Science topic

Medical and Biomedical Image Processing - Science topic

Discussing specific questions of image processing of medical and biomedical images including pattern recognition.
Questions related to Medical and Biomedical Image Processing
  • asked a question related to Medical and Biomedical Image Processing
Question
4 answers
Dear researchers.
I have recently started my research in detecting and tracking brain tumors with the help of artificial intelligence, which includes image processing.
What part of this research is valuable, and what do you suggest for the most recent part that is still useful for a PhD. research proposal?
Thank you for participating in this discussion.
Relevant answer
Answer
In current technology era, to sustain and provide healthy life to humans it is necessary to detect the diseases in early stages. We are focused on Brain tumour detection process, it is very challenging task in medical image processing. Through early diagnosis of brain, we can improve treatment possibilities and increase the survival rate of the patients. Recently, deep learning plays a major role in computer vision, using deep learning techniques to reduction of human judgements in the process of diagnosis. Proposed model is efficient than traditional model and provides best accuracy values. The experimental results are clearly showing that, the proposed model outperforms in the detection of brain tumour images.
  • asked a question related to Medical and Biomedical Image Processing
Question
4 answers
I'm working of DICOM images for the project I need multiple segmentation in one volume, anyone interested?
Relevant answer
Answer
Look the link, naybe useful.
Regards,
Shafagat
  • asked a question related to Medical and Biomedical Image Processing
Question
3 answers
I have recorded hologram is best on digital in-line holography technique. The distance between the objects and the sensor (CMOS sensor,taking from web cam) is typically 5 mm. The illumination source is LED ,that is placed at a distance of 5 cm from the sample and is filtered through a large pinhole having a diameter of 0.1 mm .
Relevant answer
Answer
Kamrul Hasan You can easily get the phase of refractive index, thickness, etc. for the desired hologram, which I edited for you.
  • asked a question related to Medical and Biomedical Image Processing
Question
2 answers
Any short introductory document from image domain please.
Relevant answer
Answer
In general, the linear feature is easier to distinguish than the nonlinear feature.
  • asked a question related to Medical and Biomedical Image Processing
Question
7 answers
Medical Imaging.
Relevant answer
Answer
Hi,
Have you already found the answer to your question?
  • asked a question related to Medical and Biomedical Image Processing
Question
6 answers
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?
Relevant answer
Answer
Image Pre-Processing Techniques are classified into four kinds, which are given below.
1. Brightness transformations/corrections for pixels
2. Geometric Transformations
3. Image Filtering and Segmentation are the third and final steps.
4. Fourier transform and image restoration are examples of Fourier transform and image restoration.
Kind Regards
Qamar Ul Islam
  • asked a question related to Medical and Biomedical Image Processing
Question
2 answers
I have searched about the REVIEW database but I can't find it for download. besides, The vampire dataset is not a public database. So, I had sent vampire non-commercial license to use the annotation tool. But they didn’t answer.
could anybody send me the datasets?
Relevant answer
Answer
So thanks Mr.
J. Rafiee
  • asked a question related to Medical and Biomedical Image Processing
Question
6 answers
I'm looking for a PhD position and opportunity in one of the English speaking university in European countries (or Australia).
I majored in artificial intelligence. I am in the field of medical image segmentation and My thesis in master was about retinal blood vessels extraction based on active contour. Skilled in Image processing, machine learning, MATLAB and C++.
So could anybody helps me to find a prof and PhD position related on my skills in one of the English speaking university?
Relevant answer
Answer
You have any examination is cleared than you go for scholarship
Like in NET etc
  • asked a question related to Medical and Biomedical Image Processing
Question
4 answers
I am an engineer/researcher  developing code for finding feature parameters of irregular shaped 2d closed curves . For example a square has 4 corners and one width/height , circle infinite corners/radius, oval infinite corners/ 2 lengths major and minor. So I was wondering if we could have various measure for real life curves as found in US cardiac images such as width of the ventricles.Any help would be deeply appreciated.
Thanks,
Sushma
Relevant answer
Answer
following
  • asked a question related to Medical and Biomedical Image Processing
Question
6 answers
I would like to research on MR images (0.5T and 3T). Can you please suggest some websites that I can download dataset including both 0.5T and 3T MR images? Thank you.
Relevant answer
Answer
Try searching the image dataset library here :
  • asked a question related to Medical and Biomedical Image Processing
Question
3 answers
I am trying to train a UNET architecture built with the PyTorch library. I am looking for the best methods to load my data into the model. the dataset has around 500 images (RGB image of around 3000x4000 pixels).
I am planning to extract patches of an image into (256,256,3) shaped patches alone with augmentation while training.
I am looking at methods that do not overload the ram at the same time not compress the data too much to lose the quality of data.
Relevant answer
Answer
The best way is to start with basic/fundamental methods/approaches, followed by the attached book.
  • asked a question related to Medical and Biomedical Image Processing
Question
2 answers
Images are required for testing and validation of a new technique.
Relevant answer
Answer
Can anyone suggest to me from where I can find ultrasound breast images with ground truth
  • asked a question related to Medical and Biomedical Image Processing
Question
16 answers
For my research I need cancer( tumor) digital pathology image database.Type of cancer is not important but the data should have labels.
