Least Absolute Shrinkage And Selection Operator Model Research Articles

Relationships Between the External and Internal Training Load in Professional Soccer: What Can We Learn From Machine Learning?

Machine learning may contribute to understanding the relationship between the external load and internal load in professional soccer. Therefore, the relationship between external load indicators (ELIs) and the rating of perceived exertion (RPE) was examined using machine learning techniques on a group and individual level. Training data were collected from 38 professional soccer players over 2 seasons. The external load was measured using global positioning system technology and accelerometry. The internal load was obtained using the RPE. Predictive models were constructed using 2 machine learning techniques, artificial neural networks and least absolute shrinkage and selection operator (LASSO) models, and 1 naive baseline method. The predictions were based on a large set of ELIs. Using each technique, 1 group model involving all players and 1 individual model for each player were constructed. These models' performance on predicting the reported RPE values for future training sessions was compared with the naive baseline's performance. Both the artificial neural network and LASSO models outperformed the baseline. In addition, the LASSO model made more accurate predictions for the RPE than did the artificial neural network model. Furthermore, decelerations were identified as important ELIs. Regardless of the applied machine learning technique, the group models resulted in equivalent or better predictions for the reported RPE values than the individual models. Machine learning techniques may have added value in predicting RPE for future sessions to optimize training design and evaluation. These techniques may also be used in conjunction with expert knowledge to select key ELIs for load monitoring.

Model selection and prediction of outcomes in recent onset schizophrenia patients who undergo cognitive training

Predicting treatment outcomes in psychiatric populations remains a challenge, but is increasingly important in the pursuit of personalized medicine. Patients with schizophrenia have deficits in cognition, and targeted cognitive training (TCT) of auditory processing and working memory has been shown to improve some of these impairments; but little is known about the baseline patient characteristics predictive of cognitive improvement. Here we use a model selection and regression approach called least absolute shrinkage and selection operator (LASSO) to examine predictors of cognitive improvement in response to TCT for patients with recent onset schizophrenia. Forty-three individuals with recent onset schizophrenia randomized to undergo TCT were assessed at baseline on measures of cognition, symptoms, functioning, illness duration, and demographic variables. We carried out 10-fold cross-validation of LASSO for model selection and regression. We followed up on these results using linear models for statistical inference. No individual variable was found to correlate with improvement in global cognition using a Pearson correlation approach, and a linear model including all variables was also found not to be significant. However, the LASSO model identified baseline global cognition, education, and gender in a model predictive of improvement on global cognition following TCT. These findings offer guidelines for personalized approaches to cognitive training for patients with schizophrenia.

Generalized interaction LASSO based on alternating direction method of multipliers for liver disease classification

Features and interaction between features of liver disease is of great significance for the classification of liver disease. Based on least absolute shrinkage and selection operator (LASSO) and interaction LASSO, the generalized interaction LASSO model is proposed in this paper for liver disease classification and compared with other methods. Firstly, the generalized interaction logistic classification model was constructed and the LASSO penalty constraints were added to the interactive model parameters. Then the model parameters were solved by an efficient alternating directions method of multipliers (ADMM) algorithm. The solutions of model parameters were sparse. Finally, the test samples were fed to the model and the classification results were obtained by the largest statistical probability. The experimental results of liver disorder dataset and India liver dataset obtained by the proposed methods showed that the coefficients of interaction features of the model were not zero, indicating that interaction features were contributive to classification. The accuracy of the generalized interaction LASSO method is better than that of the interaction LASSO method, and it is also better than that of traditional pattern recognition methods. The generalized interaction LASSO method can also be popularized to other disease classification areas.

A data-driven approach to modeling physical fatigue in the workplace using wearable sensors

Wearable sensors are currently being used to manage fatigue in professional athletics, transportation and mining industries. In manufacturing, physical fatigue is a challenging ergonomic/safety “issue” since it lowers productivity and increases the incidence of accidents. Therefore, physical fatigue must be managed. There are two main goals for this study. First, we examine the use of wearable sensors to detect physical fatigue occurrence in simulated manufacturing tasks. The second goal is to estimate the physical fatigue level over time. In order to achieve these goals, sensory data were recorded for eight healthy participants. Penalized logistic and multiple linear regression models were used for physical fatigue detection and level estimation, respectively. Important features from the five sensors locations were selected using Least Absolute Shrinkage and Selection Operator (LASSO), a popular variable selection methodology. The results show that the LASSO model performed well for both physical fatigue detection and modeling. The modeling approach is not participant and/or workload regime specific and thus can be adopted for other applications.

