Accuracy of Genomic prediction for fleece traits in Inner Mongolia Cashmere goats.

The fleece traits are important economic traits of goats. With the reduction of sequencing and genotyping cost and the improvement of related technologies, genomic selection for goats has become possible. The research collect pedigree, phenotype and genotype information of 2299 Inner Mongolia Cashmere goats (IMCGs) individuals. We estimate fixed effects, and compare the estimates of variance components, heritability and genomic predictive ability of fleece traits in IMCGs when using the pedigree based Best Linear Unbiased Prediction (ABLUP), Genomic BLUP (GBLUP) or single-step GBLUP (ssGBLUP). The fleece traits considered are cashmere production (CP), cashmere diameter (CD), cashmere length (CL) and fiber length (FL). It was found that year of production, sex, herd and individual ages had highly significant effects on the four fleece traits (P < 0.01). All of these factors should be considered when the genetic parameters of fleece traits in IMCGs are evaluated. The heritabilities of FL, CL, CP and CD with ABLUP, GBLUP and ssGBLUP methods were 0.26 ~ 0.31, 0.05 ~ 0.08, 0.15 ~ 0.20 and 0.22 ~ 0.28, respectively. Therefore, it can be inferred that the genetic progress of CL is relatively slow. The predictive ability of fleece traits in IMCGs with GBLUP (56.18% to 69.06%) and ssGBLUP methods (66.82% to 73.70%) was significantly higher than that of ABLUP (36.73% to 41.25%). For the ssGBLUP method is significantly (29% ~ 33%) higher than that with ABLUP, and which is slightly (4% ~ 14%) higher than that of GBLUP. The ssGBLUP will be as an superiors method for using genomic selection of fleece traits in Inner Mongolia Cashmere goats.

Enhancing long-term survival prediction with two short-term events: Landmarking with a flexible varying coefficient model.

Patients with cardiovascular diseases who experience disease-related short-term events, such as hospitalizations, often exhibit diverse long-term survival outcomes compared to others. In this study, we aim to improve the prediction of long-term survival probability by incorporating two short-term events using a flexible varying coefficient landmark model. Our objective is to predict the long-term survival among patients who survived up to a pre-specified landmark time since the initial admission. Inverse probability weighting estimation equations are formed based on the information of the short-term outcomes before the landmark time. The kernel smoothing method with the use of cross-validation for bandwidth selection is employed to estimate the time-varying coefficients. The predictive performance of the proposed model is evaluated and compared using predictive measures: area under the receiver operating characteristic curve and Brier score. Simulation studies confirm that parameters under the landmark models can be estimated accurately and the predictive performance of the proposed method consistently outperforms existing methods that either do not incorporate or only partially incorporate information from two short-term events. We demonstrate the practical application of our model using a community-based cohort from the Atherosclerosis Risk in Communities (ARIC) study.

Effect of Posterior Keratometry on the Accuracy of 10 Intraocular Lens Calculation Formulas: Standard Keratometry versus Total Keratometry.

PURPOSE: To investigate the effect of posterior keratometry (PK) on the accuracy of 10 intraocular lens (IOL) power calculation formulas using standard keratometry (K) and total keratometry (TK). METHODS: This is a retrospective consecutive case-series study. The IOL power was calculated using K and TK measured by IOLMaster 700 in 6 new-generation formulas (Barrett Universal II, Emmetropia Verifying Optical (EVO) 2.0, RBF Calculator 3.0, Hoffer QST, Kane, and Ladas Super Formula) and 4 traditional formulas (Haigis, Hoffer Q, Holladay 1, and SRK/T). The arithmetic prediction error (PE) and mean absolute PE (MAE) were evaluated. The locally-weighted scatterplot smoothing was performed to assess the relationship between PE and PK. RESULTS: A total of 576 patients (576 eyes) who underwent cataract surgery were included. Compared with using K, all formulas using TK showed a hyperopic shift in the whole group. Specifically, for eyes with PK exceeding -5.90 D, all formulas using TK exhibited a hyperopic shift (all P < 0.001), while eyes with PK less than -5.90 D showed a myopic shift (all P < 0.001). The MAE of new-generation formulas calculated with TK and K showed no statistical differences, while the MAE of traditional formulas with TK was larger (TK: 0.34 ~ 0.43 D; K: 0.33 ~ 0.42 D, all P < 0.05). CONCLUSIONS: The prediction bias of formulas with TK increased as PK deviated from -5.90 D. TK did not improve the prediction accuracy of new-generation formulas, and even performed worse in traditional formulas.