Relevant answer
  • asked a question related to Medical and Biomedical Image Processing
Question
11 answers
I would like to do research on MRI images. Can you please suggest some websites from where MRI images can be downloaded. Thanking you. Gautam Talukdar. 
  • asked a question related to Medical and Biomedical Image Processing
Question
15 answers
I am working on Medical image segmentation Problems, I would like to know is there any public dataset available for Lung and Liver images
Relevant answer
Answer
  • asked a question related to Medical and Biomedical Image Processing
Question
12 answers
Before displaying an ultrasound image, the machine usually performs a logarithmic compression in order to better visualise an image in terms of grayscale. In order to treat a DICOM image, is it necessary to do a logarithmic decompression for more accurate results of image analysis and processing?
Relevant answer
Answer
I am agree with Larbi Messaouda ; It depends on the type of processing and analysis you want to perform.
  • asked a question related to Medical and Biomedical Image Processing
Question
10 answers
I am working with image processing techniques on breast thermograms.
Relevant answer
Dear Sheeja,
You can use our database:
In case you decide to use it, please follow the license instructions and cite the works we recommend there.
Regards.
  • asked a question related to Medical and Biomedical Image Processing
Question
10 answers
I need a "public" benchmark image dataset for prostate segmentation. Images can be CT/MR/ultrasound (preferably MRI) but at least 3 experts, ideally more,  must have segmented each image.
  • asked a question related to Medical and Biomedical Image Processing
Question
3 answers
I am a university student and I am doing research in medical image processing.
I need the adolescent (12 to 18 years old) brain T1 and diffusion-weighted images data set.
Actually , I really need this to conduct my research. Could anyone please help me to find a dataset for that?
Relevant answer
Answer
Hi Lakshika. I suggest you register and then download data from the Human Connectome Project (HCP) database which has thousands of scans from ages across the lifespan, including you adults: https://www.humanconnectome.org/study/hcp-young-adult/data-releases
Jerome
  • asked a question related to Medical and Biomedical Image Processing
Question
3 answers
Iris recognition.
Relevant answer
Codes_xor = bitxor(IrisCode_subject , IrisCode_db);
% get the anding of the masks
Masks_and = bitand(Mask_subject , Mask_db);
total = sum(sum(Masks_and)) ; % number of pixels to be taken into consideration
hd = sum(sum(bitand(Codes_xor,Masks_and)))/total;
  • asked a question related to Medical and Biomedical Image Processing
Question
3 answers
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 ?
Relevant answer
Answer
Can someone refer to any text or website that provides a step by step guide for image reconstruction in PET using list-mode data
  • asked a question related to Medical and Biomedical Image Processing
Question
4 answers
CTU-UHB Intrapartum CTG database consists of CTG records and clinical information. The data in database has been extracted from the OB TraceVue system to an open format using software. I need the raw data of this database. The database contains .dat and .hea file. But i cann't find a way to extract the raw data in the database. Can anyone help me?
Relevant answer
Answer
where can i find the data dictionary for the clinical parameters for this data? I could not find on physionet site..Thanks for any help
  • asked a question related to Medical and Biomedical Image Processing
Question
6 answers
we need a 4D untrasound device that allows personal laptops to be the main part of the scanner with endaroscope ,echo cardioscope ,vaginal ,rectal and abdominal  sonic probes as attachments, is there any available? how effective.
Relevant answer
Answer
This is not add.
However, what is the quality of such ultraounds?
  • asked a question related to Medical and Biomedical Image Processing
Question
6 answers
I have done by both UG and PG in Biomedical Engineering. I would like to peruse my PhD in Medical image processing/Analysis with machine learning for prediction. My area of focus would be either cancer images or Neuro images. After completing PhD other than teaching what is the scope?Please help.
Relevant answer
Answer
This is a new and emerging field.
  • asked a question related to Medical and Biomedical Image Processing
Question
14 answers
I've recently started pursuing PhD in Medical Image Processing. I am struggling to find a suitable problem to carry out research.
I was initially interested in working on brain images but after doing studies and according to peers' advices lots of works have already done with brain, but still work can be done on:
1. Retinal images
2. Chest radiography
3. Estimating probability of lung cancer etc.
I would be obliged if somebody suggests me some relevant topics(related to brain or other organs) where still there are lots of scopes to work on.
Relevant answer
Answer
Well, there are are thousands of interesting topics for many imaging modalities. It might be better to decide what you want to do first, so that you will be able to limit your search.
Do you want to work on image reconstruction, segmentation, analyses or registration? Which imaging modality you want to work with (CT, MRI, ultrasound, mammography, X-rays, SPECT-CT, PET-CT, DEXA, interventional imaging, multimodal, etc)? Are you interested in 2D or 3D imaging, clinical or pre-clinical? Do you want to work with deep learning or no?
After answering some of these questions, look for research groups that are working in a related discipline and their current projects, or recent publications for an inspiration. Also check proceedings of major image processing conferences (e.g., MICCAI, NeurIPS, SPIE Medical Imaging, Fully3D, IEEE MIC)
Good luck!
