By looking at a medical image, how can i identify whether it is an ultrasound or MRI or CT image?
When you want to identify, whether it is an CT, MRI or US image, you should understand the imaging technologies.
So lets have a look:
Computer tomography (CT) scan:
CT imaging is based on X-rays. It uses the ability of high frequency electromagnetic waves to pass through soft tissue of the human body. Dense structures like bones and teeth have a high intensity in the resulting scans and can be clearly detected, whereas soft structures have a low intensity.
Magnetic resonance imaging (MRI) scan:
MRI scanners are generating a magnetic field, which stimulates hydrogen atoms to emit a radio frequent signal. Thus tissues with a lot of water have a high intensity in MR images and regions with no water have a low intensity.
Ultrasound (US) imaging:
US images are formed by measuring the amplitude and the travel time of the reflected sound echoes to reach the US transducer. If there is a high acoustic impedance mismatch between two tissue interfaces the reflected sound echo signal will have a high amplitude resulting in a high intensity pixel appearance in the formed US image. The appearance of different tissue in US images is described here: https://theultrasoundsite.co.uk/ultrasound-case-studies/starting-article-2-different-tissues-ultrasound-look-like/
To be able to answer correctly we need to know exactly the purpose of the request. Does the recognition of the technique used for diagnostic images serve for a layman (patient, relative, curious) or is it formulated by a healthcare professional?
If the question is from a layman, the answer by Sonja Jäckle is more than enough, unless more details are requested
In case the doubt comes from a healthcare professional, the first thing to ask is, what is the level of this healthcare professional? On average, healthcare professionals have this type of knowledge. We are talking about macroscopic differences that should not escape even those who only clean inside a health facility.
Dear Screepriya, the differences among these images are fundamental. The important step is to understand the imaging technique and as a result the formation of the image. I would propose to look for a book about medical imaging technology. However, if you have to tell the difference among these images and you cannot do it based on your knowledge and experience I would suggest to see the radiology report of the images or use a medical imaging software. I believe that the majority of these softwares can tell you what kind of a scan you have.
Dear Screepriya, the differences among these images are fundamental. The important step is to understand the imaging technique and as a result the formation of the image. I would propose to look for a book about medical imaging technology. However, if you have to tell the difference among these images and you cannot do it based on your knowledge and experience I would suggest to see the radiology report of the images or use a medical imaging software. I believe that the majority of these softwares can tell you what kind of a scan you have.
I think an interesting question is if an AI system can identify the type of image acquisition by looking into eg. the definition of contours, distribution of grey values and other image related features. Might be that CT versus MRI is the most challenging algorithm.
System may identify modality of the image with looking into the DICOM Tag (if it is correct).
If there are jpeg images only, it is slightly difficult to identify the modality automatically.
To understand this we take 3 to 5 years. But still will try to answer. Well for ultrasound images mostly in sector formate. For ct and mri its difficult for non medical background to understand but still one can differentiate. In ct images, the bony parts will be bright and in shades of white more. But in case of mri it gives a detailed view of muscular or soft tissue regions and the bony portion will be in darker shades of grey.
Follow the contrast between fat-water mostly and checking bone and vessel contrast might help sometimes. Cheers!
Thank you so much for the responses.
As a part of my research work, i would like to develop a program in Matlab to differentiate these medical images. Please let me know if any one of you could help me.
You're welcome! Maybe I could be for some assistance, but I have to update my Matlab first. Best wishes!
In MR images it is difficult to differentiate between bone and soft tissues. However, there are few sequences available which can be used to differentiate between bone, fat and muscle tissues.
One such sequence is ''Dixon''. It acquires an in-phase image and out of phase image which are than used to construct a "water-only" image and a "fat-only" image. For bone ultra short echo (UTE) sequence can be used to differentiate bone signal from soft tissue.
dear Sreepriya P.
Look at the link, maybe useful
https://www.independentimaging.com/abcs-imaging-difference-xray-ultrasound-mri-ct-scan/
- Badges
- Science topic