Assessment of discriminant models in infrared imaging using constrained repeated random sampling - Cross validation

Molecular classifications of prostate cancer: basis for individualized risk stratification and precision therapy

Tumour classifications play a pivotal role in prostate cancer (PCa) management. It can predict the clinical outcomes of PCa as early as the disease is diagnosed and then guide therapeutic schemes, such as active monitoring, standalone surgical intervention, or surgery supplemented with postoperative adjunctive therapy, thereby circumventing disease exacerbation and excessive treatment. Classifications based on clinicopathological features, such as prostate cancer-specific antigen, Gleason score, and TNM stage, are still the main risk stratification strategies and have played an essential role in standardized clinical decision-making. However, mounting evidence indicates that clinicopathological parameters in isolation fail to adequately capture the heterogeneity exhibited among distinct PCa patients, such as those sharing identical Gleason scores yet experiencing divergent prognoses. As a remedy, molecular classifications have been introduced. Currently, molecular studies have revealed the characteristic genomic alterations, epigenetic modulations, and tumour microenvironment associated with different types of PCa, which provide a chance for urologists to refine the PCa classification. In this context, numerous invaluable molecular classifications have been devised, employing disparate statistical methodologies and algorithmic approaches, encompassing self-organizing map clustering, unsupervised cluster analysis, and multifarious algorithms. Interestingly, the classifier PAM50 was used in a phase-2 multicentre open-label trial, NRG-GU-006, for further validation, which hints at the promise of molecular classification for clinical use. Consequently, this review examines the extant molecular classifications, delineates the prevailing panorama of clinically pertinent molecular signatures, and delves into eight emblematic molecular classifications, dissecting their methodological underpinnings and clinical utility.

Forecasting and Optimizing Dual Media Filter Performance via Machine Learning

Four different machine learning algorithms, including Decision Tree (DT), Random Forest (RF), Multivariable Linear Regression (MLR), Support Vector Regressions (SVR), and Gaussian Process Regressions (GPR), were applied to predict the performance of a multi-media filter operating as a function of raw water quality and plant operating variables. The models were trained using data collected over a seven year period covering water quality and operating variables, including true colour, turbidity, plant flow, and chemical dose for chlorine, KMnO₄, FeCl₃, and Cationic Polymer (PolyDADMAC). The machine learning algorithms have shown that the best prediction is at a 1-day time lag between input variables and unit filter run volume (UFRV). Furthermore, the RF algorithm with grid search using the input metrics mentioned above with a 1-day time lag has provided the highest reliability in predicting UFRV with a RMSE and R² of 31.58 and 0.98, respectively. Similarly, RF with grid search has shown the shortest training time, prediction accuracy, and forecasting events using a ROC-AUC curve analysis (AUC over 0.8) in extreme wet weather events. Therefore, Random Forest with grid search and a 1-day time lag is an effective and robust machine learning algorithm that can predict the filter performance to aid water treatment operators in their decision makings by providing real-time warning of the potential turbidity breakthrough from the filters.

Addressing Delicate and Variable Cancer Morphology in Spectral Histopathology Using Canine Visceral Hemangiosarcoma

Spectral histopathology has shown promise for the classification and diagnosis of tumors with defined morphology, but application in tumors with variable or diffuse morphologies is yet to be investigated. To address this gap, we evaluated the application of Fourier transform infrared (FTIR) imaging as an accessory diagnostic tool for canine hemangiosarcoma (HSA), a vascular endothelial cell cancer that is difficult to diagnose. To preserve the delicate vascular tumor tissue structure, and potential classification of single endothelial cells, paraffin removal was not performed, and a partial least square discrimination analysis (PLSDA) and Random Forest (RF) models to classify different tissue types at individual pixel level were established using a calibration set (24 FTIR images from 13 spleen specimens). Next, the prediction capability of the PLSDA model was tested with an independent test set (n = 11), resulting in 74% correct classification of different tissue types at an individual pixel level. Finally, the performance of the FTIR spectropathology and chemometric algorithm for diagnosis of HSA was established in a blinded set of tissue samples (n = 24), with sensitivity and specificity of 80 and 81%, respectively. Taken together, these results show that FTIR imaging without paraffin removal can be applied to tumors with diffuse morphology, and this technique is a promising tool to assist in canine splenic HSA differential diagnosis.