  • asked a question related to Medical and Biomedical Image Processing
Question
6 answers
I wonder if anyone has a recipe for a gel phantom for using for USS training.
for vascular grafts
Relevant answer
Answer
See our recent work on developing patient specific ultrasound phantoms with a new phantom material gel wax
  • asked a question related to Medical and Biomedical Image Processing
Question
6 answers
For medical image processing research.
Relevant answer
Answer
Apply the concept of Transfer Learning using Deep Convolutional Neural Network, and use any algorithm with slight hyperparameter tuning such as GoogleNet, VGGNet, ResNet, or AlexNet etc.
  • asked a question related to Medical and Biomedical Image Processing
  • asked a question related to Medical and Biomedical Image Processing
Question
141 answers
ISI rapid publication journals.
Relevant answer
Answer
you can choose the journal according to your work from the below links
  • asked a question related to Medical and Biomedical Image Processing
Question
3 answers
Hi,
I want to use a MRI images dataset in order to detect heart failure with image processing techniques. In the beginning I should choose the kind of map I need. T1 map or T2 map, I should choose one of them.
But I don’t know what is the differences between them and which one is better for detecting heart failure.
Can anyone share some information about that?
Relevant answer
Answer
T1 mapping allows you to quantify extracellular volume (ECV). Normal values for ECV are 25-28%, higher ECV values most usually represent interstitial myocardial fibrosis (as in hypertension, hypertrophic cardiomyopathy, severe aortic stenosis,...), values over 40% represent amyloidosis or infarction-related scar
T2 mapping measures specifically interstitial edema (thus, it is usually employed to detect the acute phase of myocardial infarction or acute myocarditis, as overtime edema will disappear)
One important point, though. Heart failure is defined CLINICALLY not by MRI techniques. MRI can help you to identify the cause of HF, but HF itself is diagnosed clinically (clinical history, physical examination, chest X ray,...)
  • asked a question related to Medical and Biomedical Image Processing
Question
5 answers
I am interested in doing an ROI-based analysis of ADNI FDG-PET data to examine regional metabolism in their control group. I am using a coregistered volumetric MRI to create anatomic ROIS and exporting these as masks to use on the PET data. Since I'm using ROIs created from each subject’s own MRI scan, is it still necessary to do a partial volume correction, or is that something you do only if you’re warping everyone’s images into a common space (e.g. MNI)? And, if I need to do partial volume correction anyway, is it a reasonable alternative to simply erode each MRI-based ROI mask by 1 voxel all around to avoid the issue, instead of doing a more complicated partial volume correction method (e.g. segmenting a coregistered MRI image, calculating a CSF dilution factor and then correcting the raw time-activity curve of the PET in each region)?
Relevant answer
Answer
My mistake on my previous answer I wanted to say that this method estimates an average radioactivity and not a homogeneous for the two background regions.
Sorry for my mistake and any misunderstanding caused.
In addition to my previous answer, the Müller-Gartner method will only correct for one region (most of the cases the GM) and will not treat the other regions of the brain.
  • asked a question related to Medical and Biomedical Image Processing
Question
4 answers
In what diameter size a lung nodule in CT scan is considered malignant nodule, for example in luna16 dataset there is annotation file called 'annotations.csv' that contains a nodule diameter, if I want to use this file for classification purpose based on diameter size, what is the range of diameter size that I can choose it as malignant nodule. kindly give me your answer based on a knowledge or a reference?
Relevant answer
Answer
Fleichner recommendations are the way to go. FDG PET is very helpful in SPN if size is more than 8 mm
  • asked a question related to Medical and Biomedical Image Processing
Question
9 answers
Hello everyone,
I am currently developing my method for 3D image segmentation of medical images (CT and MRI), so I would like to see if anyone have any published papers on this subject, along with the source code written in MATLAB so that I could compare my method against it?
I want to compare the segmentation accuracy and time efficiency and test my algorithm against existing methods, preferably that have been published recently in renowned journals with impact factor?
Thanks in advance!
Relevant answer
Answer
Hi Lambert,
Thank you for your clarification on this issue and whether or not the above mentioned approach by me is possible.
Regarding the CBCT data, the problem that I am trying to address using my approach is to extract the correct shape of the data on CBCT images.
In my case, I have maxilla scanned using a CBCT scanner, but due to several influencing factors (operator, bad setup parameters, but also bone porosity of the maxilla), there are some missing information of the data that cannot be extracted using commercial medical imaging software.
My aim is, by using my developed method, is to address those issues. So I am using a segmentation as an approach to extract the correct information from the data. And that is also the reason why I wanted to test my method on MRI data too.
I have acquired manually delineated CBCT data by expert which I am using as a golden standard to test my method as well as for two cases of MRI data too.
But now I am looking to expand my no. of cases for testing, as you suggested.
Thanks!
  • asked a question related to Medical and Biomedical Image Processing
Question
3 answers
In many dead cases there is useful information available in dead person's brain missing which can be used for discovering one's mind regarding different events happened to him or his relatives. I was wondering if there's any solution to extract information based on dead one's brain signals.
Relevant answer
Answer
Tough to say what ever "will be" possible in the future.
As far as current technology can tell, dead people do not emit brain signals, which is part of what defines them as being dead in the first place. There may be some information stored in the brain, like in the hard drive of a computer, which could be "read" with the right technology, but we are nowhere close to doing that yet. We haven't even made it to the point of recognizing thoughts in living people, so I think extracting information from the brains of the dead is something for far, far in the future, if ever.
