Medical Image Registration - Science topic

Hello, I am currently experimenting on medical image registration and I would like to do a baseline registration between a CT and an MRI scan of the liver of a patient for selective internal radiation therapy dosimetry. I would like to know if there is any data available online that I could use.

is there any publicly available dataset for conventional coronary angiography images?

Hi,

Is there any longitudinal datasets (CTscan or MRI, 2D or 3D) of tumors publicly available ? I'm doing a project where I to apply deep learning to see the evolution of lesions/tumors over time for the same patient and I'm struggling to find relevant datasets.

During preprocessing medical image data different techniques should be considered such as cropping, filtering, masking, augmentation. My query is, which techniques are frequently applied to medical image datasets during pre-processing?

I am a Master student in Biomedical (Bio-electrical) Engineering and work on " 2D-3D registration ( CT and X-Ray) as my thesis.

I created DRR from CT and want use CNN( Convolutional Neural Network)for estimate and train 6 transformation parameters of registration between DRR and X-Ray images.

I would be grateful if you could give me some information or code in MATLAB.

Thank you for your time.

Hi, I have some DTI data and I want to use ANTs to make a b0 group wise template to normalize my data to the IIT_mean_b0 image. I want to use buildtemplateparallel.sh and SyN algorithm ,but I'm the amateur one. How am I supposed to run this and what are the inputs?

Now I'm using python to do some image registration,but I found there is no useful tool for me.So I have to do some basic jobs and implement some basic algorithms by myself.It is so slowly and a little bit difficult.I wonder there are some useful toolkit to help me do image registration in python.Curiously I want to know which language the researcher use to do the image registration.Matlab? or others?

I just finished a program about Maximization Mutual Information in registration using python,but it seems very slowly,and a little bit wrong.

If you knew how to do image registration,including what toolkit I should use,which language is much better,which toolkit in python I can use.please told me.Thank you very much!!! Forgive me weak English.

I know there are some functions or methods to do image registration using Matlab.But they seem to be abstract, I did not find some underlying functions.I believe I can not do some modifications if I want to improve some algorithms.

I am interested in obtaining a spectral CT dataset to test a reconstruction algorithm. 

Thanks. 

I am studying 2D and 3D registrations. I need help and guidance in producing DRR from CT images. Is there an article or program that fully explains how to generate DRR from CT images? .

I want to use CNN networks for registration whose input is a fluoroscopic and DRR images.

thanks

Hello!

I am working on segmenting cells from Fine Needle Aspiration (FNA) images of breast. Size and shape features of the cells segmented out of the microscopic sample image shall be determined. I intend to classify the sample using my algorithm based on these feature values. I am facing problem in finding a database that contains sufficient FNA images for breast cancer. If anybody have experience working in this area, please share. Furthermore what segmentation techniques can be used for this purpose.

Dr. Arbab Masood Ahmad

I need a public benchmark MRI Prostate dataset for image registration/fusion. The dataset should be segmented with multiple segments per image i.e

1)Prostate gland

2)Peripheral zone

2)Central zone

3)Suspected Lesions

Thanks

I have downloaded BRATS 2015 training data set inc. ground truth for my project of Brain tumor segmentation in MRI. A file in .mha format contains T1C, T2 modalities with the OT. Please suggest how to access these files in MathWorks (MATLAB) and further how to proceed for segmentation procedure?