Residual network improves the prediction accuracy of genomic selection.

Genetic improvement of complex traits in animal and plant breeding depends on the efficient and accurate estimation of breeding values. Deep learning methods have been shown to be not superior over traditional genomic selection (GS) methods, partially due to the degradation problem (i.e. with the increase of the model depth, the performance of the deeper model deteriorates). Since the deep learning method residual network (ResNet) is designed to solve gradient degradation, we examined its performance and factors related to its prediction accuracy in GS. Here we compared the prediction accuracy of conventional genomic best linear unbiased prediction, Bayesian methods (BayesA, BayesB, BayesC, and Bayesian Lasso), and two deep learning methods, convolutional neural network and ResNet, on three datasets (wheat, simulated and real pig data). ResNet outperformed other methods in both Pearson's correlation coefficient (PCC) and mean squared error (MSE) on the wheat and simulated data. For the pig backfat depth trait, ResNet still had the lowest MSE, whereas Bayesian Lasso had the highest PCC. We further clustered the pig data into four groups and, on one separated group, ResNet had the highest prediction accuracy (both PCC and MSE). Transfer learning was adopted and capable of enhancing the performance of both convolutional neural network and ResNet. Taken together, our findings indicate that ResNet could improve GS prediction accuracy, affected potentially by factors such as the genetic architecture of complex traits, data volume, and heterogeneity.

Investigating the Performance of Frequentist and Bayesian Techniques in Genomic Evaluation.

The genomic evaluation process relies on the assumption of linkage disequilibrium between dense single-nucleotide polymorphism (SNP) markers at the genome level and quantitative trait loci (QTL). The present study was conducted with the aim of evaluating four frequentist methods including Ridge Regression, Least Absolute Shrinkage and Selection Operator (LASSO), Elastic Net, and Genomic Best Linear Unbiased Prediction (GBLUP) and five Bayesian methods including Bayes Ridge Regression (BRR), Bayes A, Bayesian LASSO, Bayes C, and Bayes B, in genomic selection using simulation data. The difference between prediction accuracy was assessed in pairs based on statistical significance (p-value) (i.e., t test and Mann-Whitney U test) and practical significance (Cohen's d effect size) For this purpose, the data were simulated based on two scenarios in different marker densities (4000 and 8000, in the whole genome). The simulated data included a genome with four chromosomes, 1 Morgan each, on which 100 randomly distributed QTL and two different densities of evenly distributed SNPs (1000 and 2000), at the heritability level of 0.4, was considered. For the frequentist methods except for GBLUP, the regularization parameter λ was calculated using a five-fold cross-validation approach. For both scenarios, among the frequentist methods, the highest prediction accuracy was observed by Ridge Regression and GBLUP. The lowest and the highest bias were shown by Ridge Regression and GBLUP, respectively. Also, among the Bayesian methods, Bayes B and BRR showed the highest and lowest prediction accuracy, respectively. The lowest bias in both scenarios was registered by Bayesian LASSO and the highest bias in the first and the second scenario were shown by BRR and Bayes B, respectively. Across all the studied methods in both scenarios, the highest and the lowest accuracy were shown by Bayes B and LASSO and Elastic Net, respectively. As expected, the greatest similarity in performance was observed between GBLUP and BRR ( d = 0.007 , in the first scenario and d = 0.003 , in the second scenario). The results obtained from parametric t and non-parametric Mann-Whitney U tests were similar. In the first and second scenario, out of 36 t test between the performance of the studied methods in each scenario, 14 ( P < . 001 ) and 2 ( P < . 05 ) comparisons were significant, respectively, which indicates that with the increase in the number of predictors, the difference in the performance of different methods decreases. This was proven based on the Cohen's d effect size, so that with the increase in the complexity of the model, the effect size was not seen as very large. The regularization parameters in frequentist methods should be optimized by cross-validation approach before using these methods in genomic evaluation.

A novel nonparametric time-dependent precision-recall curve estimator for right-censored survival data.