Use of a least absolute shrinkage and selection operator (LASSO) model to selected ion flow tube mass spectrometry (SIFT-MS) analysis of exhaled breath to predict the efficacy of dialysis: a pilot study

Selected ion flow tube-mass spectrometry (SIFT-MS) provides rapid, non-invasive measurements of a full-mass scan of volatile compounds in exhaled breath. Although various studies have suggested that breath metabolites may be indicators of human disease status, many of these studies have included few breath samples and large numbers of compounds, limiting their power to detect significant metabolites. This study employed a least absolute shrinkage and selective operator (LASSO) approach to SIFT-MS data of breath samples to preliminarily evaluate the ability of exhaled breath findings to monitor the efficacy of dialysis in hemodialysis patients. A process of model building and validation showed that blood creatinine and urea concentrations could be accurately predicted by LASSO-selected masses. Using various precursors, the LASSO models were able to predict creatinine and urea concentrations with high adjusted R-square (>80%) values. The correlation between actual concentrations and concentrations predicted by the LASSO model (using precursor H3O+) was high (Pearson correlation coefficient = 0.96). Moreover, use of full mass scan data provided a better prediction than compounds from selected ion mode. These findings warrant further investigations in larger patient cohorts. By employing a more powerful statistical approach to predict disease outcomes, breath analysis using SIFT-MS technology could be applicable in future to daily medical diagnoses.

18F-Fluorodeoxyglucose Positron Emission Tomography Can Quantify and Predict Esophageal Injury During Radiation Therapy

We sought to investigate the ability of mid-treatment (18)F-fluorodeoxyglucose positron emission tomography (PET) studies to objectively and spatially quantify esophageal injury invivo from radiation therapy for non-small cell lung cancer. This retrospective study was approved by the local institutional review board, with written informed consent obtained before enrollment. We normalized (18)F-fluorodeoxyglucose PET uptake to each patient's low-irradiated region (<5Gy) of the esophagus, as a radiation response measure. Spatially localized metrics of normalized uptake (normalized standard uptake value [nSUV]) were derived for 79 patients undergoing concurrent chemoradiation therapy for non-small cell lung cancer. We used nSUV metrics to classify esophagitis grade at the time of the PET study, as well as maximum severity by treatment completion, according to National Cancer Institute Common Terminology Criteria for Adverse Events, using multivariate least absolute shrinkage and selection operator (LASSO) logistic regression and repeated 3-fold cross validation (training, validation, and test folds). This 3-fold cross-validation LASSO model procedure was used to predict toxicity progression from 43 asymptomatic patients during the PET study. Dose-volume metrics were also tested in both the multivariate classification and the symptom progression prediction analyses. Classification performance was quantified with the area under the curve (AUC) from receiver operating characteristic analysis on the test set from the 3-fold analyses. Statistical analysis showed increasing nSUV is related to esophagitis severity. Axial-averaged maximum nSUV for 1 esophageal slice and esophageal length with at least 40% of axial-averaged nSUV both had AUCs of 0.85 for classifying grade 2 or higher esophagitis at the time of the PET study and AUCs of 0.91 and 0.92, respectively, for maximum grade 2 or higher by treatment completion. Symptom progression was predicted with an AUC of 0.75. Dose metrics performed poorly at classifying esophagitis (AUC of 0.52, grade 2 or higher mid treatment) or predicting symptom progression (AUC of 0.67). Normalized uptake can objectively, locally, and noninvasively quantify esophagitis during radiation therapy and predict eventual symptoms from asymptomatic patients. Normalized uptake may provide patient-specific dose-response information not discernible from dose.

Accuracy of genomic breeding values for meat tenderness in Polled Nellore cattle.