The challenge is based on the publication by Collins et al. (2017) appearing in the IEEE Transactions on Medical Imaging (vol. 36, no. 7, pp. 1502-1510) and is described by the SPIE Medical Imaging 2019 Proceedings paper please cite both of these if using data within a paper). To describe briefly, there are surgical workflow advantages if one could align preoperative liver image volume data to its intraoperative physical counterpart using sparse surface data visible during the procedure at presentation. We have developed a novel human-to-phantom framework that allows us to transpose real operating room (OR) data patterns that we acquired clinically using an optically tracked stylus onto a quantitative deforming phantom environment. This framework allows the development and testing of image-to-physical registration algorithms in the presence of deformation with quantitative subsurface targets for assessing error and within the context of realistic OR data acquisition. We note that the deformations we have imposed on the phantom mimic patterns of deformation we have seen in the OR. Specifically, the presentation of the organ allows the anterior surface of the organ to be visible at various levels of extent, and deformations are associated with the surgical packing of the organ on the posterior side of the organ. For this challenge, these states can be assumed. As part of this new challenge, we have developed a new phantom and more data patterns not previously used. We also have many more subsurface targets for characterization than in previous work (n=159 targets). In order to formerly enter your result for the challenge and have it posted on the Final Result Dashboard as complete, results need to be provided for all 112 data sets. You do not need to submit results to all data sets to interact with the challenge. The Dashboard will also track partial submissions, i.e. you can provide subsets for analysis while you are developing and these results will be provided on the Dashboard. Only latest results will be retained. *** Also, new to the challenge, on the data site, we do have sample results among the data set available so that you can check your algorithms before submitting to the dashboard.***

The challenge was officially released in February of 2019. With respect to official end-dates to the challenge, the intent is to leave this available to the community for an extended period of time. With sufficient participation, a review and analysis will be forthcoming.

Hello,

Already we have some similarity metric such as MI, NCC, CR, etc, As needed for task purposes.

I want to know does anyone know what the best or newest similarity metric is to registering two multi modality images??

I know that in List mode data the scintillating events are stored along with the time stamps of the event. I have a doubt regarding image reconstruction from the list mode data.

Is each (x,y coordinate of) photon count binned into an image matrix or are the photon counts integrated in certain time interval and treated with Anger algorithm or MLEM algorithm then binned into a image matrix? If the latter is the case then what is the value of the time interval?

If my question is not clear or wrong can anybody suggest some sources which will help me in understanding this ?

Dear colleagues,

We are pleased to announce the 9th International Workshop on Biomedical Image Registration, WBIR2020, hosted in Portorož, Slovenia! The workshop will be held in the Congress Centre Bernardin in Portorož, on 16 and 17 June, 2020.

The workshop brings together leading researchers in the area of biomedical image registration to present and discuss recent developments in the field, including methodological innovations and advances in the performance and validation on existing and novel applications. The workshop will include both oral and poster presentations, exciting keynote lectures, all with ample opportunities for discussion. At the social events you will enjoy the warm and relaxing Adriatic seaside along with authentic Mediterranean cuisine and excellent drinks.

WEBSITE https://wbir2020.org

IMPORTANT DATES Paper submission deadline: Jan 10, 2020 Notification of acceptance: Feb 21, 2020 Camera-ready deadline: March 20, 2020 Conference dates: June 16 and 17, 2020

AIMS AND SCOPE Submissions are invited in all areas of biomedical image registration. Topics of interest include, but are not limited to:

- Novel registration methodology: 2D/3D/4D, spatiotemporal/dynamic, pairwise / groupwise, slice-to-volume, projective, single/multi-modal, intra/inter-subject, model-based, patch-based, multi-channel, tracking

- Mathematical aspects of image registration: continuous/discrete optimization, real- time, similarity measures, diffeomorphisms, LDDMM, stationary velocity, inverse consistency, multi-scale

- Machine learning and deep learning techniques for registration: unsupervised / supervised / reinforcement learning, convolutional / recurrent / transformer networks, neural networks for feature extraction and matching, correspondence weighting and prediction, attention modeling, deformation learning, deep encoder- decoder networks

- Biomedical applications of registration: computer-assisted interventions, image- guided therapy, treatment planning/delivery, diagnosis/prognosis, atlas-based segmentation, label fusion, histopathology correlation, serial studies, pathology detection and localization, morphometry, biomechanics, image retrieval/restoration/fusion, imaging biomarkers for precision medicine, radiomics & radiogenomics, early proofs of concept

- Validation of registration: quantitative and qualitative methods, benchmarking, comparison studies, phantom studies, correlation to outcome, validation protocols and performance metrics, uncertainty estimation

All accepted full paper submissions will be published as a volume in the Springer's Lecture Notes in Computer Science (LNCS) series.