  • asked a question related to Medical and Biomedical Image Processing
Question
5 answers
Could you please point me out to some successful Signal, image, or video processing real life applications using partial differential equation?
Preferably, involving heat, reaction-diffusion, Poisson, or Wave equation.
If possible in fuzzy environment.
Best regards
Sarmad.
  • asked a question related to Medical and Biomedical Image Processing
Question
4 answers
Hello all,
I am a researcher working on color image processing and have knowledge in Machine Learning and Programming. Currently, I am planning to deviate my research towards the Medical and Biomedical application. Suggest some research topics in Medical/Biomedical application which has scope in future?
Relevant answer
Answer
Biomedical sciences are applied sciences applying portions of natural science to interventions, or technology that are of use in healthcare or public health such as medical microbiology, clinical virology, clinical epidemiology, genetic epidemiology, and biomedical engineering are medical sciences. I researched in .
  • asked a question related to Medical and Biomedical Image Processing
Question
9 answers
I have developed a method using which the edges in images are get enhanced. Which evaluation metric shall I use to state that edges in image are improved or not? I am working on brain tumor images.
Relevant answer
Answer
Follow two steps:
1. Build the edge map from your image.
2. Compare PSNR and SSIM for the edge map.
  • asked a question related to Medical and Biomedical Image Processing
Question
7 answers
hi all
The benefits of Multilevel Image Segmentation Versus 2-level Image Segmentation? 
thanks
Relevant answer
Answer
In multilevel image segmentation you are able to segment images into several part instead of twp part. So if you have more than 1 object you can not segment all object by using bi-level segmentation but multilevel can be helpful in this case. In the other hand determine the optimum number of segment is handicap in this subject.
  • asked a question related to Medical and Biomedical Image Processing
Question
4 answers
Suppose I have a dicom file with pixel size (0.5,0.5), and I want to make the pixel size to (0.7,0.7) or vice versa. How can I change/modify the pixel size of a dicom file?
  • asked a question related to Medical and Biomedical Image Processing
Question
7 answers
I'm using the batch scripts generated by SPM12 in my own Matlab codes. I'm trying to use "deploytool" in Matlab2015b to compile my codes into an "exe" file. I've included SPM12 as the required files to run the program. But when I run the "exe" file, I got errors as follows:
> Undefined function or variable 'cfg_mlbatch_appcfg_master'.
> Error in cfg_util>local_initapps (line 1440)
> Error in cfg_util (line 743)
> Error in spm_jobman (line 189)
One link I found online related to this problem is not helpful: https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind1606&L=spm&F=&S=&P=612689
I don't know if anyone knows how to solve this problem.
Thanks!
Relevant answer
Answer
Well, you will have to make changes small changes and hacks in core codes for doing this...
Core codes will have if conditions as
>> if(isdeployed)
which basically means it will run different codes when you are running deployed codes, but such deployement is done by you, and you would not require such different codes, (in fact you dont even have them..., thus the errors crop up..)
Here is what you could do:
Comment out the extra checks of isdeployed in the core codes..
You could do that just for the codes you are using, in this case just corresponding to matlab batch which exists in spm/matlabbatch..
The files to be changed are basically
cfg_util.m, cfg_ui.m and cfg_mlbatch_appcfg.m ...
Regards
Chaithya G R
  • asked a question related to Medical and Biomedical Image Processing
Question
5 answers
I am working on emotion recognition using a pretrained DCNN for feature extraction. I have obtained 1 X 1000 features from the DCNN. And i want to reduce its dimensionality. PCA have been suggested. But I want to know if there are better alternatives for PCA.
Relevant answer
Answer
Hi Shehu,
regarding the answer given by Amir, look at this two links
and paper 
you can easily implement autoencoder and get non linear dimensional reduction.
  • asked a question related to Medical and Biomedical Image Processing
Question
6 answers
The files has the extention .vol, and the format is called "Kretzfile". It is a 4D ultrasonic study for fetal heart analysis. Doctors called the analysis STIC (Spatio-Temporal Image Correlation). The images were adquired using an ultrasound system "Voluson 730" of General Electric. I need the images for my research project related to segmentation and optical flow estimation.
Relevant answer
Answer
I've implemented a GE/Kretz 3D ultrasound image reader in 3D Slicer. After you have loaded the image you can use all the awesome tools in 3D Slicer to visualize and process it (for example, to create a 3D-printable model). You can see a demo here:
It's not perfect yet (spherical to Cartesian conversion is not fully accurate), but it's completely free and open-source - fixes and improvements are welcome. For further details and questions please post to the 3D Slicer forum:
  • asked a question related to Medical and Biomedical Image Processing
Question
15 answers
I am interested to segment it (somehow) and get the motion information from an echo of the heart. Any comment also on this part?
