\documentclass[10pt]{amsart} \usepackage[left=.75in,right=.75in,top=.75in,bottom=.75in]{geometry} \geometry{letterpaper} % ... or a4paper or a5paper or ... %\geometry{landscape} % Activate for for rotated page geometry %\usepackage[parfill]{parskip} % Activate to begin paragraphs with an empty line rather than an indent \usepackage{color,graphicx} \usepackage{amssymb} \usepackage{amsmath} \usepackage{verbatim} %\DeclareGraphicsRule{.tif}{png}{.png}{`convert #1 `dirname #1`/`basename #1 .tif`.png} \newcommand{\eqn}[1]{(\ref{#1})} \newcommand{\picdir}{./pdffig} %\renewcommand{\labelenumi}{\arabic{enumi}. } %http://latex2rtf.sourceforge.net/latex2rtf_1_9_19.html#Conditional-Parsing %Starting with LaTeX2RTF 1.9.18, there is a handy method for %controlling which content should be processed by LaTeX or by %LaTeX2RTF . Control is achieved using the standard \if facility of %TeX. If you include the following line in the preamble of your document % % \newif\iflatextortf %Then you will create a new \iflatextortf command in LaTeX . TeX %sets the value of this to false by default. Now, LaTeX2RTF %internally sets \iflatextortf to be true, and to ensure that this %is always the case, LaTeX2RTF ignores the command %\latextortffalse. This means that you can control how different %applications process your document by % % \iflatextortf % This code is processed only by latex2rtf % \else % This code is processed only by latex % \fi %Note that \iflatextortf will only work within a section; you %cannot use this command to conditionally parse code that crosses %section boundaries. Also, it will only work on complete table or %figure environments. Due to the mechanism used by LaTeX2RTF in %processing these environments, at this time the only way to %conditionally parse tables and figures is to include two complete %versions of the environment in question, nested within an %appropriate \iflatextortf structure. % \newif\iflatextortf \iflatextortf %do nothing \else %pdflatex \usepackage{boxedminipage,float} \usepackage{wrapfig,setspace} \usepackage[pdftex, plainpages=false, colorlinks=true, citecolor=black, filecolor=black, linkcolor=black, urlcolor=black]{hyperref} %% textureanat_ATROPOS_GMM.png %% textureanat_ATROPOS_GMM_VOLUME_TO_SURFACE_AREA_RATIO.png %% textureanat_DENOISE.png %% textureanat_ATROPOS_GMM_CANNY.png %% textureanat_ClusterProminence_1.png textureanat_ClusterProminence_5.png %% textureanat_Energy_1.png textureanat_Energy_5.png %% textureanat_Inertia_1.png textureanat_Inertia_5.png %% textureanat_ClusterShade_1.png textureanat_ClusterShade_5.png %% textureanat_Entropy_1.png textureanat_Entropy_5.png %% textureanat_InverseDifferenceMoment_1.png textureanat_InverseDifferenceMoment_5.png %% textureanat_Correlation_1.png textureanat_Correlation_5.png %% textureanat_HaralickCorrelation_1.png textureanat_HaralickCorrelation_5.png %% %% %% %% %% %% %% \title{Image Feature Data Service} \author{ D.~Fuentes\textsuperscript{1} } \date{ \small The University of Texas M.D. Anderson Cancer Center,\\ Departments of \textsuperscript{1}Imaging Physics, \textsuperscript{2}Diagnostic Radiology, %\textsuperscript{3}Gastrointenstinal Oncology, %and \textsuperscript{4}Biostatistics, Houston TX 77030, USA \\ %Email: \texttt{jhazle@mdanderson.org} \\ Received: date / Accepted: date % Webpage: \texttt{http://wiki.ices.utexas.edu/dddas} } \begin{document} \maketitle \begin{itemize} \item Hypothesis: Volumetric RECIST\cite{tacher2015comparison} provides more reliable response to therapy than RECIST 1.x \begin{itemize} \item 'more reliable' = earlier response ? \item 'more reliable' = more accurate ? \item How do we quantify this ? \end{itemize} \end{itemize} \begin{figure}[h] \begin{minipage}{0.7\textwidth} \includegraphics[width=1\textwidth]{\picdir/timeline.png} \\ \includegraphics[width=1\textwidth]{\picdir/Kaplan.png} \end{minipage}~\begin{minipage}{0.39\textwidth} \includegraphics[width=1\textwidth]{\picdir/ResponseCriteria.png} \end{minipage} \caption{Typical treatment timeline, Response Criteria, and Kaplan-Meier Analysis~\cite{tacher2015comparison}. KM analysis demonstrates that volumes statistically stratify `responders' from `non-responders'. }\label{Fig:Kaplan} \end{figure} \begin{itemize} \item {\color{red} How do we show impact on patient care? } How do we show that early identification of non-response will improve disease-specific outcomes and most important, patient outcomes? \item All volumes are generated from segmentations. What is convincing evidence that segmentations are accurate? Comparison to pathology? \end{itemize} \clearpage \pagebreak \begin{itemize} \item Key Personnel to streamline workflow \begin{itemize} \item database programmer: build rules for automated data retrieval (???) \item research assistant: data curation of survival times and imaging qa (MD) \item research programmer: data analysis (PhD) \item statistician: (auston) \end{itemize} \end{itemize} \begin{figure}[h] \includegraphics[width=1\textwidth]{\picdir/ManualAuto} \caption{ Step 1. Show this with manual segmentations to demonstrate an impact. Step 2. Investigate if image features also stratify a survival difference. Step 3. Automate the segmentation process to improve throughput. }\label{Fig:ManualAuto} \end{figure} \begin{thebibliography}{1} \bibitem{tacher2015comparison} Vania Tacher, MingDe Lin, Rafael Duran, Hooman Yarmohammadi, Howard Lee, Julius Chapiro, Michael Chao, Zhijun Wang, Constantine Frangakis, Jae~Ho Sohn, et~al. \newblock Comparison of existing response criteria in patients with hepatocellular carcinoma treated with transarterial chemoembolization using a 3d quantitative approach. \newblock {\em Radiology}, 278(1):275--284, 2015. \end{thebibliography} \clearpage \pagebreak \begin{figure}[h] \IfFileExists{workdir/StatSummary.pdf}{\includegraphics[width=.8\textwidth]{workdir/StatSummary.pdf} }{\fbox{\begin{picture}(10,10)\put(0,5){ not found } \end{picture}}} \IfFileExists{workdir/TopPredictors.csv}{\verbatiminput{workdir/TopPredictors.csv}}{predictors not found} \caption{Group Summary}\label{fig:GroupSummary} \end{figure} \clearpage \pagebreak \section{Usage} \begin{figure}[h] \includegraphics[width=1\textwidth]{\picdir/DataFlow.png} \caption{ Data Service Workflow. Medically trained user transfers studies from PACS to image processing dicom server for manual labeling, data curating, and building a training database. GPFS file system is shared between the image processing server and eagle. Analysis-ready data is transferred to HPC server to texture features and image feature generation. Ideally, MetaEngine should manage all queries, processing, dataflow, and collect all results for visualization and interpretation. }\label{Fig:DataFlow} \end{figure} \begin{itemize} \item Currently setup on IPVL5. TODO - move to eagle \item Use WinSCP (\href{https://winscp.net/}{https://winscp.net/}) or CyberDuck (\href{https://cyberduck.io/}{https://cyberduck.io/}) to upload image sets to: \begin{itemize} \item \texttt{igilab@192.168.24.101:/workarea/igilab/github/MelanomaImageFeatures/DataDirectory/} \end{itemize} \item The corresponding processed images will appear at: \begin{itemize} \item \href{http://192.168.24.101/igilab/github/MelanomaImageFeatures/Processed/}{http://192.168.24.101/igilab/github/MelanomaImageFeatures/Processed/} \end{itemize} \end{itemize} Each image set should be in a separate directory and should follow the below naming convention \textit{exactly}: \begin{verbatim} $ ls DataDirectory/MRN/MM.DD.YYYY/ Truth.nii.gz Ven.nii.gz \end{verbatim} \pagebreak {\small \begin{verbatim} Dependencies ============ http://otbcb.readthedocs.org/en/latest/index.html $ module show otb/4.2.0 ------------------------------------------------------------------- /usr/local/Modules/3.2.10/modulefiles/otb/4.2.0: module-whatis set environment variables to build OTB OTB-4.2.0 module-whatis simple build module-whatis cmake -DCMAKE_CXX_COMPILER=/usr/bin/g++-4.4 -DCMAKE_C_COMPILER=/usr/bin/gcc-4.4 -DCMAKE_BUILD_TYPE=Debug -DBUILD_SHARED_LIBS=Off -DOTB_USE_EXTERNAL_LIBKML=ON -DBUILD_APPLICATIONS=OFF -DBUILD_EXAMPLES=OFF -DBUILD_TESTING=OFF -DCMAKE_VERBOSE_MAKEFILE=ON -DCMAKE_INSTALL_PREFIX=$OTB_HOME ../OTB-4.2.0 setenv OTB_VERSION OTB-4.2.0 setenv OTB_SOURCE /opt/apps/OTB/OTB-4.2.0 setenv OTB_HOME /opt/apps/OTB/OTB-4.2.0-gcc-4.4.7-dbg setenv OTB_DIR /opt/apps/OTB/OTB-4.2.0-gcc-4.4.7-dbg/lib/otb ------------------------------------------------------------------- http://sourceforge.net/p/advants/kaena/ci/master/tree/ $ module help ants/dev ----------- Module Specific Help for 'ants/dev' ------------------- This module loads ANTsR dev ######### need latest version of R, edit sources.list $ tail /etc/apt/sources.list deb http://cran.r-project.org/bin/linux/ubuntu/ precise/ local({r <- getOption('repos'); r['CRAN'] <- 'http://cran.r-project.org'; options(repos=r)}) install.packages('Rcpp',type='source') mypkg<-c('signal','timeSeries','mFilter','MASS','robust','magic','knitr','pixmap','rgl','misc3d') for ( x in mypkg ) { install.packages(x) } sudo apt-get update sudo apt-get install r-base sudo apt-get install r-base-dev ######### install from github wget https://raw.github.com/stnava/RMI/master/stnava/install_anstr_packages.sh ./install_anstr_packages.sh /opt/apps/ANTsR/dev/ 0 1 ######### edit cmake file to use dbg mode ?? /opt/apps/ANTsR/dev//ANTsR_src/ANTsR/src/CMakeLists.txt diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt index c4b1c0c..b88126c 100644 --- a/src/CMakeLists.txt +++ b/src/CMakeLists.txt @@ -20,6 +20,7 @@ ExternalProject_Add( ANTS -D COPY_SCRIPT_FILES_TO_BIN_DIR=ON # for useful things like buildtemplateparallel -D BUILD_SHARED_LIBS=OFF # R requires shared objects -D BUILD_TESTING=OFF # reduces build time -D BUILD=DEBUG \end{verbatim} } \end{document}