In order to assess prognostic risk for individuals in precision health research, risk prediction models are increasingly used, in which statistical models are used to estimate the risk of future outcomes based on clinical and nonclinical characteristics. The predictive accuracy of a risk score must be assessed before it can be used in routine clinical decision making, where the receiver operator characteristic curves, precision-recall curves, and their corresponding area under the curves are commonly used metrics to evaluate the discriminatory ability of a continuous risk score. Among these the precision-recall curves have been shown to be more informative when dealing with unbalanced biomarker distribution between classes, which is common in rare event, even though except one, all existing methods are proposed for classic uncensored data. This paper is therefore to propose a novel nonparametric estimation approach for the time-dependent precision-recall curve and its associated area under the curve for right-censored data. A simulation is conducted to show the better finite sample property of the proposed estimator over the existing method and a real-world data from primary biliary cirrhosis trial is used to demonstrate the practical applicability of the proposed estimator.

A Nonlinear Local Approximation Approach for Catchment Classification.

Catchment classification plays an important role in many applications associated with water resources and environment. In recent years, several studies have applied the concepts of nonlinear dynamics and chaos for catchment classification, mainly using dimensionality measures. The present study explores prediction as a measure for catchment classification, through application of a nonlinear local approximation prediction method. The method uses the concept of phase-space reconstruction of a time series to represent the underlying system dynamics and identifies nearest neighbors in the phase space for system evolution and prediction. The prediction accuracy measures, as well as the optimum values of the parameters involved in the method (e.g., phase space or embedding dimension, number of neighbors), are used for classification. For implementation, the method is applied to daily streamflow data from 218 catchments in Australia, and predictions are made for different embedding dimensions and number of neighbors. The prediction results suggest that phase-space reconstruction using streamflow alone can provide good predictions. The results also indicate that better predictions are achieved for lower embedding dimensions and smaller numbers of neighbors, suggesting possible low dimensionality of the streamflow dynamics. The classification results based on prediction accuracy are found to be useful for identification of regions/stations with higher predictability, which has important implications for interpolation or extrapolation of streamflow data.

A divide-and-conquer method for sparse risk prediction and evaluation.

Divide-and-conquer (DAC) is a commonly used strategy to overcome the challenges of extraordinarily large data, by first breaking the dataset into series of data blocks, then combining results from individual data blocks to obtain a final estimation. Various DAC algorithms have been proposed to fit a sparse predictive regression model in the $L_1$ regularization setting. However, many existing DAC algorithms remain computationally intensive when sample size and number of candidate predictors are both large. In addition, no existing DAC procedures provide inference for quantifying the accuracy of risk prediction models. In this article, we propose a screening and one-step linearization infused DAC (SOLID) algorithm to fit sparse logistic regression to massive datasets, by integrating the DAC strategy with a screening step and sequences of linearization. This enables us to maximize the likelihood with only selected covariates and perform penalized estimation via a fast approximation to the likelihood. To assess the accuracy of a predictive regression model, we develop a modified cross-validation (MCV) that utilizes the side products of the SOLID, substantially reducing the computational burden. Compared with existing DAC methods, the MCV procedure is the first to make inference on accuracy. Extensive simulation studies suggest that the proposed SOLID and MCV procedures substantially outperform the existing methods with respect to computational speed and achieve similar statistical efficiency as the full sample-based estimator. We also demonstrate that the proposed inference procedure provides valid interval estimators. We apply the proposed SOLID procedure to develop and validate a classification model for disease diagnosis using narrative clinical notes based on electronic medical record data from Partners HealthCare.

Accuracy or novelty: what can we gain from target-specific machine-learning-based scoring functions in virtual screening?

Machine-learning (ML)-based scoring functions (MLSFs) have gradually emerged as a promising alternative for protein-ligand binding affinity prediction and structure-based virtual screening. However, clouds of doubts have still been raised against the benefits of this novel type of scoring functions (SFs). In this study, to benchmark the performance of target-specific MLSFs on a relatively unbiased dataset, the MLSFs trained from three representative protein-ligand interaction representations were assessed on the LIT-PCBA dataset, and the classical Glide SP SF and three types of ligand-based quantitative structure-activity relationship (QSAR) models were also utilized for comparison. Two major aspects in virtual screening campaigns, including prediction accuracy and hit novelty, were systematically explored. The calculation results illustrate that the tested target-specific MLSFs yielded generally superior performance over the classical Glide SP SF, but they could hardly outperform the 2D fingerprint-based QSAR models. Although substantial improvements could be achieved by integrating multiple types of protein-ligand interaction features, the MLSFs were still not sufficient to exceed MACCS-based QSAR models. In terms of the correlations between the hit ranks or the structures of the top-ranked hits, the MLSFs developed by different featurization strategies would have the ability to identify quite different hits. Nevertheless, it seems that target-specific MLSFs do not have the intrinsic attributes of a traditional SF and may not be a substitute for classical SFs. In contrast, MLSFs can be regarded as a new derivative of ligand-based QSAR models. It is expected that our study may provide valuable guidance for the assessment and further development of target-specific MLSFs.