Relevant answer
Answer
I've implemented a GE/Kretz 3D ultrasound image reader in 3D Slicer. After you have loaded the image you can use all the awesome tools in 3D Slicer to visualize and process it (for example, to create a 3D-printable model). You can see a demo here:
It's not perfect yet (spherical to Cartesian conversion is not fully accurate), but it's completely free and open-source - fixes and improvements are welcome. For further details and questions please post to the 3D Slicer forum:
  • asked a question related to Medical and Biomedical Image Processing
Question
2 answers
I am using Dice score to compare image segmentation.
But what is the significance of Kappa score, Positive predictive Value and Jaccard coefficient with respect to Image segmentation?
How can use these values to tune my algorithm?
Relevant answer
Answer
The correctly and incorrectly classified instances show the percentage of test instances that were correctly and incorrectly classified. The raw numbers are shown in the confusion matrix, with a and b representing the class labels. Here there were 100 instances, so the percentages and raw numbers add up, aa + bb = 59 + 12 = 71, ab + ba = 27 + 2 = 29.
The percentage of correctly classified instances is often called accuracy or sample accuracy. It has some disadvantages as a performance estimate (not chance corrected, not sensitive to class distribution), so you'll probably want to look at some of the other numbers. ROC Area, or area under the ROC curve, is my preferred measure.
Kappa is a chance-corrected measure of agreement between the classifications and the true classes. It's calculated by taking the agreement expected by chance away from the observed agreement and dividing by the maximum possible agreement. A value greater than 0 means that your classifier is doing better than chance (it really should be!).
The error rates are used for numeric prediction rather than classification. In numeric prediction, predictions aren't just right or wrong, the error has a magnitude, and these measures reflect that.
Hopefully that will get you started.
  • asked a question related to Medical and Biomedical Image Processing
Question
4 answers
Hi. I am currently working on a medical image processing project that segments the coronary artery blobs from the axial CT slices of the human heart and converts the segmented coronary blob slices to a 3D coronary artery model. I wish to extend my project by mapping the 3D model to the 2D axial input slices, ie, by clicking on a point in the 3D model, it must display the appropriate CT slice and also the point to the position in that slice. Are there any methods ( techniques ) or softwares available to do this? If so, how?
I am using OpenCV to implement this project.
Relevant answer
Answer
Suggest using OpenGL
1. set your camera so screen projection plane is parallel to the slice...
2. clear screen buffer as usual with glClearColor set to background color
3. Clear your depth buffer with glClearDepth set to Z-coordinate of the slice in camera space
4. set glDepthFunc(GL_EQUAL)
5. render mesh
glClearColor( 0.0,0.0,0.0,0.0 ); // <0.0,1.0> r,g,b,a
glClearDepth( 0.5 ); // <0.0,1.0> ... 0.0 = z_near, 1.0 = z_far
glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);
glDepthFunc(GL_EQUAL);
This will render only the slice for which fragments have Z==Slice coordinate. This can be also done by GLSL by throwing away all fragments with different Z. The DirectX should have something similar (I do not use it so I do not know for sure).
As most meshes are BR models (hollow) then you will obtain circumference of your slice so you most likely need to fill it afterwards to suite your needs...
  • asked a question related to Medical and Biomedical Image Processing
Question
17 answers
By physical dimensions, I mean that the spatial resolution of the images is specified in terms of millimetres. 
Relevant answer
Answer
Hello Dr. Gupta,
That is true, but at best it is a guess based on an assumption of continuity at a minimum across the region of interpolation, at worst a distortion that will result in incorrect decisions.  And there is no way to know, in an individual situation, which it is. Should the data on which the assumption is made be noisy, incorrect, or imprecise (as is nearly always the case), to try to extract resolution that is not in the image is risky.
I don't do that.  I increase the resolution of the imaging modality if I can or else I live with the resolution that I have.  Anything else is too risky because of the potential for misinterpretation.
At one of the annual major medical imaging conventions in Chicago, I saw an image supposedly correctly up-sampled using the latest gee-whiz algorithm.  When I compared it to the original, I brought to the attention of the presenter the fact that after the algorithm, it appeared that the patient had an extra rib.
The problem is not the algorithm, the problem is the lack of understanding that the algorithm in question does not apply to the application in question, and further, the lack of understanding of the risks incurred by using it.
I have seen similar problems with supposedly correct applications of algorithms both before and since, many times.
Best regards,
Peter
  • asked a question related to Medical and Biomedical Image Processing
Question
21 answers
I'm trying to analyse images of animal tissues produced with a 3DHistech slide scanner. I usually use ImageJ but have problems fitting large enough images in the memory. Are there any other free alternatives, or tricks that could be used with ImageJ? My analysis is simple, just detecting & measuring circular holes of certain size (adipocytes). The images are like 25000 pixels x 25000 pixels.
Relevant answer
Answer
QuPath Open source software for digital pathology
Orbit Image Analysis
Both software are easier than matlab. There cut images into tiles automatically.
  • asked a question related to Medical and Biomedical Image Processing
Question
23 answers
Thanks in advance for your replies.
Relevant answer
Answer
I've opened a repository in github with our script so it can be available to everyone :)
The link also includes instructions for using the script :)
Let us know if it's working for you, and on the other hand
If you experience any trouble using it, let us know in the issues section!
  • asked a question related to Medical and Biomedical Image Processing
Question
17 answers
Does anybody know any software which can convert pictures from MHA format to .raw format?
