm and 10 mm transmission cells for measurements in the first and second overtone region, respectively. Water wavelengths affected by temperature variation were identified. Aquagrams (water spectral patterns) highlight slightly different responses in the first and second overtone regions. The influence of increasing temperature on the peak absorbance of the juice was largely a lateral wavelength shift in the first overtone region and a vertical amplitude shift in the second overtone region of water. With the same data set, we investigated the use of external parameter orthogonalisation (EPO) and extended multiple scatter correction (EMSC) pre-processing to assist in building temperature-independent partial least square regression models for predicting soluble solids concentration (SSC) of kiwifruit juice. The interference component selected for correction was the first principal component loading measured using pure water samples taken at the same three temperatures (20, 25, and 30 °C). The results show that the EMSC method reduced SSC prediction bias from 0.77 to 0.1 °Brix in the first overtone region of water. Using the EPO method significantly reduced the prediction bias from 0.51 to 0.04 °Brix, when applying a model made at one temperature (30 °C) to measurements made at another temperature (20 °C) in the second overtone region of water.

EXPRESS: Understanding and Neutralizing the Expense Prediction Bias: the Role of Accessibility, Typicality, and Skewness

Consumers display an expense prediction bias in which they underpredict their future spending. The authors propose this bias occurs in large part because: 1) consumers base their predictions on typical expenses that come to mind easily during prediction, 2) taken together, typical expenses lead to a prediction near the mode of a consumer’s expense distribution rather than the mean, and 3) expenses display positive skew with mode < mean. Accordingly, the authors also propose that prompting consumers to consider reasons why their expenses might be different than usual increases predictions – and therefore prediction accuracy – by bringing atypical expenses to mind. Ten studies ( N = 6,044) provide support for this account of the bias and the “atypical intervention” developed to neutralize it.

Web-based and mixed-mode cognitive large-scale assessments in higher education: An evaluation of selection bias, measurement bias, and prediction bias

Educational large-scale studies typically adopt highly standardized settings to collect cognitive data on large samples of respondents. Increasing costs alongside dwindling response rates in these studies necessitate exploring alternative assessment strategies such as unsupervised web-based testing. Before respective assessment modes can be implemented on a broad scale, their impact on cognitive measurements needs to be quantified. Therefore, an experimental study on N = 17,473 university students from the German National Educational Panel Study has been conducted. Respondents were randomly assigned to a supervised paper-based, a supervised computerized, and an unsupervised web-based mode to work on a test of scientific literacy. Mode-specific effects on selection bias, measurement bias, and predictive bias were examined. The results showed a higher response rate in web-based testing as compared to the supervised modes, without introducing a pronounced mode-specific selection bias. Analyses of differential test functioning showed systematically larger test scores in paper-based testing, particularly among low to medium ability respondents. Prediction bias for web-based testing was observed for one out of four criteria on study-related success factors. Overall, the results indicate that unsupervised web-based testing is not strictly equivalent to other assessment modes. However, the respective bias introduced by web-based testing was generally small. Thus, unsupervised web-based assessments seem to be a feasible option in cognitive large-scale studies in higher education.

Development of an Easy-to-Use Prediction Equation for Body Fat Percentage Based on BMI in Overweight and Obese Lebanese Adults

An accurate estimation of body fat percentage (BF%) in patients who are overweight or obese is of clinical importance. In this study, we aimed to develop an easy-to-use BF% predictive equation based on body mass index (BMI) suitable for individuals in this population. A simplified prediction equation was developed and evaluated for validity using anthropometric measurements from 375 adults of both genders who were overweight or obese. Measurements were taken in the outpatient clinic of the Department of Nutrition and Dietetics at Beirut Arab University (Lebanon). A total of 238 participants were used for model building (training sample) and another 137 participants were used for evaluating validity (validation sample). The final predicted model included BMI and sex, with non-significant prediction bias in BF% of −0.017 ± 3.86% (p = 0.946, Cohen’s d = 0.004). Moreover, a Pearson’s correlation between measured and predicted BF% was strongly significant (r = 0.84, p < 0.05). We are presenting a model that accurately predicted BF% in 61% of the validation sample with an absolute percent error less than 10% and non-significant prediction bias (−0.028 ± 4.67%). We suggest the following equations: BF% females = 0.624 × BMI + 21.835 and BF% males = 1.050 × BMI − 4.001 for accurate BF% estimation in patients who are overweight or obese in a clinical setting in Lebanon.

Accurate brain age prediction with lightweight deep neural networks

Deep learning has huge potential for accurate disease prediction with neuroimaging data, but the prediction performance is often limited by training-dataset size and computing memory requirements. To address this, we propose a deep convolutional neural network model, Simple Fully Convolutional Network (SFCN), for accurate prediction of brain age using T1-weighted structural MRI data. Compared with other popular deep network architectures, SFCN has fewer parameters, so is more compatible with small dataset size and 3D volume data. The network architecture was combined with several techniques for boosting performance, including data augmentation, pre-training, model regularization, model ensemble and prediction bias correction. We compared our overall SFCN approach with several widely-used machine learning models. It achieved state-of-the-art performance in UK Biobank data (N = 14,503), with mean absolute error (MAE) = 2.14y in brain age prediction and 99.5% in sex classification. SFCN also won (both parts of) the 2019 Predictive Analysis Challenge for brain age prediction, involving 79 competing teams (N = 2,638, MAE = 2.90y). We describe here the details of our approach, and its optimisation and validation. Our approach can easily be generalised to other tasks using different image modalities, and is released on GitHub.HighlightsA lightweight deep learning model, Simple Fully Convolutional Network (SFCN), is presented, achieving state-of-the-art brain age prediction and sex classification performance in UK Biobank MRI brain imaging data.Even with limited number of training subjects (e.g., 50), SFCN performs better than widely-used regression models.A semi-multimodal ensemble strategy is proposed and achieved first place in the PAC 2019 brain age prediction challenge.Linear regression can remove brain age prediction bias (even on unlabelled data) while maintaining state-of-the-art performance.