4 ms·

Because scans are not booked wall to wall every hour of every day. Needs more likely follow a poisson distribution with ample waits in between uses. To try and

Because scans are not booked wall to wall every hour of every day. Needs more likely follow a poisson distribution with ample waits in between uses. To try and pack it that densely, with the large variation in how people show up at the hospital, would cause bottlenecks and loss of service. 20 a week is a more common number for most places. Here's actual data [1] measured on site. The usage is indeed around 50%, so the 20 is a much better answer. [1] https://www.researchgate.net/figure/MRI-scanner-utilization-rate-over-the-3-week-data-collection-period-MRI-scanner-type-and_fig1_51852320 https://www.researchgate.net/figure/MRI-scanner-utilization-...

DuskStar 7y ago
40% utilization means that it was in use for 67 hours per week. I don't think an MRI takes three hours. (In the case of the study listed, the utilization figures might be out of ~35 hours, which would make things different - but they also exclude scans scheduled in advance)
- ChrisLomont 7y ago
  The link states "As a result, in our study the maximum hours per week on each scanner was limited to 35 h for the Vision (no hours on Tuesdays), 39 h for the Harmony, 32 h for the Sonata (no hours on Wednesdays) and 38 h for the Avanto." Why guess 67 when you can look at the answer?
  - DuskStar 7y ago
    Because from my reading that's not the number of hours they used the machines, but the number of hours in which the machines were available to be used. And since they also state that their utilization figures only include emergency cases (and exclude scheduled scans) I don't consider it a particularly relevant source in the first place.
    - ChrisLomont 7y ago
      >I don't consider it a particularly relevant source in the first place. Care to provide a relevant source?
- mattkrause 7y ago
  MRIs take as long as you can afford. I have actually run ~3 hour MRIs for experiments, but we're being very fussy and trying to get the best possible image with the smallest possible voxel, which you do by acquiring tons of volumes and averaging them. Obviously, you can be in and out of there much quicker if you're just looking for a gaping hole.
snowwrestler 7y ago
This is data from a hospital. In the U.S. at least, a lot of MRI scans are done in outpatient centers, scheduled well in advance, and so utilization can be much higher. This would certainly be true of prophylactic scans.
- ChrisLomont 7y ago
  The utilization could also be much lower. If not enough people need to sign up for each and every day, then it will not be full. It's extremely unlikely the needs in a region match the capacity anywhere near 100%. Such a tight capacity would result in many problems, so reasonable planning would have excess capacity to deal with busy seasons/holidays/weekends/etc. Just like the hospital case, things like this likely follow a Poisson distribution, and it is therefore reasonable to assume similar usage patterns. Have a link with data for your claim? I looked and found none.
lostlogin 7y ago
A lot has changed since 2011 but even back then the MR scanner should always have been full and the next patient should have been waiting outside. If a scanner is very early or late you call the next patients and alter scheduling. Anything else is a waste of time and money.
- ChrisLomont 7y ago
  There's a reason we design roads, electric grids, food supply, and so on, to not perfectly match needs. It's impossible and foolish to assume constant rates of needs. For the same reason, no hospital would assume MRI scanners are 100% in use all the time. You cannot schedule people needs that perfectly without fail, on holidays, summer versus winter, day and night, weekends versus Mondays, and so on.
roywiggins 7y ago
Utilization is less in a hospital, especially when you're measuring "active scanning time" and not all the time that you're really using a room. Hospitals also don't book scans so aggressively since someone might need a scan pronto- in an emergency, or near-emergency- and it can't be scheduled weeks in advance like you do in outpatient settings. Complex cases where you need anesthesia, especially, may involve a lot of mucking about before and after the "active scanning" period, where the room is being used, but the scanner is not actively scanning anyone. Also, paediatrics cases. Utilization is a lot higher in outpatient settings where you can line patients up and turn the room around in a few minutes.
pbhjpbhj 7y ago
>Because scans are not booked wall to wall every hour of every day. // I had a scan in the UK, and this is exactly how it was done. Mine was actually in a mobile unit set up in a car park (for several months). It took about 20mins and I was in the ante-room getting prepped as someone else was in the scanner, someone else was in the waiting room waiting to go next. At least a couple an hour and there were a couple of shifts of workers too (I asked). As an aside the workers were all foreign, how fortunate we were/are and how sad for those countries to lose their talent to us.