The Impact of Preprocessing Methods for a Successful Prostate Cell Lines Discrimination Using Partial Least Squares Regression and Discriminant Analysis Based on Fourier Transform Infrared Imaging

Fourier transform infrared spectroscopy (FT-IR) is widely used in the analysis of the chemical composition of biological materials and has the potential to reveal new aspects of the molecular basis of diseases, including different types of cancer. The potential of FT-IR in cancer research lies in its capability of monitoring the biochemical status of cells, which undergo malignant transformation and further examination of spectral features that differentiate normal and cancerous ones using proper mathematical approaches. Such examination can be performed with the use of chemometric tools, such as partial least squares discriminant analysis (PLS-DA) classification and partial least squares regression (PLSR), and proper application of preprocessing methods and their correct sequence is crucial for success. Here, we performed a comparison of several state-of-the-art methods commonly used in infrared biospectroscopy (denoising, baseline correction, and normalization) with the addition of methods not previously used in infrared biospectroscopy classification problems: Mie extinction extended multiplicative signal correction, Eiler’s smoothing, and probabilistic quotient normalization. We compared all of these approaches and their effect on the data structure, classification, and regression capability on experimental FT-IR spectra collected from five different prostate normal and cancerous cell lines. Additionally, we tested the influence of added spectral noise. Overall, we concluded that in the case of the data analyzed here, the biggest impact on data structure and performance of PLS-DA and PLSR was caused by the baseline correction; therefore, much attention should be given, especially to this step of data preprocessing.

Repeated double cross-validation applied to the PCA-LDA classification of SERS spectra: a case study with serum samples from hepatocellular carcinoma patients

Intense label-free surface-enhanced Raman scattering (SERS) spectra of serum samples were rapidly obtained on Ag plasmonic paper substrates upon 785 nm excitation. Spectra from the hepatocellular carcinoma (HCC) patients showed consistent differences with respect to those of the control group. In particular, uric acid was found to be relatively more abundant in patients, while hypoxanthine, ergothioneine, and glutathione were found as relatively more abundant in the control group. A repeated double cross-validation (RDCV) strategy was applied to optimize and validate principal component analysis-linear discriminant analysis (PCA-LDA) models. An analysis of the RDCV results indicated that a PCA-LDA model using up to the first four principal components has a good classification performance (average accuracy was 81%). The analysis also allowed confidence intervals to be calculated for the figures of merit, and the principal components used by the LDA to be interpreted in terms of metabolites, confirming that bands of uric acid, hypoxanthine, ergothioneine, and glutathione were indeed used by the PCA-LDA algorithm to classify the spectra.

The Sample, the Spectra and the Maths-The Critical Pillars in the Development of Robust and Sound Applications of Vibrational Spectroscopy

The last two decades have witnessed an increasing interest in the use of the so-called rapid analytical methods or high throughput techniques. Most of these applications reported the use of vibrational spectroscopy methods (near infrared (NIR), mid infrared (MIR), and Raman) in a wide range of samples (e.g., food ingredients and natural products). In these applications, the analytical method is integrated with a wide range of multivariate data analysis (MVA) techniques (e.g., pattern recognition, modelling techniques, calibration, etc.) to develop the target application. The availability of modern and inexpensive instrumentation together with the access to easy to use software is determining a steady growth in the number of uses of these technologies. This paper underlines and briefly discusses the three critical pillars-the sample (e.g., sampling, variability, etc.), the spectra and the mathematics (e.g., algorithms, pre-processing, data interpretation, etc.)-that support the development and implementation of vibrational spectroscopy applications.

Current applications of machine learning in the screening and diagnosis of glaucoma: a systematic review and Meta-analysis

AIM

To compare the effectiveness of two well described machine learning modalities, ocular coherence tomography (OCT) and fundal photography, in terms of diagnostic accuracy in the screening and diagnosis of glaucoma.