Lower refractive prediction accuracy of total keratometry using intraocular lens formulas loaded onto a swept-source optical biometer.

PURPOSE: To compare refractive outcomes calculated using intraocular lens (IOL) power calculation formulas loaded onto the IOLMaster 700 with the employment of anterior keratometry (K) and total keratometry (TK). METHODS: A total of 225 eyes of 225 patients underwent uneventful cataract surgery and implantation of a single model of nontoric monofocal IOL by a single surgeon. All eyes underwent preoperative ocular biometric measurements with the IOLMaster 700. Refractive outcomes, including the mean numerical prediction error (MNE); standard deviation (SD); adjusted mean absolute prediction error (MAE); adjusted median absolute prediction error (MedAE); percentages of eyes with adjusted prediction error (PE) within ± 0.25, ± 0.50, ± 0.75, and ± 1.00 diopter; and IOL Formula Performance Index (FPI), were compared between the K-based formula and the TK-based formula of Barrett Universal II (BUII), Haigis, SRK/T, Holladay 2, and Hoffer Q. Axial length (short, medium, and long) subgroup analyses and anterior and posterior keratometry (flat, medium, and steep) subgroup analyses were conducted. RESULTS: The K-based formula performed better than the TK-based formula in the accuracy of refractive prediction of each IOL calculation formula: BUII-K (FPI 0.690), BUII-TK (0.677), Haigis-K (0.617), Haigis-TK (0.584), SRK/T-K (0.608), SRK/T-TK (0.595), Holladay 2-K (0.419), Holladay 2-TK (0.406), Hoffer Q-K (0.364), and Hoffer Q-TK (0.356). The subgroup analyses of refractive prediction outcomes showed that TK influenced the refractive outcomes in eyes with relatively normal ranges of axial length and anterior keratometry. CONCLUSIONS: Using TK instead of K leads to lower refractive prediction accuracy of the IOL power calculation formulas loaded on the IOLMaster 700.

Modification of the Barrett Universal II formula by the combination of the actual total corneal power and virtual axial length.

PURPOSE: This study aims to investigate whether a combination of the total corneal power (TCP) and virtual axial length (AL) based on Gaussian optics makes the refractive prediction accuracy of the Barrett Universal II (BUII) formula better than the conventional anterior keratometry (K) and axial length. METHODS: The TCP and the virtual AL were calculated in two ways: the corneal index strategy and the TK index strategy. The former uses the corneal refractive index n1 as a variable, and the latter uses the TK index nx as a variable. In a dataset of 225 eyes, the calculated TCP and the virtual AL were input into the BUII formula along with the anterior chamber depth, lens thickness, and white-to-white measured with the IOLMaster 700, and the refractive prediction accuracy was evaluated by the mean numerical prediction error (MNE), standard deviation (SD), mean absolute prediction error (MAE), median absolute prediction error (MedAE), percentages of eyes with prediction error (PE) within ± 0.50 diopter, and IOL formula performance index (FPI). The refractive prediction outcomes also underwent subgroup analyses and were compared with those of the anterior keratometry-based BUII-K of the IOLMaster 700. RESULTS: In the corneal index strategy, the FPI had the highest value at approximately n1 = 1.346. In the TK index strategy, the FPI had the highest value at approximately nx = 1.3858. There was no tendency for the refractive prediction outcomes of the BUII-n1 = 1.346 and the BUII-nx = 1.3858 to be inferior to those of the BUII-K, particularly in the medium range of subgroups. CONCLUSION: The combination of the actual TCP and the virtual AL based on Gaussian optics may lead to a better refractive prediction accuracy of the BUII formula than that of BUII-K.