I used visualization-MITK workbench2016.11.0 software, but it has not such convertor. 
Relevant answer
Answer
@Reuben Reyes: Thank you Reyes. I converted my all image in MHA format use VV. It's the one that I needed.
  • asked a question related to Medical and Biomedical Image Processing
Question
6 answers
I want to evaluate my 3D tumor segmentation and I do not have any idea how to evaluate that?
Regards
Ali
Relevant answer
Answer
Use BRATS dataset..
  • asked a question related to Medical and Biomedical Image Processing
Question
4 answers
I would like to get sample tomosynthesis imaging
Relevant answer
Answer
yes i want this type of image
  • asked a question related to Medical and Biomedical Image Processing
Question
9 answers
I have 83 ct dicom series image  and I want to remove patient table from these image.for this purpose I write the following code in the MATLAB software:
a = dicomread('E:\dicom\44.dcm');
b=double(a);
sample=a(50:100,250:300);
background=mean(mean(sample));
for i=1:512;
for j=1:512;
if i>385;
b(i,j)=background;
end
end
end
image=dicomwrite(b,uint16(b),'44.dcm');
after I run this code patient table  remove from my image but image contrast is unacceptable.
how to  adjust image contrast with matlab?
please help me.
Any help would be much appreciated. Thanks.
Relevant answer
Answer
dear Dr Luminita Moraru 
I want to use dicom image for monte carlo simulation in radiotherapy with beamnrc/dosxyznrc code.
  • asked a question related to Medical and Biomedical Image Processing
Question
6 answers
Lung Imaging
Relevant answer
Answer
Yes i agree with Prof Sanad,receptor persent in lung are the best for diagnosis of particular disease by phramaceutical imaging via SPECT and PET..
  • asked a question related to Medical and Biomedical Image Processing
Question
2 answers
Respected all,
I need Follicular Lymphomas tissue sample for my project. If anyone of you have information regarding that please share to me 
Thanks in advance
Relevant answer
Answer
Dear sir 
Please visit Nikon's public database for your research https://www.microscopyu.com/galleries/pathology
Hope this will help you
  • asked a question related to Medical and Biomedical Image Processing
Question
19 answers
I would like to detect edges on 3D MRI images but with intensities variations in the some regions, I keep obtaining false edges.
I applied canny edge detection on the image prceeded with anisotropic filtering and you can see the false edges in the attached picture (green circles )
Is there some processing strategy that can solve this problem?
Relevant answer
Answer
Hi Amal,
Basic suggestion: did you vary the Canny thresholds? This wasn't mentioned in the prior discussion. It might give you a sense of the strength of the noisy edges that you want to avoid, versus edges that you want to detect. 
Peter
  • asked a question related to Medical and Biomedical Image Processing
Question
8 answers
I know several quantitative quality measures for contrast enhancement in medical images, but sometimes is not enougth and I need the opinion of physicians about the visual quality of the processing.
Are there procedures, surveys, instruments, software to do that?
Relevant answer
Answer
Use entropy, mean, standard deviation and contrast
  • asked a question related to Medical and Biomedical Image Processing
Question
4 answers
I need some public digital tomosynthesis datasets for detection of lung nodules.
Relevant answer
Answer
If you still need them let me know
  • asked a question related to Medical and Biomedical Image Processing
Question
4 answers
Is there any method for the selection of best wavelet for the texture classification for medical images? 
Relevant answer
Answer
Depends on which imaging modality you are working with and what your images look like. In fact, depending on your images, wavelets for texture discrimination might not even be the right approach. You might want to look at gray-level cooccurrence matrices: 
Also, you could also look at histogram-based methods for discrimination. Here is a publication that might interest you:
  • asked a question related to Medical and Biomedical Image Processing
Question
2 answers
I'm doing an imaging study in patients with sciatic nerve radiculopathy, and I'm trying to find out the vertebral level that is adjacent to where the L4, L5, and S1 nerve roots enter the cord.
I understand that there will likely be a large amount of variability among people, and most of the resources I've seen so far suggest that they enter the cord somewhere between the level of L1, T12, and T11 vertebrae.
Does anyone know of any studies (e.g. fiber tracing, DWI) that have attempted to more formally answer this question? 
Thanks for your help!
Dan
Relevant answer
Answer
HI Dirk, thanks for your response. I am interested in tracking the roots - L4 to S1. However, my original question should have been more clear: I'm actually interested in finding out where the roots terminate in the cord. Because I'm interested in sensory afferents carrying nociceptive information, and because these fibers enter into Lissauer's tract, traveling up or down several segments before synapsing, I wondered if there were any good references describing where they are most likely to terminate. 
If anyone is interested, a colleague directed me to the following publication, which may have some insights into the question:
Thanks,
Dan
  • asked a question related to Medical and Biomedical Image Processing
Question
5 answers
simple inflamed appendicitis has limitation in diagnosis using simple and cheap ultrasound machines. Any other way from new modern ultrasounds apart from clinical assessment and costly CT and MRI..
Relevant answer
Answer
Ultrasound is very useful in the diagnosis of acute appendicitis, particularly in children where the burden of radiation exposure from CT scans are more of an issue than in adults.  Consider the use of both high frequency linear trandsducers, and also mid-low frequency curvi-linear transducers.  Ultrasound has a high negative predictive value, even when the appendix is not directly seen with ultrasound, and there is a lack of secondary signs that are mentioned above.