ORGANIZING COMMITTEE Ziga Spiclin, University of Ljubljana, Ljubljana, Slovenia Jamie McClelland, University College London, London, UK Jan Kybic, Czech Technical University in Prague, Prague, Czech Republic Orcun Goksel, ETH Zurich, Zurich, Switzerland

SPONSORS The WBIR 2020 is a MICCAI Society Endorsed Event (www.miccai.org).

I am working on DWI images of GBM patients and I want to subtract ADC values of post treatment from ADC valuse of pre reatment images. I did not use any marker, how can I find the same slices for registration in two image sets?

Best Regards

Hi all,

I'm interested in X-y location of a point on an image. If I interpolate the image to be 1000 by 1000 pixels, then downsample it to 200 *200 pixels. Then measure X-Y location of the point. Do the interpolation and downsampling generate an artifact that might affect the accurate x-y position of the point?

Why do we need to down sample the histology image when we try the registration to a medical image?. one reason that I know is to reduce the noise. Also, to help the similarity metrics to find a corresponding points then robust the registration. I would know if there is more significant reasons.

Thanks,

I'm facing a difficulty in evaluating the registration accuracy by using Elastix, since I'm using an artificial images.

I would like to have an advice on how to evaluate the accuracy of the registration process.

Thanks

A detailed study of medical image registration technique with application and validation. I am looking for suggestion to select a >1.4 impact factored and relatively fast review journal to send my manuscript.

Thanks

We are working with lung images from LIDC available through The Cancer Imaging Archive.  The XML file provides the ground truth data.  Is there any possibility to generate ground truth images from the available data.

In the development of a Co-registration method to compare two 3D MRI exams (before and post chemotherapy treatment for one patient using the same MRI modality) ==>you can see the problematic on the image uploaded enclose<==. The results show a correct alignment at the visual level. However, this is surely not enough. The first thing I have to think about is to validate the findings by comparing the anatomical points of interest (Landmarks). Are there any more practical propositions?

I thank the community in advance.

Dear guys,

I'm looking for a data set containing both CT and MR images of Head:

* Task description: Implementing INTRA-subject CT-MR volumetric image registration.

* Data requirement: the field of view MUST include mouth regions (i.e.,  jaw, tongue, mandible, etc).

Thanks in advance for your suggestions and helps,

Fetal movement and nonrigid deformation of uterine organs are severe artifacts deteriorating in utero magnetic resonance imaging (MRI) of the moving fetus. Many image registration tools have been developed to correct motion artifacts in brain MRI, such as FSL, AFNI, ANTs, IRTK, Elastix, and so forth. Most of them are similar each other since they are based on common theories of image registration. Moreover, advances have been made to cope with the severe motion artifacts in MRI of the moving fetus during decades. I already have used some of existing tools for image registration in functional MRI of the fetus. But, has anyone quantitatively compared those tools for in utero functional MRI (for example, of the placenta and fetal brain)? Which image registration tool would you recommend for in vivo functional MRI of the fetal brain?

I am working in image processing in the area of brain cancers. For evaluating my diagnosis algorithm I need a T2-weight MRI tumor brain database. if someone can help me in this way, please contact me.

The noise properties of the CBCT images depends directly on the voxel dimension, so comparing CNR obtained on different systems without a proper scaling is not correct.

Reconstructing the same set of projection with different voxel size will allow to have an experimental relation between the noise and the voxel size itself, but without the possibility to recover the raw projections (system installed in clinical enviroment) this experimental approach is not possible.

Is there a teoretical method for that sacling/normalization calculation? The type of reconstruction (FDK or Iterative) will influence that normalization process?

Please provide me some useful materials to have a basic understanding on any Image Registration based application

Previously, I manage to develop 3-D model of the lower lumbar spine from the MR image segmentation based on the full-color image slices from Visible Human Project. I believe the image was obtained from a cadaver in the supine position. I was wondering if there are any available resources (similar to the one offered by Visible Human Project) which are in extension and flexion state? 

Rigid registration, although trivial, is often used as a pre-registration step in Medical Image Registration or Remote Sensing. I would like to know what are the current best methods with respect to :

i) computation time and

ii) precision.