Accuracy of refractive outcomes using standard or total keratometry for intraocular lens power formulas in conventional cataract surgery.

PURPOSE: To evaluate if total keratometry (TK) is better than standard keratometry (K) for predicting an accurate intraocular lens (IOL) refractive outcome in virgin eyes using four IOL power calculation formulas. METHODS: 447 eyes that underwent monofocal intraocular lens implantation were enrolled in this study. IOLMaster 700 (Carl Zeiss Meditech, Jena, Germany) was used for optical biometry. Prediction error (PE), mean absolute prediction error (MAE), median absolute prediction error (MedAE), proportions of eyes within ± 0.25 diopters (D), ± 0.50 D, ± 0.75 D, ± 1.00 D, ± 2.00 D prediction error, and formula performance index (FPI) were calculated for each K- and TK-based formula. RESULTS: Overall, the accuracy of each TK and K formula was comparable. The MAEs and MedAEs showed no difference between most of the K-based and the TK-based formula; only the MAE of TK was significantly higher than that of K using the Haigis. The percent of eyes within ± 0.25 D PE for TK was not significantly different from that for K. The analysis of PE across various optical dimensions revealed that TK had no effect on the refractive results in eyes with different preoperative axial length, anterior chamber depth, keratometry, and lens thickness. The K-based Barrett Universal II formula performed excellently, showed the leading FPI score, and had the best refractive prediction outcomes among the four formulas. CONCLUSION: TK and K can be used for IOL power calculation in monofocal IOL implantation cataract surgery in virgin eyes, as both are comparable. In all investigated formulas, the predictive accuracy of TK-based formulas is not superior to that of standard K-based formulas.

Weighted genomic prediction for growth and carcass-related traits in Nelore cattle.

This study aimed to assess the impact of differential weighting in genomic regions harboring candidate causal loci on the genomic prediction accuracy and dispersion for growth and carcass-related traits in Nelore cattle. The dataset contained 168 793 phenotypic records for adjusted weight at 450 days of age (W450), 83 624 for rib eye area (REA), 24 480 for marbling (MAR) and 82 981 for subcutaneous backfat thickness (BFT) and rump fat thickness (RFT). The pedigree harbored information from 244 254 animals born between 1977 and 2016, including 6283 sires and 50 742 dams. Animals (n = 7769) were genotyped with the low-density panel (Clarifide® Nelore 3.0), and the genotypes were imputed to a panel containing 735 044 markers. A linear animal model was applied to estimate the genetic parameters and to perform the weighted single-step genome-wide association study (WssGWAS). A total of seven models for genomic prediction were evaluated combining the SNP weights obtained in the iterations of the WssGWAS and the candidate QTL. The heritability estimated for W450 (0.35) was moderate, and for carcass-related traits, the estimates were moderate for REA (0.27), MAR (0.28) and RFT (0.28), and low for BFT (0.18). The prediction accuracy for W450 incorporating reported QTL previously described in the literature along with different SNPs weights was like those described for the default ssGBLUP model. The use of the ssGWAS to weight the SNP effects displayed limited advantages for the REA prediction accuracy. Comparing the ssGBLUP with the BLUP model, a meaningful improvement in the prediction accuracy from 0.09 to 0.63 (700%) was observed for MAR. The highest prediction accuracy was obtained for BFT and RFT in all evaluated models. The application of information obtained from the WssGWAS is an alternative to reduce the genomic prediction dispersion for growth and carcass-related traits, except for MAR. Furthermore, the results obtained herein pointed out that is possible to improve the prediction accuracy and reduce the genomic prediction dispersion for growth and carcass-related traits in young animals.

Predicting milk protein fractions using infrared spectroscopy and a gradient boosting machine for breeding purposes in Holstein cattle.

