Diagnosis of dementias using partial least squares discriminant analysis

An Adaptive Partial Least-Squares Regression Approach for Classifying Chicken Egg Fertility by Hyperspectral Imaging

Partial least-squares (PLS) regression is a well known chemometric method used for predictive modelling, especially in the presence of many variables. Although PLS was not initially developed as a technique for classification tasks, scientists have reportedly used this approach successfully for discrimination purposes. Whereas some non-supervised learning approaches, including, but not limited to, PCA and k-means clustering, do well in identifying/understanding grouping and clustering patterns in multidimensional data, they are limited when the end target is discrimination, making PLS a preferable alternative. Hyperspectral imaging data on a total of 672 fertilized chicken eggs, consisting of 336 white eggs and 336 brown eggs, were used in this study. Hyperspectral images in the NIR region of the 900-1700 nm wavelength range were captured prior to incubation on day 0 and on days 1-4 after incubation. Eggs were candled on incubation day 5 and broken out on day 10 to confirm fertility. While a total number of 312 and 314 eggs were found to be fertile in the brown and white egg batches, respectively, the total number of non-fertile eggs in the same set of batches was 23 and 21, respectively. Spectral information was extracted from a segmented region of interest (ROI) of each hyperspectral image and spectral transmission characteristics were obtained by averaging the spectral information. A moving-thresholding technique was implemented for discrimination based on PLS regression results on the calibration set. With true positive rates (TPRs) of up to 100% obtained at selected threshold values of between 0.50 and 0.85 and on different days of incubation, the results indicate that the proposed PLS technique can accurately discriminate between fertile and non-fertile eggs. The adaptive PLS approach was, thereby, presented as suitable for handling hyperspectral imaging-based chicken egg fertility data.

Machine Learning for Prediction, Classification, and Identification of Immobilized Enzymes for Biocatalysis

BACKGROUND

Enzyme immobilization is a beneficial component involved in biocatalytic strategies. Understanding and evaluating the enzyme immobilization system plays an important role in the successful development and implementation of the biocatalysis route. Ensuring the implementation of a successful enzyme immobilization process is vital for realizing a highly functioning and well suited biocatalytic process within pharmaceutical development.

AIM

To develop a method which can accurately and objectively identify and classify differences within enzyme immobilization systems, sample preparation methods, and data collection parameters.

METHODS

Raman hyperspectral imaging was used to obtain a total of eight spectral data sets from enzyme immobilization samples. Partial least squares discriminant analysis (PLS-DA) was used to classify and identify the samples based on their differences.

RESULTS

Several two-class, four-class, and eight-class PLS-DA models were built to classify the different sample data sets. All models reached between 92-100% accuracy after cross-validation and external validation, illustrating great success of the models for identifying differences between the samples.

CONCLUSION

Raman hyperspectral imaging with machine learning can be used to investigate, interpret, and classify different data collection parameters, sample preparation methods, and enzyme immobilization supports, providing crucial insight into enzyme immobilization process development.

Changes in Sulfur Metabolism in Mouse Brains following Radon Inhalation

Therapy using hot springs, including the high-level radioactive gas "radon", is traditionally conducted as an alternative treatment for various diseases. Oxidative-stress-related diseases are inhibited by the enhancement of antioxidative functions following radon inhalation. We have reported that radon inhalation increased the level of anti-oxidants, such as glutathione (G-SH), in the brain and had a protective antioxidative effect against transient global cerebral ischemic injury. However, no studies have yet revealed the changes in G-SH associated substances after radon inhalation. In this study, we comprehensively analyzed several metabolites, focusing on G-SH. Mice were exposed to radon at concentrations of 200, 2000, or 20,000 Bq/m³ for 1, 3, or 10 days. We detected 27 metabolites in the mouse brains. The result showed that the L-methionine levels increased, whereas the levels of urea, glutathione, and sulfite ion decreased under any condition. Although the ratio of G-SH to oxidized glutathione (GS-SG) decreased, glutathione monosulfide (G-S-SH) and cysteine monosulfide (Cys-S-SH) increased after radon inhalation. G-S-SH and Cys-S-SH can produce a biological defense against the imbalance of the redox state at very low-dose irradiation following radon inhalation because they are strong scavengers of reactive oxygen species. Additionally, we performed an overall assessment of high-dimensional data and showed some specific characteristics. We showed the changes in metabolites after radon inhalation using partial least squares-discriminant analysis and self-organizing maps. The results showed the health effects of radon, especially the state of sulfur-related metabolites in mouse brains under the exposure conditions for radon therapy.