METHODS

A systematic search of Embase and PubMed databases was undertaken up to 1^st of February 2019. Articles were identified alongside their reference lists and relevant studies were aggregated. A Meta-analysis of diagnostic accuracy in terms of area under the receiver operating curve (AUROC) was performed. For the studies which did not report an AUROC, reported sensitivity and specificity values were combined to create a summary ROC curve which was included in the Meta-analysis.

RESULTS

A total of 23 studies were deemed suitable for inclusion in the Meta-analysis. This included 10 papers from the OCT cohort and 13 from the fundal photos cohort. Random effects Meta-analysis gave a pooled AUROC of 0.957 (95%CI=0.917 to 0.997) for fundal photos and 0.923 (95%CI=0.889 to 0.957) for the OCT cohort. The slightly higher accuracy of fundal photos methods is likely attributable to the much larger database of images used to train the models (59 788 vs 1743).

CONCLUSION

No demonstrable difference is shown between the diagnostic accuracy of the two modalities. The ease of access and lower cost associated with fundal photo acquisition make that the more appealing option in terms of screening on a global scale, however further studies need to be undertaken, owing largely to the poor study quality associated with the fundal photography cohort.

A Novel Artificial Intelligence Technique to Estimate the Gross Calorific Value of Coal Based on Meta-Heuristic and Support Vector Regression Algorithms

Gross calorific value (GCV) is one of the essential parameters for evaluating coal quality. Therefore, accurate GCV prediction is one of the primary ways to improve heating value as well as coal production. A novel evolutionary-based predictive system was proposed in this study for predicting GCV with high accuracy, namely the particle swarm optimization (PSO)-support vector regression (SVR) model. It was developed based on the SVR and PSO algorithms. Three different kernel functions were employed to establish the PSO-SVR models, including radial basis function, linear, and polynomial functions. Besides, three benchmark machine learning models including classification and regression trees (CART), multiple linear regression (MLR), and principle component analysis (PCA) were also developed to estimate GCV and then compared with the proposed PSO-SVR model; 2583 coal samples were used to analyze the proximate components and GCV for this study. Then, they were used to develop the mentioned models as well as check their performance in experimental results. Root-mean-squared error (RMSE), correlation coefficient (R2), ranking, and intensity color criteria were used and computed to evaluate the GCV predictive models developed. The results revealed that the proposed PSO-SVR model with radial basis function had better accuracy than the other models. The PSO algorithm was optimized in the SVR model with high efficiency. These should be used as a supporting tool in practical engineering to determine the heating value of coal seams in complex geological conditions.

Discriminant analysis and feature selection in mass spectrometry imaging using constrained repeated random sampling - Cross validation (CORRS-CV)

The identification of biomarkers through Mass spectrometry imaging (MSI) is gaining popularity in the clinical field. However, considering the complexity of spectral and spatial variables faced, data mining of the hyperspectral images can be troublesome. The discovery of markers generally depends on the creation of classification models which should be validated to ensure the statistical significance of the discriminants m/z detected. Internal validation using resampling methods such as cross validation (CV) are widely used for model selection, the estimation of its generalization performance and biomarker discovery when sample sizes are limited and an independent test set is not available. Here, we introduce for first time the use of Constrained Repeated Random Subsampling CV (CORRS-CV) on multi-images for the validation of classification models on MSI. Although several aspects must be taken into account (e.g. image size, CORRS-CV∂value, the similarity across spatially close pixels, the total computation time), CORRS-CV provides more accurate estimates of the model performance than k-fold CV using of biological replicates to define the data split when the number of biological replicates is scarce and holding images back for testing is a waste of valuable information. Besides, the combined use of CORRS-CV and rank products increases the robustness of the selection of discriminant features as candidate biomarkers which is an important issue due to the increased biological, environmental and technical variabilities when analysing multiple images, especially from human tissues collected in clinical studies.

Fluorescent electronic tongue based on soluble conjugated polymeric nanoparticles for the discrimination of heavy metal ions in aqueous solution

A fluorescence sensing array (or fluorescent electronic tongue) based on six sorts of soluble conjugated polymeric nanoparticles (SCPNs) decorated with PEG chains is designed for the rapid identification of heavy metal ions in water.