In recent years, increasing attention has been focused on the genetic evaluation of protein fractions in cow milk with the aim of improving milk quality and technological characteristics. In this context, advances in high-throughput phenotyping by Fourier transform infrared (FTIR) spectroscopy offer the opportunity for large-scale, efficient measurement of novel traits that can be exploited in breeding programs as indicator traits. We took milk samples from 2,558 Holstein cows belonging to 38 herds in northern Italy, operating under different production systems. Fourier transform infrared spectra were collected on the same day as milk sampling and stored for subsequent analysis. Two sets of data (i.e., phenotypes and FTIR spectra) collected in 2 different years (2013 and 2019-2020) were compiled. The following traits were assessed using HPLC: true protein, major casein fractions [αS1-casein (CN), αS2-CN, ß-CN, κ-CN, and glycosylated-κ-CN], and major whey proteins (ß-lactoglobulin and α-lactalbumin), all of which were measured both in grams per liter (g/L) and proportion of total nitrogen (% N). The FTIR predictions were calculated using the gradient boosting machine technique and tested by 3 different cross-validation (CRV) methods. We used the following CRV scenarios: (1) random 10-fold, which randomly split the whole into 10-folds of equal size (9-folds for training and 1-fold for validation); (2) herd/date-out CRV, which assigned 80% of herd/date as the training set with independence of 20% of herd/date assigned as the validation set; (3) forward/backward CRV, which split the data set in training and validation set according with the year of milk sampling (FTIR and gold standard data assessed in 2013 or 2019-2020) using the "old" and "new" databases for training and validation, and vice-versa with independence among them; (4) the CRV for genetic parameters (CRV-gen), where animals without pedigree as assigned as a fixed training population and animals with pedigree information was split in 5-folds, in which 1-fold was assigned to the fixed training population, and 4-folds were assigned to the validation set (independent from the training set). The results (i.e., measures and predictions) of CRV-gen were used to infer the genetic parameters for gold standard laboratory measurements (i.e., proteins assessed with HPLC) and FTIR-based predictions considering the CRV-gen scenario from a bi-trait animal model using single-step genomic BLUP. We found that the prediction accuracies of the gradient boosting machine equations differed according to the way in which the proteins were expressed, achieving higher accuracy when expressed in g/L than when expressed as % N in all CRV scenarios. Concerning the reproducibility of the equations over the different years, the results showed no relevant differences in predictive ability between using "old" data as the training set and "new" data as the validation set and vice-versa. Comparing the additive genetic variance estimates for milk protein fractions between the FTIR predicted and HPLC measures, we found reductions of -19.7% for milk protein fractions expressed in g/L, and -21.19% expressed as % N. Although we found reductions in the heritability estimates, they were small, with values ranging from -1.9 to -7.25% for g/L, and -1.6 to -7.9% for % N. The posterior distributions of the additive genetic correlations (ra) between the FTIR predictions and the laboratory measurements were generally high (>0.8), even when the milk protein fractions were expressed as % N. Our results show the potential of using FTIR predictions in breeding programs as indicator traits for the selection of animals to enhance milk protein fraction contents. We expect acceptable responses to selection due to the high genetic correlations between HPLC measurements and FTIR predictions.

Assessing optimal frequency for image acquisition in computer vision systems developed to monitor feeding behavior of group-housed Holstein heifers.

Computer vision systems have emerged as a potential tool to monitor the behavior of livestock animals. Such high-throughput systems can generate massive redundant data sets for training and inference, which can lead to higher computational and economic costs. The objectives of this study were (1) to develop a computer vision system to individually monitor detailed feeding behaviors of group-housed dairy heifers, and (2) to determine the optimal frequency of image acquisition to perform inference with minimal effect on feeding behavior prediction quality. Eight Holstein heifers (96 ± 6 d old) were housed in a group and a total of 25,214 images (1 image every second) were acquired using 1 RGB camera. A total of 2,209 images were selected and each animal in the image was labeled with its respective identification (1-8). The label was annotated only on animals that were at the feed bunk (head through the feed rail). From the labeled images, 1,392 were randomly selected to train a deep learning algorithm for object detection with YOLOv3 ("You Only Look Once" version 3) and 154 images were used for validation. An independent data set (testing set = 663 out of the 2,209 images) was used to test the algorithm. The average accuracy for identifying individual animals in the testing set was 96.0%, and for each individual heifer from 1 to 8 the accuracy was 99.2, 99.6, 99.2, 99.6, 99.6, 99.2, 99.4, and 99.6%, respectively. After identifying the animals at the feed bunk, we computed the following feeding behavior parameters: number of visits (NV), mean visit duration (MVD), mean interval between visits (MIBV), and feeding time (FT) for each heifer using a data set composed by 8,883 sequential images (1 image every second) from 4 time points. The coefficient of determination (R2) was 0.39, 0.78, 0.48, and 0.99, and the root mean square error (RMSE) were 12.3 (count), 0.78, 0.63, and 0.31 min for NV, MVD, MIBV, and FT, respectively, considering 1 image every second. When we moved from 1 image per second to 1 image every 5 (MIBV) or 10 (NV, MDV, and FT) s, the R2 observed were 0.55 (NV), 0.74 (MVD), 0.70 (MIBV), and 0.99 (FT); and the RMSE were 2.27 (NV, count), 0.38 min (MVD), 0.22 min (MIBV), and 0.44 min (FT). Our results indicate that computer vision systems can be used to individually identify group-housed Holstein heifers (overall accuracy = 99.4%). Based on individual identification, feeding behavior such as MVD, MIBV, and FT can be monitored with reasonable accuracy and precision. Regardless of the frequency for optimal image acquisition, our results suggested that longer time intervals of image acquisition would reduce data collecting and model inference while maintaining adequate predictive performance. However, we did not find an optimal time interval for all feeding behavior; instead, the optimal frequency of image acquisition is phenotype-specific. Overall, the best R2 and RMSE for NV, MDV, and FT were achieved using 1 image every 10 s, and for MIBV it was achieved using 1 image every 5 s, and in both cases model inference and data storage could be drastically reduced.

