Imaging Trends

[medinah]

hans, f_w, and others...thanks for sharing

About imaging trends, I read that CT and MRI are most lucrative nowadays and that PET and PET/CT are on the up and up. You guys think that the latter are going to surpass the former in the near future? Also, is it true that mammograms actually cause a decrease in profit (due to decrease in reimbursements) and that's why mostly hospitals conduct them?

 

[hans19]

hans, f_w, and others...thanks for sharing

About imaging trends, I read that CT and MRI are most lucrative nowadays and that PET and PET/CT are on the up and up. You guys think that the latter are going to surpass the former in the near future? Also, is it true that mammograms actually cause a decrease in profit (due to decrease in reimbursements) and that's why mostly hospitals conduct them?

CT/PET is integrally, part CT. In the near future look for MR/PET. PET will not replace, so much as augment those modalites. PET alone actually does not reimburse as well as MR or CT alone.

Mammography does not make the group much profit. Rather, when rads groups contract with a hospital, part of the contract calls for providing mammography services as well as things such as vascular access and on call IR coverage, so its part of a package deal.

 

[medinah]

So do you think that PET/CT and PET/MR will, at some point, become more lucrative than CT or MR? If so, how many years from now would it be noticeable? And, in general, do you expect imaging reimbursements to change eventually?

 

[f_w]

About imaging trends, I read that CT and MRI are most lucrative nowadays and that PET and PET/CT are on the up and up.

'lucrative' is relative. If you can participate in the technical component of the reimbursement, the equation looks different from a situation where your income is based on the professional fee alone (but at the same time, you are more exposed to reimbursement cuts as some people are starting to find out this quarter).

You guys think that the latter are going to surpass the former in the near future?

I don't think so.
PET-CT won't help me to diagnose a torn meniscus, an extruding disc or a solid renal lesion. It is however quite useful for a subset of cancer patients and certainly starts to dominate staging and re-staging for certain cancers (if available).

Also, is it true that mammograms actually cause a decrease in profit (due to decrease in reimbursements) and that's why mostly hospitals conduct them?

There are actually enterpreneurial mammography practices in existence that make quite a good living on breast imaging alone.
If you batch-read screening mammos and you are one of these people that are able to crank them out at a high pitch (I am not), you are going to make bank.
The downside in mammo is the liability exposure. 60% of liability expenses in radiology are related to mammo but only 10% of the revenue.

 

[medinah]

Do radiology residencies train in PET, PET/CT, and SPECT?

 

[hans19]

Do radiology residencies train in PET, PET/CT, and SPECT?

Most residencies should. However, the only thing mandated by the ACGME for radiology residency is 4 months of nucear medicine. It does not mandate that the facility have PET, or SPECT. You would hope that in choosing your residency you would take these factors into consideration.

 

[medinah]

So can it be assumed that most PET and PET/CT scans in the real world are read by nuclear radiologists or nuclear med. physicians?

 

[hans19]

So can it be assumed that most PET and PET/CT scans in the real world are read by nuclear radiologists or nuclear med. physicians?

They are read by radiologists who have had nuclear medicine fellowships, or by radiologists otherwise comfortable reading nuclear medicine studies.

There are very few jobs for nuclear medicine physicians in the 'real world'.

 

[f_w]

They are read by radiologists who have had nuclear medicine fellowships, or by radiologists otherwise comfortable reading nuclear medicine studies.

A very smart clinical scientist at Siemens-CTI told me about 15 years ago that one day F18DG will just be another contrast agent for CT.

 

[Neuronix]

I'm still unsure how the field will handle these combined imaging/nuc med modalities. Vendors are making and showing off all their new technologies that integrate say MRI and PET and CT and SPECT. It's probable that one day we will have all of these modalities combined into one machine. Now who will read these multi-modality coregistered studies? As it is my hospital has to send all its coregistered CT + nuclear med studies to an affiliated hospital for reads by one of the rare guys who's dual boarded.

As a technical note f_w, I think it's a little misleading to say 18FDG will be a CT contrast agent. When I think of a CT contrast agents I think of something that changes the density of the area they reside in blocking the passage of X-rays. But since 18FDG gives off gamma waves, very similar in frequency to X-rays, the positron decays can be detected by the same detectors. Hence one can (and people have) build combined CT/SPECT scanners. The idea is that you can whirl the CT around with the X-ray source on and get an anatomical set, then inject contrast and whirl the detectors around without a source and get a functional set.