Thank you.
  • asked a question related to Medical and Biomedical Image Processing
Question
8 answers
Hi, 
I have a set of segmented images and i want to select ,for each new image, the best match in this collection. witch algorithm can i use for such task?  
Thanks 
Relevant answer
Answer
Scale Invariant Feature Transform (SIFT) is the best nowadays, SURF is a semilar approach but faster. To match these features, there are different well-known similarity measurements.
  • asked a question related to Medical and Biomedical Image Processing
Question
6 answers
Which classification tool is better for Medical Image analysis. I am planning on working on medical images to facilitate image works within help sector.
Relevant answer
Answer
For medical image analysis you can use Wndchrm.
  • asked a question related to Medical and Biomedical Image Processing
Question
28 answers
I am finding difficult to read medical images (.dcm) in MATLAB. Please suggest.
Relevant answer
Answer
Hi,you can use the following code to read and show dicom image with matlab
[x,map,alpha,overlays]=dicomread('full directory path of your dicom file')
imshow('x')
note, this code doesn't work for dicom image series.it works only for single dicom image.
  • asked a question related to Medical and Biomedical Image Processing
Question
5 answers
I have several synthesized quantum dots and magnetic particles. I feel that they can be used as contrast agent for bio imaging. So for that I want to do some lab based experiment. Then go for animal model. So please guide me what are the primarily experiments that are possible to proceed in this area.    
Relevant answer
Answer
Thanks all for your responses.
  • asked a question related to Medical and Biomedical Image Processing
Question
5 answers
I want to do research about retinal images and finally extract vessel and optic disk and exudates. but i cant find angiogram database.
who can help me?
and
you think which one is better for my Thesis? segmentation from fondus images(DRIVE) or angiography images?
Relevant answer
Answer
Dear Mrs.Shaabani
there are some appropriate database as your need:
moreover below dataset site can be helpful:
bellow survey reference can be useful for second Question about data selection:
[1] Fraz, Muhammad Moazam, et al. "Blood vessel segmentation methodologies in retinal images–a survey." Computer methods and programs in biomedicine 108.1 (2012): 407-433.
:
  • asked a question related to Medical and Biomedical Image Processing
Question
6 answers
A saliency map (2D image)and a segmented image(2D image) of same image is already having. On the basis of saliency map, i have to calculate average saliency for each region of segmented image.
Relevant answer
Answer
if you look up the graph based saliency toolbox in matlab on the web, you can use that to segment your region in a way that its pretty easy to just run a boxcar saliency indicator over. an example would be statistical noise via a local STDev estimate. I can point you to some change detection publications that explain the process.
rereading your other posts - why not just do a binary map of the region of interest with some morphological tool, then run a window of saliency estimates of your choice (see my prior comment) along the whole mxn matrix?
  • asked a question related to Medical and Biomedical Image Processing
Question
4 answers
We are trying to understand how capability of 3D tissue scaffolds made of polymers and ceramics is affected by their internal architectures. Currently, we have obtained a series of micro-CT images (about 300) of a polymer-made scaffold. We want to use imageJ to achieve 3D reconstruction and get to know overall porosity, surface area and distribution of pores and ratio between open and closed pores. 
I wonder if there are any imageJ plugins can do this. If not what other software can be used?
Your help will be very appreciated. 
Relevant answer
Answer
HI,
If you have Micro CT software, I think you can achieve what you want to see. Basically, micro CT software can reconstruct 3D Images and you can see clear of pores properties.
  • asked a question related to Medical and Biomedical Image Processing
Question
9 answers
Are there any metrics to compare the robustness of 2 features for an image? I need to compare and choose which feature vector would be apt for my spine image dataset.
Relevant answer
Answer
i found out that HOG features are best when compared with eigen features or fisher features. i don't know particularly about the spine images in specific, but i can tell you that you can start with finding the HOG features. 
i worked on face images. you can see my paper explaining about finding the HOG features and the importance of HOG features. 
instead of spending time in identifying the methodology for the best feature vector, you can work with already existing and experimentally proven features one by one. i hope you will be succeeded with the first experiment.
MATLAB is the best platform for finding the features for images. 
plenty of MATLAB programs are available in the internet.
previously i tried other features like SIFT and so on. i didnot get as much output as HOG features. 
all the best
  • asked a question related to Medical and Biomedical Image Processing
Question
3 answers
i am trying to start research in medical image processing, which type of medical image is in recent research and which one is to better?
Relevant answer
Answer
Brain parts and tissues, mammograms, lungs and biometric access. 
