Tutorial for the use of ILAA

Handling multicollinearity with ILAA. Impact on Linear and Logistic Regression Modeling are provided #Multicollinearity, #Decorrelation, #LinearTransformations #ERT

Evaluation of the 70 Gene Signature based on the data from the "vanDeVijver" validation dataset.

How to use the FRESA.CAD RRPlot() function to validate and evaluate Survival Biomarkers

Sample code for the use of RRPlots for the evaluation of Survival Models

Testing the feature decorrelation algorithm on the DARWIN data set

Demo of the use of Feature Decorrelation on the multiple Feature Data set.

Testing the Feature Decorrelation Algorithm, the Filtered Fit on the Arcene Data Set.

A demo of the use of filteredfit() to train and validate Machine Learning models with data decorrelation

A demo showcasing options to compute the decorrelation matrices on the Vehicle data set.

The Discovery of Features associated with an outcome is enhanced by data transformations that simplify the machine learning process

Filtered Fit GDSTM on Arcene Data Set

Tutorial of Filteredfit() function with decorrelation

Full description of the GDSTMDecorrelation function of FRESA.CAD

Sample use of GDSTMDecorrelation

Demo of FRESA.CAD::GDSTMDecorrelation()

Vignette of Cross-Validation using FRESA.CAD

Testing FRESA.CAD 3.3.0 f

Hospitalizaciones y Defunciones de adultos Mayores (>60) en México por COVID-19

Radiomics and Melanoma Diagnosis

Testing Feature Decorrelation on the Parkisons Data Set

Testing the effect of feature decorrelation on the ARCENE daata set

Here is an example of the options of the targeted feature decorrelation function in FRESA.CAD.

Multidimensional data decorrelation. Here I show the advantage of non-ortogonal decorrelation of multidimensional signals. Top basis vectors are sequentially drawn from the data matrix, features are decorrelated creating a new set of non-correlated features.

Classifier ML methods in FRESA.CAD. Various ML methods, with filters and feature decorrelation

Impact of Feature Decorrelation on ML methods

Running all the FRESA.CAD modeling functions and filter for Binary Outcomes BSWiMS,KNN,NB,PCA,Signature,LASSO,BESS, etc...

Mexico COVID

Country Peaks Estimations of COVID-19

Predictions of New Hospitalizations and Fatalities in Mexico

Estimating the probability of Discharge from Hospital

Forecasting the expected % of deaths due to COVID-19 I'm not an epidemiologist so don´t believe 100% of the estimations. Models where updated 4/8/2020 with 4/7/2020 data from Kaggle

Ensemble Model of Logit functions for the diagnosis of COVID-19. The model is not valid. It is just a proof of concept of ML for the diagnosis of COVID-19 based on self-reported symptoms and diagnosis

Estimating the probability of Hospital Discharge based on Patient Symptoms

Models and features that predict death from COVID-19. Diabetes, hypertension, and dyspnea.

Benchmarking Caret-optimized classifiers with FRESA.CAD

Testing Boosting with FRESA.CAD ADA boost, Gradient Boost, RF, LDA, QDA, KNN, KNN-BSWiMS

Gradient Boost, Ada Boost, KNN, Random Forest, LASSO, SVM, LDA, QDA, Logistic Regression, RPART. Blanced, Proportional, Boostraping Cross-Validation

FRESA.CAD Benchmarking the prostate cancer data set

Evaluation of Filters on a Logistic Regression Fit: IDI,NRI,Kendall, T-test,wilcoxon, etc....

Classifiers that have FRESA.CAD interface for random cross-validation

FRESA.CAD Benchmarking

Comparing several ML methods on the DIABETES dataset

The use of FRESA.CAD Cross-validation function and FRESA.CAD binary benchmark to compare ML methods.

Binary Classification Benchmarking with the ARCENE dataset

Comparing ADABOOST vs Other Common classifiers

Ada Boost and Hybrid Learning with FRESA.CAD benchmark

Benchmarking Classifiers

Genes associated with BRCA recurrence

FRESA.CAD benchmark on stagec prostate cancer data set

FRESA.CAD on Milles per Galon Dataset

FRESA.CAD benchmarking Classification models on the White Wine Dataset

Benchmarking Ordinal Algorithms using FRESA.CAD

Use of the benchmark function for the evaluation of classification algorithms.

Predicting KL Scores using quantitative JSW OAI measurements

Ordinal Modeling of the Low Birth Weight data set

KL Score modeling with FRESA.CAD

DEXA Body Fat estimation from anthropometric measurement

"Data obtained from http://biostat.mc.vanderbilt.edu/DataSets".

Benchmarking B:SWiMS, LASSO, SVM, RF

Benchmark of classifiers and Feature Selection: B:SWiMS, LASSO, Random Forest and SVM with FeaLect

Lymphoma Benchmark with B:SWiMS, SVM, RF and SVM

Prediction of ONCOTYPE DX from Radiomics Analysis of Mammograms

Predicting ADAS13 from MRI's Structural quantitative analysis

FRESA.CAD´s B:SWiMS approach is compared to SVM, RF and LASSO on the ARCENE data set

Colon Benchmark: BSWIMS,SVM,LASSO and RF

Wisconsin Breast Cancer Data Set analysis using FRESA.CAD Comparison between LASSO, Random Forest, Support Vector Machine (SVM), and the use of FeaLect for feature selection.

Benchmarking FRESA.CAD with the SONAR data set.

FRESA.CAD Madelon and SVM

Example of the use of FRESA.CAD to find and analyze models

FRESA:CAD on the ARCENE data set

Risk Models of Breast Cancer Recurrence using FRESA.CAD

Predicting Survival

BRCA recurrence with logistic models

Analysis of the prostate cancer DB

BSWIMS test on lymphoma data set

Prediction of KL scores and TKR using X-Ray Quantitative Image assessments Data form OAI

This document shows how to adjust longitudinal data to site bias and anthropometric differences

Scrip showing the use of FRESA.CAD for the creation and validation of breast cancer signatures