DuskStar 7y ago

40% utilization means that it was in use for 67 hours per week. I don't think an MRI takes three hours. (In the case of the study listed, the utilization figures might be out of ~35 hours, which would make things different - but they also exclude scans scheduled in advance)

ChrisLomont 7y ago
The link states "As a result, in our study the maximum hours per week on each scanner was limited to 35 h for the Vision (no hours on Tuesdays), 39 h for the Harmony, 32 h for the Sonata (no hours on Wednesdays) and 38 h for the Avanto." Why guess 67 when you can look at the answer?
- DuskStar 7y ago
  Because from my reading that's not the number of hours they used the machines, but the number of hours in which the machines were available to be used. And since they also state that their utilization figures only include emergency cases (and exclude scheduled scans) I don't consider it a particularly relevant source in the first place.
  - ChrisLomont 7y ago
    >I don't consider it a particularly relevant source in the first place. Care to provide a relevant source?
mattkrause 7y ago
MRIs take as long as you can afford. I have actually run ~3 hour MRIs for experiments, but we're being very fussy and trying to get the best possible image with the smallest possible voxel, which you do by acquiring tons of volumes and averaging them. Obviously, you can be in and out of there much quicker if you're just looking for a gaping hole.

snowwrestler 7y ago

This is data from a hospital. In the U.S. at least, a lot of MRI scans are done in outpatient centers, scheduled well in advance, and so utilization can be much higher. This would certainly be true of prophylactic scans.

ChrisLomont 7y ago
The utilization could also be much lower. If not enough people need to sign up for each and every day, then it will not be full. It's extremely unlikely the needs in a region match the capacity anywhere near 100%. Such a tight capacity would result in many problems, so reasonable planning would have excess capacity to deal with busy seasons/holidays/weekends/etc. Just like the hospital case, things like this likely follow a Poisson distribution, and it is therefore reasonable to assume similar usage patterns. Have a link with data for your claim? I looked and found none.

lostlogin 7y ago

A lot has changed since 2011 but even back then the MR scanner should always have been full and the next patient should have been waiting outside. If a scanner is very early or late you call the next patients and alter scheduling. Anything else is a waste of time and money.

ChrisLomont 7y ago
There's a reason we design roads, electric grids, food supply, and so on, to not perfectly match needs. It's impossible and foolish to assume constant rates of needs. For the same reason, no hospital would assume MRI scanners are 100% in use all the time. You cannot schedule people needs that perfectly without fail, on holidays, summer versus winter, day and night, weekends versus Mondays, and so on.

roywiggins 7y ago

Utilization is less in a hospital, especially when you're measuring "active scanning time" and not all the time that you're really using a room. Hospitals also don't book scans so aggressively since someone might need a scan pronto- in an emergency, or near-emergency- and it can't be scheduled weeks in advance like you do in outpatient settings. Complex cases where you need anesthesia, especially, may involve a lot of mucking about before and after the "active scanning" period, where the room is being used, but the scanner is not actively scanning anyone. Also, paediatrics cases. Utilization is a lot higher in outpatient settings where you can line patients up and turn the room around in a few minutes.

pbhjpbhj 7y ago

>Because scans are not booked wall to wall every hour of every day. // I had a scan in the UK, and this is exactly how it was done. Mine was actually in a mobile unit set up in a car park (for several months). It took about 20mins and I was in the ante-room getting prepped as someone else was in the scanner, someone else was in the waiting room waiting to go next. At least a couple an hour and there were a couple of shifts of workers too (I asked). As an aside the workers were all foreign, how fortunate we were/are and how sad for those countries to lose their talent to us.