Development of a prediction model for target positioning by using diaphragm waveforms extracted from CBCT projection images.

PURPOSE: To develop a prediction model (PM) for target positioning using diaphragm waveforms extracted from CBCT projection images. METHODS: Nineteen patients with lung cancer underwent orthogonal rotational kV x-ray imaging lasting 70 s. IR markers placed on their abdominal surfaces and an implanted gold marker located nearest to the tumor were considered as external surrogates and the target, respectively. Four different types of regression-based PM were trained using surrogate motions and target positions for the first 60 s, as follows: Scenario A: Based on the clinical scenario, 3D target positions extracted from projection images were used as they were (PMCL ). Scenario B: The short-arc 4D-CBCT waveform exhibiting eight target positions was obtained by averaging the target positions in Scenario A. The waveform was repeated for 60 s (W4D-CBCT ) by adapting to the respiratory phase of the external surrogate. W4D-CBCT was used as the target positions (PM4D-CBCT ). Scenario C: The Amsterdam Shroud (AS) signal, which depicted the diaphragm motion in the superior-inferior direction was extracted from the orthogonal projection images. The amplitude and phase of W4D-CBCT were corrected based on the AS signal. The AS-corrected W4D-CBCT was used as the target positions (PMAS-4D-CBCT ). Scenario D: The AS signal was extracted from single projection images. Other processes were the same as in Scenario C. The prediction errors were calculated for the remaining 10 s. RESULTS: The 3D prediction error within 3 mm was 77.3% for PM4D-CBCT , which was 12.8% lower than that for PMCL . Using the diaphragm waveforms, the percentage of errors within 3 mm improved by approximately 7% to 84.0%-85.3% for PMAS-4D-CBCT in Scenarios C and D, respectively. Statistically significant differences were observed between the prediction errors of PM4D-CBCT and PMAS-4D-CBCT . CONCLUSION: PMAS-4D-CBCT outperformed PM4D-CBCT , proving the efficacy of the AS signal-based correction. PMAS-4D-CBCT would make it possible to predict target positions from 4D-CBCT images without gold markers.

Four kinds of body temperatures and their relationships with thermal perception.

Body temperature serves as the principal factor in thermal perception determination. Current thermal comfort researches mainly focused on skin temperature, while other kinds of body temperatures were often ignored. In laboratory with strictly controlled environment, 26 subjects (13 males and 13 females) remained seated for a duration of 130 min in two thermal environments (19 °C and 35 °C), arranged in a particular order; four kinds of body temperatures (skin temperature, oral temperature, auditory canal temperature and breath temperature) and three kinds of thermal perception votes (thermal sensation, thermal comfort and thermal acceptable) were regularly collected. The analysis results showed that, skin temperature and breath temperature significantly changed with ambient temperature (p < 0.001); the difference between average value of core temperature in two conditions was small (≤0.3 °C), but a significant difference was almost observed in auditory canal temperature of males (p = 0.07). Both skin temperature and breath temperature were significantly related with three subjective votes (p < 0.001), meanwhile, the prediction accuracy of breath temperature for thermal perception was in no way inferior to skin temperature. Although oral temperature and auditory canal temperature had partial significant correlations with thermal perception, they were difficult to be carried out in practical application due to their weak explanatory powers (correlation coefficient <0.3). In summary, this research tried to establish correlation laws between body temperatures and thermal perception votes during a temperature step-change experiment, while finding the potential of utilizing breath temperature for thermal perception prediction, which is expected to be further promoted in the future.