  • asked a question related to Medical and Biomedical Image Processing
Question
9 answers
Hello Friends:
I'm sending out a general call for your help in deciding on a thesis topic for my MS degree. I had an MSEE and had worked as an electronics engineer for 20 years (in US Defense, FiberOptics, Data Storage, Machine Vision industries). Now decided to leave it behind and returned to school this past January 2016 for an MSBE degree (Ph.D. later on, God willing) since I wanted to meld my knowledge in electronics with the new field of bioengineering(new to me anyways). The problem is I have no life sciences background (Biology, Chemistry, BioChem, etc.) to speak of since high school (and that was a really long time ago!). All thesis project ideas I have seen are so biology/life science-based which I, unfortunately, can't do (without spending another year or two acquiring that knowledge). Designing standard bioelectronics devices such as heart rate monitors/pacemakers etc. bores me silly and honestly don't think one can submit that as a thesis topic as still be an honest individual. I have the following interests:
- Image Processing ... I designed an IP board as my MSEE thesis back in the day). Analyzing MRI scan images seem to have been done to death. Is it still a topic worthy of an MS degree? If so, any new challenges here?
- Bioimpedance measurements ... I'm new to this but find the whole concept intriguing and fascinating. But designing body fat measurement devices is also too boring and technically dumb. Not worth an MS degree in my opinion. Sorry!
- Biosensors ... new to them as a concept and find them also very intriguing and fascinating. But I'm afraid my lack of biology/biochemistry would be a hindrance and burden to me.
Any suggested project must be realistically doable within a 4-5 month timeframe, tops. I'd start in Spring 2017 then finish in the following summer.
Thanks All :)
- David
Relevant answer
Answer
A rapidly evolving area of very diverse  needs may be described broadly as “Independence Engineering”: Solutions for people living with disabilities, chronic disease or who are simply frail and aging, can incorporate significant innovation or adaptation of our existing technologies. It may not seem as sexy as many of the historical advances that leap to mind when thinking of Biomedical Engineering, but the marketplace is huge and growing, and many of the innovations require less life science knowledge. Understanding of neurotransmission, the cardiac cycle or the workings of the immune system are hardly vital to be effective .  Just an idea, Good Luck!....GW
  • asked a question related to Medical and Biomedical Image Processing
Question
4 answers
My requirement is to find the volume of the sample within the MRI image which consists of the sample, water and air. I have the data in the form of intensities in a 3D array of size 128*128*128 in which the sample corresponds to black pixels and the water volume corresponds to lighter pixels. Also, the outside air volume and the air bubbles present which are black. I am trying to use watershed function to find the connected sequence of pixels and finding the one with maximum frequency (I assume the sample to be the largest connected sequence of dark pixels), but somehow it is not detecting the sample volume as connected one. Also, bwconncomp and the labelmatrix are not of much help. Any suggestions/ideas on how to find this ?
Relevant answer
Answer
1. Find all connected clusters of voxels in a certain intensity range
2. Merge clusters that belong to your sample (the exact strategy would depend on your sample shape and location characteristics). You can do it manually as first approximation and then see if cluster merging can be automated.
  • asked a question related to Medical and Biomedical Image Processing
Question
3 answers
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.
Relevant answer
Answer
when you employ clustering with inhomogeneity correction the number of misregistrations is reduced without loss of accuracy thus increasing robustness as compared to the standard non-inhomogeneity corrected and equidistant binning based registration.
  • asked a question related to Medical and Biomedical Image Processing
Question
1 answer
 When i used the command "patch" on Matlab it depends on the size of vertices and faces and i can't modified it ?
Relevant answer
Answer
I think you need to take a look at the matlab documentation: the vertices are controlled by the input parameters.
"patch(X,Y,C) adds a filled 2-D patch object to the current axes. A patch object is one or more polygons defined by the coordinates of its vertices. The elements of X and Y specify the vertices of a polygon . If X and Y are m-by-n matrices, MATLAB® draws n polygons with m vertices. C determines the color of the patch. For more information on color input requirements, see Coloring Patches."
  • asked a question related to Medical and Biomedical Image Processing
Question
4 answers
For my research, I am doing MRI image analysis. I have the data in the form of gray levels (only 1 and 0 for dark and bright pixels) in a 3D array. I am looking for a solution to find the boundary between dark and white area and mark the boundary with gray(Similar to the red boundary of the black box as shown in the image). Is there a way to identify the whole boundary?
I have tried using bwboundaries() and bwtraceboundary() functions, but of not much use.
Thanks in advance
Relevant answer
Answer
One easy solution would be to use a dilation morphological operator. Dilate your image by one voxel, then substract the original image from your dilated image. The result is the boundary region. You can then multiply by 0.5 and sum with your original image to have the boundary region in gray:
ana = analyze75read('test.hdr');
dil = imdilate(ana, ones(3,3,3)); 
boundary = dil-ana;
output = 0.5*boundary + ana; %may need to convert to float, since it'll round up if you're using ints 
Matlab also has a tool for 2D images called bwmorph. If a slice-by-slice approach is acceptable, you can run:
ana = analyze75read('test.hdr');
slice = ana(:,:,30);
bwmorph(slice, 'remove')
And then just repeat per slice.
  • asked a question related to Medical and Biomedical Image Processing
Question
9 answers
I need implementation code for retinal blood vessels segmentation from funds images using LINE OPERATOR. Here is the published paper on it. 
" Retinal Blood Vessel Segmentation Using Line Operators and Support Vector Classification " .
Link to paper is attached. 
Relevant answer
Answer
 Dear Mahtab and everyone who is following the post,
Here is the code attached to this text for your information, I hope it will work for you.
  • asked a question related to Medical and Biomedical Image Processing