Zebu () cattle, mostly of the Nellore breed, comprise more than 80% of the beef cattle in Brazil, given their tolerance of the tropical climate and high resistance to ectoparasites. Despite their advantages for production in tropical environments, zebu cattle tend to produce tougher meat than Bos taurus breeds. Traditional genetic selection to improve meat tenderness is constrained by the difficulty and cost of phenotypic evaluation for meat quality. Therefore, genomic selection may be the best strategy to improve meat quality traits. This study was performed to compare the accuracies of different Bayesian regression models in predicting molecular breeding values for meat tenderness in Polled Nellore cattle. The data set was composed of Warner-Bratzler shear force (WBSF) of longissimus muscle from 205, 141, and 81 animals slaughtered in 2005, 2010, and 2012, respectively, which were selected and mated so as to create extreme segregation for WBSF. The animals were genotyped with either the Illumina BovineHD (HD; 777,000 from 90 samples) chip or the GeneSeek Genomic Profiler (GGP Indicus HD; 77,000 from 337 samples). The quality controls of SNP were Hard-Weinberg Proportion -value ≥ 0.1%, minor allele frequency > 1%, and call rate > 90%. The FImpute program was used for imputation from the GGP Indicus HD chip to the HD chip. The effect of each SNP was estimated using ridge regression, least absolute shrinkage and selection operator (LASSO), Bayes A, Bayes B, and Bayes Cπ methods. Different numbers of SNP were used, with 1, 2, 3, 4, 5, 7, 10, 20, 40, 60, 80, or 100% of the markers preselected based on their significance test (-value from genomewide association studies [GWAS]) or randomly sampled. The prediction accuracy was assessed by the correlation between genomic breeding value and the observed WBSF phenotype, using a leave-one-out cross-validation methodology. The prediction accuracies using all markers were all very similar for all models, ranging from 0.22 (Bayes Cπ) to 0.25 (Bayes B). When preselecting SNP based on GWAS results, the highest correlation (0.27) between WBSF and the genomic breeding value was achieved using the Bayesian LASSO model with 15,030 (3%) markers. Although this study used relatively few animals, the design of the segregating population ensured wide genetic variability for meat tenderness, which was important to achieve acceptable accuracy of genomic prediction. Although all models showed similar levels of prediction accuracy, some small advantages were observed with the Bayes B approach when higher numbers of markers were preselected based on their -values resulting from a GWAS analysis.

Least absolute shrinkage and selection operator and dimensionality reduction techniques in quantitative structure retention relationship modeling of retention in hydrophilic interaction liquid chromatography

The objective of this study was to model the retention of nucleosides and pterins in hydrophilic interaction liquid chromatography (HILIC) via QSRR-based approach.Two home-made (Amino-P-C18, Amino-P-C10) and one commercial (IAM.PC.DD2) HILIC stationary phases were considered. Logarithm of retention factor at 5% of acetonitrile (logkACN) along with descriptors obtained for 16 nucleosides and 11 pterins were used to develop QSRR models. We used and compared the predictive performance of three regression techniques: partial least square (PLS), the least absolute shrinkage and selection operator (LASSO), and the LASSO followed by stepwise multiple linear regression.The highest predictive squared correlation coefficient (QLOOCV2) in PLS analysis was found for Amino-P-C10 (QLOOCV2=0.687) and IAM.PC.DD2 (QLOOCV2=0.506) and the lowest for IAM.PC.DD2 (QLOOCV2=−0.01). Much higher values were obtained for the LASSO model. The QLOOCV2 equaled 0.9 for Amino-P-C10, 0.66 for IAM.PC.DD2 and 0.59 for Amino-P-C18. The combination of LASSO with stepwise regression provided models with comparable predictive performance as the LASSO, however with possibility of calculating the standard error of estimates.The use of LASSO itself and in combination with classical stepwise regression may offer greater stability of the developed models thanks to more smooth change of coefficients and reduced susceptibility towards chance correlation. Application of QSRR-based approach, along with the computational methods proposed in this work, may offer a useful approach in the modeling of retention of nucleoside and pterin compounds in HILIC.

A New Sparsity-Based Band Selection Method for Target Detection of Hyperspectral Image

Band selection (BS) plays an important role in the dimensionality reduction of hyperspectral data. However, as to the existing BS methods, few are specially designed for target detection. In this letter, we combine the target detection and BS process together and put forward a new BS method for target detection, named least absolute shrinkage and selection operator (LASSO)-based BS (LBS). Interestingly, by using a linear regression model with L1 regularization (LASSO model), LBS transforms the discrete BS problem into the continuous optimization problem, which cannot only avoid the complicated subset selection process but also evaluate the importance of all the bands simultaneously. The experiments on real hyperspectral data demonstrate that LBS is a very effective BS method for target detection.

A gradient method for the monotone fused least absolute shrinkage and selection operator

We propose the monotone fused least absolute shrinkage and selection operator (LASSO) model and develop a continuous algorithm for it. The LASSO model is a special case of the fused LASSO model. The LASSO technique improves prediction accuracy and reduces the number of predictors, while the fused LASSO procedure also encourages flatness of the regression predictors. The monotone fused LASSO model describes regression with monotonic constraints better than the fused LASSO model. We adapt Nesterov's fast gradient methods to the monotone fused LASSO model, we give closed-form solutions for each iteration and prove the boundedness of the optimal solution set, and we provide convergence results. Numerical examples are provided and discussed. Our approach can easily be adapted to related problems, such as the monotone regression and the fused LASSO model.

