cgplab / radar Goto Github PK

A tool to perform comprehensive analysis of high-dimensional radiomic datasets

License: MIT License

R 100.00%

radiomics radiomics-analysis data-analysis imaging imaging-genomics

radar's Introduction

RadAR

Radiomics Analysis with R (RadAR) is a package for R to perform comprehensive analysis of high-dimensional radiomic datasets.

Introduction

Quantitative analysis of biomedical images, referred to as radiomics, is emerging as a promising approach to facilitate clinical decisions and improve patient stratification. The typical radiomic workflow includes image acquisition, segmentation, feature extraction, and analysis of high-dimensional datasets. While procedures for primary radiomic analyses have been established in recent years, processing the resulting radiomic datasets remains a challenge due to the lack of specific tools for doing so.

Here we present RadAR (Radiomics Analysis with R), a new software to perform comprehensive analysis of radiomic features. RadAR allows users to process radiomic datasets in their entirety, from data import to feature processing and visualization, and implements multiple statistical methods for analysis of these data. We used RadAR to analyse the radiomic profiles of more than 850 cancer patients from publicly available datasets and showed that it was able to recapitulate expected results. These results demonstrate RadAR as a reliable and valuable tool for the radiomics community.

Installation

RadAR is freely available under MIT license at https://github.com/cgplab/RadAR. First, install biocViews to facilitate the installation of the package dependecies:

if (!requireNamespace("BiocManager", quietly = TRUE))
    install.packages("BiocManager")

BiocManager::install("biocViews")

Then, you can install RadAR from github with:

# install.packages("devtools")
devtools::install_github("cgplab/RadAR")

To include vignettes use:

# install.packages("devtools")
devtools::install_github("cgplab/RadAR", build_vignettes = T)

Usage

To show how to use RadAR for the analysis of radiomic datasets, a step-by-step tutorial is included in the package.

Citation

Benelli M, Barucci A, Zoppetti N, Calusi S, Redapi L, Della Gala G, Piffer S, Bernardi L, Fusi F, Pallotta S. Comprehensive analysis of radiomic datasets by RadAR. Cancer Research 80 (15), 3170-3174. https://cancerres.aacrjournals.org/content/80/15/3170

radar's People

Contributors

Stargazers

Watchers

Forkers

sands58 escri11 guangzhaohui jimmyye1 romagnolid willmika

radar's Issues

Error in plot_features

I have got another error as follow: Error in boxplot.default(split(mf[[response]], mf[-response], drop = drop, :
invalid first argument

The features have been extracted by Pyradiomics. I have the same error when putting the software example (LungExample) in the code.
Please advice.

ERROR: dependency 'mRMRe' is not available for package 'RadAR'

Package mRMRe has recently been removed from CRAN. Not sure if this has to do with a long-standing bug where mRMRe may return erroneous results for large dataset. Bummer.

Error in viewing correlation matrix of features by heatmap

Hi,
Thanks for your comprehensive package for radiomics feature analysis.
I have a problem when I want to view the correlation matrix of features by heatmap. Below you can see the error, although there is no NaN or inf or 0 value in my extracted features by Pyradiomics.
Please advice.
Error in hclust(d, method = method) :
NA/NaN/Inf in foreign function call (arg 10)

label in hierarchical cluster

Hi again,

I cannot see a proper label in this plot

plot_heatmap_hcl(rdr = rdr, annotation_tracks = c("type"))

Could you, please, tell me how to get type or any other label rather than this long label

Issue about drawing a clusterized heat map

When I set only one clinical factor (e.g.,"Overall.Stage") as annotation_tracks, its name is not correctly displayed on a heat map.
Could you teach me how to solve this?

Below is the command I tried.
plot_heatmap_hcl(rdr = rdr,
annotation_tracks = c("Overall.Stage"))

argument function import_radiomic_table()

Just I minor typo ... probably due to copy & paste

In the vignette you wrote

rdr <- import_radiomic_table(dir = "PATH/TO/RADIOMIC_TABLE")

and the argument for this function is file instead of dir

import_radiomic_table <- function (file = NULL)

Juan

PS\ One of my PhD is going to do a PR for another issue you have with boxplots

problem with pre-processing (normalization)

I got this error:

> library(RadAR)
> lung1_rdr <- normalize_feature_values(rdr = lung1_rdr, 
+                                 which_data = "scaled")
Error in t.default(data) : argument is not a matrix

Do you know what I am doing wrong?

adding RIA as another input

Hi,

first of all I would like to congratulate you for the excellent R library you have created! it is really useful.

I was wondering whether you could create a set of function for RIA (https://cran.r-project.org/web/packages/RIA/index.html) since it is another really nice R package to compute radiomic features like pyradiomics, 3Dslicer or LifeX. The advantadge would be that if so, one could integrate everything in a single R pipeline.

I know that I can use import_radiomic_table() but it would be nice to integrate RIA into RadAR as you did with pyradiomics.

Thanks
Juan

How to perform multi-series MRI-omics analysis

Hi，
Recently, I'm trying to analyze T1, T2, T1enhance...etc MRI sequences and their radiomics features，which means one case has at least 3 or 4 csv-files of the same features' name(1 csv file /each MRI sequence, gernerated by Pyradiomics)，so I manually added a suffix like '_T1_ROI', '_T2_ROI'（using python). After I fused 2 csv-file (for example, T1 and T2 's features ) into 1 csv-file，the program failed to load that csv-file.

Here is my Question:
I think I should formatting the csv-file first, but (Question one) is it possible for RadAR to analyze such kind of complex datasets(muti-sequences MRIomics features)?

And even successfully loaded 1 csv, it seems that I dont have the 'Normailized' and 'Scale' data. (Question two) How to gernerate those kind of data in RadAR and how those data store in the Rdata?

(Question three) how to format my csv-file ?(RadAR cant load that csv, i already have 'feature_table' kind of csv)