But this comes back to my question, you've just done two "separate" studies. Now who reads it? Given this future, it seems screwy to me that Rads and Nucs are even two separate entities. But, I don't know much about such things.

 

[f_w]

As a technical note f_w, I think it's a little misleading to say 18FDG will be a CT contrast agent.

Semantics.

The point being that today we are using FDG in a suitable scanner to map metabolic activity of tissues in much the same way we use Iodinated benzene compounds to map perfusion.

Hence one can (and people have) build combined CT/SPECT scanners.

The 'Hawkeye', great device. Unfortunately, reimbursement for these combined studies never compensated for the additional expense and the vendors sort of phased them out.

But this comes back to my question, you've just done two "separate" studies. Now who reads it?

Ideally a 'nuclear radiologist'. That is a diagnostic radiologist with the additional year of nucs training of a nucs fellowship.

 

[UnderDoc]

Neuronix --

While it would be nice, I don't understand how the machine you describe could ever be built. As you know, PET technology relies on nanosecond resolution timing circuitry to allow for the co-localization of positron annihilation events along a linear axis. Multiple axes are samples allowing for the triangulation of these events within two-dimensional space. Without a dedicated 360 deg detector ring, I don't see how you could ever hope to localize said events in space. The 511keV gamma rays produced wouldn't undergo consistent enough attenuation in soft tissues to allow for their localization based on attenuation alone, as far as I can understand it. Also, these high energy photons are quite penetrating and wouldn't be detected by the CT digitizer.

 

[Neuronix]

PET and CT on a single detector... What I was saying works for CT/SPECT, but you're right it would never work for CT/PET. Still, there is at least one idea for doing PET/CT, see for example:

CT Acquisition Using PET Detectors and Electronics
Berard, P.; Pepin, C.M.; Rouleau, D.; Cadorette, J.; Lecomte, R
IEEE TRANSACTIONS ON NUCLEAR SCIENCE, VOL. 52, NO. 3, JUNE 2005

So yes, by whirling around what I was really referring to was SPECT, not PET. You got me there. I didn't think that one through all the way. Though, you can still combine PET and CT into a single detector by detecting X-rays on a 360 degree ring.

Since I was curious I started looking up all kinds of things:

SPECT/CT have been combined into a single detector (what I was really referring to in my other post).

A single CdZnTe detector for simultaneous CT/SPECT imaging
William C. Barber, Koji Iwata, Bruce H. Hasegawa, Paul R. Bennett, Len J. Cirignano and Kanai S. Shah
Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment
Volume 505, Issues 1-2 , 1 June 2003, Pages 595-598

The Hawkeye has two separate detectors.

That being said, there's other tricks people are using to combine modalities like PET and MR too. For example:

Development of a combined microPET-MR system.
Lucas AJ, Hawkes RC, Ansorge RE, Williams GB, Nutt RE, Clark JC, Fryer TD, Carpenter TA.
Technol Cancer Res Treat. 2006 Aug;5(4):337-41.

These guys stuck the detectors in the magnet. One of many possibilities.

 

[f_w]

But since 18FDG gives off gamma waves, very similar in frequency to X-rays, the positron decays can be detected by the same detectors.

Lets see:
CT 140kVp photons
PET 511kV coincidence pairs

Yes, there used to be Anger type nucs scanners known as 'coincidence cameras' that where able to detect both (after switching out collimators), but there are some inherent difficulties with the thick NaI crystals used on these setups.
Also, the geometry of acquiring a coincidence and a transmission image at the same time could be challenging.

There was a concept known as the 'ECAT-ART' from CTI. In order to save on the expense of a ring-detector (the most common PET setup), they used two small sections of the ring and spun it around the patient. The folks that developed the ART once bolted a Siemens 2-slice detector and tube combo on the back of the gantry and that way had a coregistered dataset. (However, the PET still took 35min to acquire while the CT took 30seconds. So in the end, they didn't have more of a coregistration than any of the other designs).