Multi-phase optimisation model predicts manual lifting motions with less reliance on experiment-based posture data.

Optimisation-based predictive models are widely-used to explore the lifting strategies. Existing models incorporated empirical subject-specific posture constraints to improve the prediction accuracy. However, over-reliance on these constraints limits the application of predictive models. This paper proposed a multi-phase optimisation method (MPOM) for two-dimensional sagittally symmetric semi-squat lifting prediction, which decomposes the complete lifting task into three phases-the initial posture, the final posture, and the dynamic lifting phase. The first two phases are predicted with force- and stability-related strategies, and the last phase is predicted with a smoothing-related objective. Box-lifting motions of different box initial heights were collected for validation. The results show that MPOM has better or similar accuracy than the traditional single-phase optimisation (SPOM) of minimum muscular utilisation ratio, and MPOM reduces the reliance on experimental data. MPOM offers the opportunity to improve accuracy at the expense of efforts to determine appropriate weightings in the posture prediction phases. Practitioner summary: Lifting optimisation models are useful to predict and explore the human motion strategies. Existing models rely on empirical subject-specific posture constraints, which limit their applications. A multi-phase model for lifting motion prediction was constructed. This model could accurately predict 2D lifting motions with less reliance on these constraints.

A coupled CEEMD-BiLSTM model for regional monthly temperature prediction.

Temperature is an important indicator of climate change. With the gradual increase of global warming, a well-chosen model can improve the accuracy of temperature prediction. It is of great significance and value for future disaster prevention and mitigation and economic development. Monthly temperature is influenced by solar activity, monsoon, and other factors, with significant uncertainty, ambiguity, and randomness. A coupled CEEMD-BiLSTM temperature model is constructed based on the good decomposition-reconstruction characteristics of CEEMD for uncertain time series and the advantages of BiLSTM for solving stochastic prediction, and it is applied to the prediction of monthly temperature in Zhengzhou City. The results show that the minimum relative error of the coupled CEEMD-BiLSTM model is 0.01%, the maximum relative error is 0.99%, and the average relative error is 0.22%, and the prediction accuracy of this coupled model for monthly temperature in Zhengzhou is higher than that of the CEEMD-LSTM model, EEMD-BiLSTM model, and BP neural network model, with better stability and adaptability.

ACO-KELM: Anti Coronavirus Optimized Kernel-based Softplus Extreme Learning Machine for Classification of Skin Cancer.

Due to the presence of redundant and irrelevant features in large-dimensional biomedical datasets, the prediction accuracy of disease diagnosis can often be decreased. Therefore, it is important to adopt feature extraction methodologies that can deal with problem structures and identify underlying data patterns. In this paper, we propose a novel approach called the Anti Coronavirus Optimized Kernel-based Softplus Extreme Learning Machine (ACO-KSELM) to accurately predict different types of skin cancer by analyzing high-dimensional datasets. To evaluate the proposed ACO-KSELM method, we used four different skin cancer image datasets: ISIC 2016, ACS, HAM10000, and PAD-UFES-20. These dermoscopic image datasets were preprocessed using Gaussian filters to remove noise and artifacts, and relevant features based on color, texture, and shape were extracted using color histogram, Haralick texture, and Hu moment extraction approaches, respectively. Finally, the proposed ACO-KSELM method accurately predicted and classified the extracted features into Basal Cell Carcinoma (BCC), Squamous Cell Carcinoma (SCC), Actinic Keratosis (ACK), Seborrheic Keratosis (SEK), Bowen's disease (BOD), Melanoma (MEL), and Nevus (NEV) categories. The analytical results showed that the proposed method achieved a higher rate of prediction accuracy of about 98.9%, 98.7%, 98.6%, and 97.9% for the ISIC 2016, ACS, HAM10000, and PAD-UFES-20 datasets, respectively.