Thank you, my lecturers just don’t get to the point and waffle without precise definitions. The future of universities should replace crappy lectures with video segments like this
Thanks for the vote 🗳. Pass these videos to your friends. These videos are open to everyone so anyone can use in a lecture or watch ahead and do activities during lecture time
Sir you r just doing a very good job No one make videos on radiography in India as like u I am becoming a huge fan of yours ❤❤ Mashallah 😊 Thank you sir for helping us Lottts of luv from students like me --From India 🇮🇳
It’s very nicely explained. I have a question. What does it mean when MTF exceeds 1? Is it good when it is much larger than 1 or it should be as close as possible to 1?
The MTF is normalize to the value at the zero frequency, so if other values go above one that means that the kernel is doing boosting, ie some frequencies have been artificially enhanced. This is common in lung imaging for instance in CT. These boosted kernels are good for visualization but should not be used for quantitative measurements. Hope that helps 😉
The line spread function is directly related to the MTF, the MTF is the Frequency space representation of the LSF. The blue in the image is due to the detector blur and focal spot blur
Great video and thanks for sharing! I have one question about the line-pair QA. I wonder why we need to remove the Cu plate when performing line-pair QA? Thank you!
Thanks Mathew, since that test is for the limiting spatial resolution the idea is use a small spot and remove the filtration so more X-rays get to the detector. This test will represent as well as the system could do.
Once you have the PSF can you use the FWHM to determine the spatial resolution of the system or should you always need to pass from the MTF ant take R10?
The PSF does contain the information on the spatial resolution and the FWHM is a good metric for overall resolution. Taking the mtf and looking at the 10% or 4% gives a better understanding of the limiting resolution that you can visualize
@@HowRadiologyWorks thank you so much for answering. One more question. Having a wire phantom, once you get the point spread function from it, would you suggest getting the FT of the fit or getting the MTF using a bar resolution phantom? I guess the second one has more sense
Ayla- changing the focal spot changes the resolution but does not change the field of view and changing the collimation does not in general change the resolution ( there are some exceptions in fluoroscopy).
Like as always, it depends. In normal x-ray and CT imaging that is true. In newer systems there is more nonlinear behavior with iterative reconstruction in CT or advanced denoising in x-ray. However since these tools are well established they are often applied even when the system is not exactly linear.
@@HowRadiologyWorks So is it why manufacturers nowadays use the DQE metric to reflect image quality? Because from my understanding, noise and MTF are both embedded in the DQE equation. Do you think you can make a video to clarify that?
@@authman-alshibly that is right and the short answer is because it is still the most accepted option. For a certain task there are more specific methods but nothing that generally covers everything like DQE. I will make one on DQE next year as I have a few other requests ahead.
Great video. Better description than any Physics text I've ever read on mtf.
Thanks Prof Tim. Much Appreciated!
Thank you, my lecturers just don’t get to the point and waffle without precise definitions. The future of universities should replace crappy lectures with video segments like this
Thanks for the vote 🗳. Pass these videos to your friends. These videos are open to everyone so anyone can use in a lecture or watch ahead and do activities during lecture time
Sir you r just doing a very good job
No one make videos on radiography in India as like u
I am becoming a huge fan of yours ❤❤
Mashallah 😊
Thank you sir for helping us
Lottts of luv from students like me
--From India 🇮🇳
You are welcome 🙏 Thanks for these nice words ☺️
Thank you for this clear explanation👍
You’re welcome ☺️
Nice presentation, thanks!
Glad you liked it! Thanks for leaving the comment
Thank you for this remarkable lecture. Which system would have the highest MTF?
What are the choices? In standard radiology mammography has highest resolution
@@HowRadiologyWorks thank you, or as we say in my country 'JazakAllah'
It’s very nicely explained. I have a question. What does it mean when MTF exceeds 1? Is it good when it is much larger than 1 or it should be as close as possible to 1?
The MTF is normalize to the value at the zero frequency, so if other values go above one that means that the kernel is doing boosting, ie some frequencies have been artificially enhanced. This is common in lung imaging for instance in CT. These boosted kernels are good for visualization but should not be used for quantitative measurements. Hope that helps 😉
@@HowRadiologyWorks Thanks a lot
@howradiologyworks,
Is there a significant impact to MTF and final image clarity based on the original LSF and Focal Point measurements?
The line spread function is directly related to the MTF, the MTF is the Frequency space representation of the LSF. The blue in the image is due to the detector blur and focal spot blur
Great video and thanks for sharing! I have one question about the line-pair QA. I wonder why we need to remove the Cu plate when performing line-pair QA? Thank you!
Thanks Mathew, since that test is for the limiting spatial resolution the idea is use a small spot and remove the filtration so more X-rays get to the detector. This test will represent as well as the system could do.
Once you have the PSF can you use the FWHM to determine the spatial resolution of the system or should you always need to pass from the MTF ant take R10?
The PSF does contain the information on the spatial resolution and the FWHM is a good metric for overall resolution. Taking the mtf and looking at the 10% or 4% gives a better understanding of the limiting resolution that you can visualize
@@HowRadiologyWorks thank you so much for answering. One more question. Having a wire phantom, once you get the point spread function from it, would you suggest getting the FT of the fit or getting the MTF using a bar resolution phantom? I guess the second one has more sense
Thank you so much!
Your welcome.
Can MTF exceed 1?!
Some books say yes it ican be in digital radiography system!
Is this true or false?
The MTF can exceed 1 especially when the vendors apply a boost to mid frequencies and the MTF is typically normalized at zero frequency
@@HowRadiologyWorks 🙏
Hi! I was wondering if making the focal spot smaller is the same as collimating an image as they both change the size of the primary beam?
Ayla- changing the focal spot changes the resolution but does not change the field of view and changing the collimation does not in general change the resolution ( there are some exceptions in fluoroscopy).
@@HowRadiologyWorks thank you!!!
So is it true here that the imaging system is a linear system and that the resultant image is the convolution of the ideal image with the PSF?
Like as always, it depends. In normal x-ray and CT imaging that is true. In newer systems there is more nonlinear behavior with iterative reconstruction in CT or advanced denoising in x-ray. However since these tools are well established they are often applied even when the system is not exactly linear.
@@HowRadiologyWorks
So is it why manufacturers nowadays use the DQE metric to reflect image quality? Because from my understanding, noise and MTF are both embedded in the DQE equation. Do you think you can make a video to clarify that?
@@authman-alshibly that is right and the short answer is because it is still the most accepted option. For a certain task there are more specific methods but nothing that generally covers everything like DQE. I will make one on DQE next year as I have a few other requests ahead.