Metabolomics reveals determinants of weight loss during lifestyle intervention in obese children

The amount of weight loss in obese children during lifestyle intervention differs strongly between individuals. The metabolic processes underlying this variability are largely unknown. We hypothesize that metabolomics analyses of serum samples might help to identify metabolic predictors of weight loss. In this study, we investigated 80 obese children aged 6–15 years having completed the one-year lifestyle intervention program ‘Obeldicks’, 40 that achieved a substantial reduction of their body mass index standard deviation score (BMI-SDS) during this intervention (defined as BMI-SDS reduction ≥ 0.5), and 40 that did not improve their overweight status (BMI-SDS reduction < 0.1). Anthropometric and clinical parameters were measured and baseline fasting serum samples of all children were analyzed with a mass spectrometry-based metabolomics approach targeting 163 metabolites. Both univariate regression models and a multivariate least absolute shrinkage and selection operator (LASSO) approach identified lower serum concentrations of long-chain unsaturated phosphatidylcholines as well as smaller waist circumference as significant predictors of BMI-SDS reduction during intervention (p-values univariate models: 5.3E−03 to 1.0E−04). A permutation test showed that the LASSO model explained a significant part of BMI-SDS change (p = 4.6E−03). Our results suggest a role of phosphatidylcholine metabolism and abdominal obesity in body weight regulation. These findings might lead to a better understanding of the mechanisms behind the large inter-individual variation in response to lifestyle interventions, which is a prerequisite for the development of individualized intervention programs.

Non-invasive continuous glucose monitoring: improved accuracy of point and trend estimates of the Multisensor system

Non-invasive continuous glucose monitoring (NI-CGM) sensors are still at an early stage of development, but, in the near future, they could become particularly appealing in diabetes management. Solianis Monitoring AG (Zurich, Switzerland) has proposed an approach for NI-CGM based on a multi-sensor concept, embedding primarily dielectric spectroscopy and optical sensors. This concept requires a mathematical model able to estimate glucose levels from the 150 channels directly measured through the Multisensor. A static multivariate linear regression model (with order and parameters common to the entire population of subjects) was proposed for such a scope (Caduff et al., Biosens Bioelectron 26:3794-3800, 2011). The aim of this work is to evaluate the accuracy in the estimation of glucose levels and trends that the NI-CGM Multisensor platform can achieve by exploiting different techniques for model identification, namely, ordinary least squares, subset variable selection, partial least squares and least absolute shrinkage and selection operator (LASSO). Data collected in human beings monitored for a total of 45 study days were used for model identification and model test. Several metrics of standard use in the diabetes scientific community to measure point and clinical accuracy of glucose sensors were used to assess the models. Results indicate that the LASSO technique is superior to the others shrinking many channel weights to zero thus leading to smoother glucose profiles and resulting in a more robust model to possible artifacts in the Multisensor data. Although, as expected, the performance of the NI-CGM system with the LASSO model is not yet comparable with that of enzyme-based needle glucose sensors, glucose trends are satisfactorily estimated. Considering the non-invasive nature of the multi-sensor platform, this result can have an immediate impact in the current clinical practice, e.g., to integrate sparse self-monitoring of blood glucose data with an indication of the glucose trend to aid the diabetic patient in dealing with, or even preventing in the short time scale, the threats of critical events such as hypoglycaemia.

LASSO based stimulus frequency recognition model for SSVEP BCIs

Steady-state visual evoked potential (SSVEP) has been increasingly used for the study of brain–computer interface (BCI). How to recognize SSVEP with shorter time and lower error rate is one of the key points to develop a more efficient SSVEP-based BCI. To achieve this goal, we make use of the sparsity constraint of the least absolute shrinkage and selection operator (LASSO) for the extraction of more discriminative features of SSVEP, and then we propose a LASSO model using the linear regression between electroencephalogram (EEG) recordings and the standard square-wave signals of different frequencies to recognize SSVEP without the training stage. In this study, we verified the proposed LASSO model offline with the EEG data of nine healthy subjects in contrast to canonical correlation analysis (CCA). In the experiment, when a shorter time window was used, we found that the LASSO model yielded better performance in extracting robust and detectable features of SSVEP, and the information transfer rate obtained by the LASSO model was significantly higher than that of the CCA. Our proposed method can assist to reduce the recording time without sacrificing the classification accuracy and is promising for a high-speed SSVEP-based BCI.