Thank you in advance for your help.

I am looking for various approaches for cancer segmentation and classification and was recently looking for Active Shape Model. The machine learning part of ASM is mainly based on Principal Component Analysis. I was looking for the feasibility of mixing ASM with deep learning? Can that be a feasible approach for segmentation pertaining to cancerous regions?

I urgently need any material that explains the general and basic standard for calculating lesion profile- the extent of vacuolation or spongiform degeneration.

is image geo-referencing and image registration is synonymous?

Local optima one of the optimization problems. while I am reading a MIND: Modality independent neighbourhood descriptor for multi-modal deformable registration paper I found the authors say: 

"The main disadvantage is that MI is intrinsically a global measure and therefore its local estimation is difficult, which can lead to many false local optima in non-rigid registration.The main disadvantage is that MI is intrinsically a global measure and therefore its local estimation is difficult, which can lead to many false local optima in non-rigid registration."

I'm looking for MRI images prior to artifact corrections. Preferably the scans from different body areas.

I am applying OTSU segmentation algorithm in order to segment the skin lesions, so for some of the lesion images I am getting optimal segmentation results but for most of the images I am not getting segmentation properly.

for example-

1.Ideal Segmentation Result-

The first image shows original lesion and second image i.e. Sampe1seg shows its segmentation result which is ideal.

2.Not Ideal segmentation-

The third image is the original image of lesion and forth image shows the poor result of segmentation.

What do you think about normalization patients with tumors? I analyzed each individual patient usually and I didn't do normalization.  How best to proceed?

I would like to do analysis of ROI in different planes by using DPARSF program.

what are the new methods or techniques used for registration of MRI CT images? What are the different methods i can try by using image processing? I am focusing on accuracy, low computational cost and Speed. Please help me.

I did my research in mathematics and i applied it for boundary of a medical image. I sent my paper to journal in computer vision and reviewer wrote that "A possible comparison would be against image registration methods, since the distorted image and boundary, after registration with the original image, would provide the original boundary". any body knows about image data in image registration?what does mean this commends?thanks n regards

i used matlab toolbox which use intensity based registration,

the image that i want to register to the reference image doesn't have good intensity, and makes algorithm not to work properly.

here is my two images

I am registered two images. For that, I stored images in matrix. Then I found inverse of one matrix which multiply with 1st image matrix. For multiplicaiton, first matrix's column must same with second matrix row. and it is not possible in different size of image. Can you solve this?

Has anyone previously used K-nearest neighbour search method to verify co-registration of MRI Images? Greatly appreciate any insight into how this method can be applied for exactly that purpose.

I'm Interested in doing image registration of two mri brain tumor images, hence i need the tumor images i.e. in the initial stage and the extended/later stage of the same patient. so that i can determine the prognosis. Any relevant sites?? thanks

I'm Working on detection and classification of MRI brain tumor images, hence for classification i need the data sets of benign and malignant mri brain tumor images. Thanks

and how to evaluate using RIRE project?

I have developed my multi-modal image registration for brain images. What evaluation methods are the best for my application?

I came across to RIRE project (http://www.insight-journal.org/rire/information.php). I read some related articles, but it is a bit confusing for me and I do not know how to evaluate my method using it. I would be thankful if anyone could help me.

what are the new trends for medical image classification. In other words, what are the current works on medical images ?

Are there any standard medical dataset?

Additionally, what are good feature selection methods for such type of datasets?

Thank you.

i am doing research on texture and shape feature for brain tumor images. so i need a latest techniques used in texture and shape features.

Contrast agent for CEUS: SonoVue

1,5 Tesla MRI

Dear sir

I converted MRI dicom files to nii  file using  dcm2niigui.

Previously i used same program for my data, then i got one nii images.

But this time, i got 3 nii images. 

one image name is sample.nii (22,529KB), the other is osample.nii(22,529KB) and the last one's name is cosample.nii (13922 KB).

I wonder what kind of image do i use for my further analysis.

If you have some idea, please help me.