So you think you can PLS-DA?

BACKGROUND

Partial Least-Squares Discriminant Analysis (PLS-DA) is a popular machine learning tool that is gaining increasing attention as a useful feature selector and classifier. In an effort to understand its strengths and weaknesses, we performed a series of experiments with synthetic data and compared its performance to its close relative from which it was initially invented, namely Principal Component Analysis (PCA).

RESULTS

We demonstrate that even though PCA ignores the information regarding the class labels of the samples, this unsupervised tool can be remarkably effective as a feature selector. In some cases, it outperforms PLS-DA, which is made aware of the class labels in its input. Our experiments range from looking at the signal-to-noise ratio in the feature selection task, to considering many practical distributions and models encountered when analyzing bioinformatics and clinical data. Other methods were also evaluated. Finally, we analyzed an interesting data set from 396 vaginal microbiome samples where the ground truth for the feature selection was available. All the 3D figures shown in this paper as well as the supplementary ones can be viewed interactively at http://biorg.cs.fiu.edu/plsda CONCLUSIONS: Our results highlighted the strengths and weaknesses of PLS-DA in comparison with PCA for different underlying data models.

Multimodal Classification of Mild Cognitive Impairment Based on Partial Least Squares

In recent years, increasing attention has been given to the identification of the conversion of mild cognitive impairment (MCI) to Alzheimer's disease (AD). Brain neuroimaging techniques have been widely used to support the classification or prediction of MCI. The present study combined magnetic resonance imaging (MRI), 18F-fluorodeoxyglucose PET (FDG-PET), and 18F-florbetapir PET (florbetapir-PET) to discriminate MCI converters (MCI-c, individuals with MCI who convert to AD) from MCI non-converters (MCI-nc, individuals with MCI who have not converted to AD in the follow-up period) based on the partial least squares (PLS) method. Two types of PLS models (informed PLS and agnostic PLS) were built based on 64 MCI-c and 65 MCI-nc from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. The results showed that the three-modality informed PLS model achieved better classification accuracy of 81.40%, sensitivity of 79.69%, and specificity of 83.08% compared with the single-modality model, and the three-modality agnostic PLS model also achieved better classification compared with the two-modality model. Moreover, combining the three modalities with clinical test score (ADAS-cog), the agnostic PLS model (independent data: florbetapir-PET; dependent data: FDG-PET and MRI) achieved optimal accuracy of 86.05%, sensitivity of 81.25%, and specificity of 90.77%. In addition, the comparison of PLS, support vector machine (SVM), and random forest (RF) showed greater diagnostic power of PLS. These results suggested that our multimodal PLS model has the potential to discriminate MCI-c from the MCI-nc and may therefore be helpful in the early diagnosis of AD.

Spatially aggregated multiclass pattern classification in functional MRI using optimally selected functional brain areas

In previous works, boosting aggregation of classifier outputs from discrete brain areas has been demonstrated to reduce dimensionality and improve the robustness and accuracy of functional magnetic resonance imaging (fMRI) classification. However, dimensionality reduction and classification of mixed activation patterns of multiple classes remain challenging. In the present study, the goals were (a) to reduce dimensionality by combining feature reduction at the voxel level and backward elimination of optimally aggregated classifiers at the region level, (b) to compare region selection for spatially aggregated classification using boosting and partial least squares regression methods and (c) to resolve mixed activation patterns using probabilistic prediction of individual tasks. Brain activation maps from interleaved visual, motor, auditory and cognitive tasks were segmented into 144 functional regions. Feature selection reduced the number of feature voxels by more than 50%, leaving 95 regions. The two aggregation approaches further reduced the number of regions to 30, resulting in more than 75% reduction of classification time and misclassification rates of less than 3%. Boosting and partial least squares (PLS) were compared to select the most discriminative and the most task correlated regions, respectively. Successful task prediction in mixed activation patterns was feasible within the first block of task activation in real-time fMRI experiments. This methodology is suitable for sparsifying activation patterns in real-time fMRI and for neurofeedback from distributed networks of brain activation.