Hello everyone, I'm working on a computer vision project in which I have to do segmentation of cervical spine vertebras and after that I have to track these segmented Vertebras in the whole fluoroscopic video sequence. I am done with the segmentation part and now I want to track those vertebras, any suggestions for tracking algorithm? video file is attached here.

Is there a well established methodology to register two different 3D microscopy images ?

I'm currently working with confocal 3D+Time acquisition in vivo, but I need to find a way to register the stacks from different time frames. For 2D registration we have lots of libraries,  but register  each 2D layer independently is out of question, because of movement in Z direction.

Any clues ?

Best regards

I have some DICOM images for testing and discovered some tools like Invesalius, (Seg3D; Biomesh, from SCI/Utah) and Gmsh. However, could some one give me a hint on other softwares - or even the trick - about how to segment the images in order to have both marrow and bone exported as a volume ready for FE processing?

I want to study aneurysm growth (in volume) in a set of subjects, based on Time of Flight images (MR) images at two time points.

For each subject, I want to do a rigid registration between time of flight images as well as a segmentation of the cerebral vascular system. And then, I want to compare the two registered segmentation volumes in order to quantify the aneurysm growth.

For doing this, some teams used in-house registration tools (not distributed as far as I know, AnToNIA for ex) and the registration is performed using a volume of interest around the aneurysm sac (and not using the whole brain).

I currently know free registration tools not specific to aneurysms (FLIRT of FSL for exemple: http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/FLIRT). Additional question: Do you think these kind of methods is appropriate to do "aneurysms registration" ? 

Thanks in advance,

I need it for image registration and dose distribution applications.

I am trying to register US image and CT scan image of cardiac. CT images are in axial plane, however, the US images are in parasternal view( sagittal plane). Is there any methods that can I use to do this registration or I have to acquired the US images in same plane as the CT images? Please help me.

Thank you. 

What is a reasonable distance in mm when applying registration on longitudinal mammograms images ? I would like some state of art measure or personal opinion of some clinical/radiographic specialist.

Dear all, I would like to ask you something about current research in medical image registration. Could you give me some current major challenges and issues in medical image registration, especially stained histological sections (different dye per section). Do you know where would be possible to get some more data to work on? Thanks

- ftp://cmp.felk.cvut.cz/pub/cmp/articles/borovec/Borovec-ISBI2013.pdf

- ftp://cmp.felk.cvut.cz/pub/cmp/articles/kybic/Kybic-ISBI2014.pdf

- ftp://cmp.felk.cvut.cz/pub/cmp/articles/kybic/Kybic-BIC2015.pdf

Can anybody suggest an Image registration method with available implantation for and benchmark comparison on medical images? I found couple: ITK, Elastix, bUnwarpJ, TurboReg... Some other suggestion?

I am curious to know how VTK estimates the transformation matrix in the VTKLandmarkTransform method? Does it involve a solution of equations given a minimum set of points in the source and target OR does this estimation (estimation of transformation matrix) involve some sort of optimisation, for example, optimising the fiducial registration error?

Thanks in advance..

I am doing research on processing medical images which contain some cells which adhere to other cells. Considering the fact that I should extract the cell’s number, I need to find a method in order to make these cells separated.

It would be kind of you to give me your valued suggestions.

Thank you.

Marjan

Can PCA based dimensionality reduction be used as preprocess ?

GPU based image registration ?

I am acquiring image sets from both modalities in an attempt to use the resolution of CT and the soft tissue contrast of MRI for more accurate pathological tissue characterization. I am not sure whether I need to limit my FOV in the MRI to the lowest setting of our uCT in order to be able to do meaningful image analysis. Any help would be greatly appreciated.

Intensity based, mutual information.

I have 6 spectral channels, every channel as an independent image.

I want to make an RGB image, I need a formal method or scientific proof for this making method.

I want to apply a shape-based template matching detection

Can I register a pair of CT brain images using cubic b-spline in hierarchical form for monomodal image registration?

As I am doing my research on non rigid registration of brain MRI images

I am doing research in nonrigid registration of brain images, now first I want to implement my code on 2D images as my code is for multimodal registration