"The manufacturer quoted a price of £2,000 to the trust as the cost of sourcing this MOD unit."
They obviously never tried ebay ... http://preview.tinyurl.com/bbmjcs6
A hospital trust says it will cost the NHS £2,000 to dig out a copy of a patient's ultrasound scan of his heart and hand it over to him. The steep bill is, we're told, due to the data being held on a magneto-optical disc, and Worcestershire Acute Hospitals NHS Trust would need to fly in kit from America to access it. The …
"What happened to "Any data held about a named data subject in machine readable form must be made available to the data subject for a nominal fee" Data Protection Act 1984 - did this disappear in the later revision?"
I'm guessing they are talking about the "machine readable" bit...
That said, I am left thinking "Mountain ... Mohammed" just send the disk by courier to someone who still has a working drive.
Okay, it might be considered a data risk, but if this gent was reasonable he would waive it- I can't think of anyone doing something nefarious with this old scan data.
Can't people be reasonable?
To be fair, I think the article title is a little misleading. It is quite clear from the article text that the £2000 figure is coming from Philips US as the cost to ship over a new drive.
It seems to me that the hospital is acting very responsibly - they are checking with the Information Commissioner what their obligations are, and they did the background work to see what would be required to comply even before they received an answer. They have said neither that they won't provide the image nor that they would pass on any bill to the patient, just that £2000 on this instance would not be an efficient use of public money - and on that I agree with them.
Of course, the question remains - if they have the disc in this format, they must have had a drive for it at some time, and presumably some other UK hospitals were using similair tech. Isn't there a single working drive in the whole UK they can borrow to just read the disc and get the file out, without needing to BUY a drive (new or eBay)?
From the article "In the event that any patient requires future treatment all clinical decisions and treatment are based on diagnostic tests undertaken at the time of admittance and as required rather than from historical records"
Having just read the BBC article - its a shame theyre not allowed to just tell him to piss off or to recreate the image himself. He's just looking for attention.
"In the event that any patient requires future treatment all clinical decisions and treatment are based on diagnostic tests undertaken at the time of admittance and as required rather than from historical records"
In some cases sure, a completely new scan is all that is needed, but I'm sure that in some cases there is some value to being able to have a before-and-after scan, and the comparison of the old scan with the new one can be medically relevant.
There is an awful lot of research going into medical imaging (using some *very* big beasts - Oak Ridge National Laboratory, I'm looking at you). One of the threads is comparing scans from before, with scans from now. The software can spot minute differences which could be diagnostic.
Also, part of the research is to aggregate old scans and outcomes to build up an expert system (showing my age there) of what is "normal" and "abnormal" for future use.
So old data still has a value.
"Because old scans aren't medically important and thus it would be a waste of public money?
That would depend. In my case, I'm being denied an insurance claim on a severe accident which has left me with injuries to my back because I have no proof that the injuries shown in a recent MRI weren't present long before as I never had an MRI on that region. An MRI from just five years ago would settle the argument. Fortunately I have the recourse to sue the insurance provider and am doing so.
Paris, obsolete eight years ago, as well.
"Why is the old data not migrated to new storage/format at the time of retiring the old kit?"
Probably because migration is a costly and time consuming process. For non-essential records it's probably cheaper to take the risk that they wont be needed any more.
I'd say the real question is why doesn't the NHS use its buying power more efficiently. You want an £x million supply contract with the NHS for data storage? Well then you're required to keep one of each drive type, for occasional loan, for Y years after the contract ends.
Working with some smaller medical offices has shown me that the cost is insane. I've seen quotes on conversion for many times what the already expensive costs. Many clinics do it on a patent by patient basis. If you visit the clinic or call in they will manually migrate the records over in the time the old system is still running, but if you call in 10 years later, it's pretty much 'gone'.
> although the scan is still on file ... it is stored in a format that can no longer be read by the hospital's computers.
A backup (or archive) is only a backup if you can restore it. Just copying stuff to some other location/format is pointless until you've verified you can read it back.
Although the hospital says they can read the file (given access to the right kit), they can't be sure it hasn't been corrupted. So until they try to access the scan there's no guarantee the scan still actually exists - just that the physical medium on which it's stored is present.
You'd hope that if the NHS does go to the effort of acquiring hardware to read this scan, they'll take the opportunity of transferring all their other MOD media onto something a little more modern, rather than just doing a one-off and shipping the reader back from whence it came. Presumably there are lots of other NHS shops in the same situation, so with even the slightest use of initiative, the cost of this activity could be shared. I hold out very little hope ...
I am the patient in this story and if it was only my scan they couldnt access I might well have let it gho, but I established it is hundreds or thousands of scans and that is ( I think) a matter of public interest
in my locality
1. The NHS is supposed to back up and securely store data
2. 2006 guidance (NHS) required them to prepare for new platforms
and ensure readability of data
3. There are 200 disks s which they say they cannot read due to a lack of MOD drive
and these extend up to 2008 which isn't that long ago
4. Old ultrasound data is potentially clinically useful or even crucial, to compare one scan with another, to facilitate a 2nd opinion, to identify incompetent practice, or even for a NHS hospital to defend against
I have purchased MRI and X ray in the past, all managed efficiently and for a tenner a go
I will say in the medical systems I've seen recently, that standards are becoming more common. One system I recently migrated to creates pdf files of most records as a copy outside the database and imaging is stored in a TIFF format. The previous version was a ball full of suck, it used an ancient ASA database that the images were stored in the database bloating it to huge sizes. The practice didn't migrate the records over from the old system because the costs were staggering, when you break it down to per record costs it was cheaper to have people manually move each record over. That's just gouging by the software firms. That's why we end up in situations like the article.
PDF and TIFF? Can't say I'm a fan of that idea. That's again using proprietary stuff (are there any patents stuffed up the sleeve of PDF, a la GIF?) and, well, the whole thing is just fairly unmalleable. How best would you import into your new system from a PDF?
I don't understand what the NHS buys to make everything so incompatible. These days everything should be using the DICOM standard for storing, transmitting and sharing medical images. It's been around since 1985, so most people should have caught up by now!
DICOM, more usefully than TIFF (or PDF) stores the image but also stores several shitwads of metadata information about the picture, such as patient name/ID/age/etc, when the image was taken, what it was taken of (i.e. Right Knee), then lots of other useful information such as exposure time (in the case of X-Rays) or co-ordinates (in the case of MRI/CT scans). It also allows for annotations to be stored which can be superimposed on the other end if required.
Source: I implement and maintain PACS systems.
Tiff actually allows any amount of metadata that you want. It also supports layers and the layers suport degrees of transparency too. The layers can contain both bitmapped and vector content, and bitmapped content can be in different resolutions to each other. Oh and varying degrees of compression, which can be lossless. A Tiff file can also be multiple pages, and lower capability renderers tend to just show what they can, leaving out what they don't understand. Tiff has come along way since just being 1-dimensional Huffman coding.
...good old fashioned stupidity on behalf of the purchasing authority.
I stumped up the cash for a CD containing the images of my MRI scans a couple of years ago, or this could have been the start of 2011.
The problem was that it didn't contain the images themselves, but bespoke files which needed a piece of software to read them and then generate the images from them.
The software (which came on the CD along with the files) was compatible with Microsoft and Apple, (I run Linux) but the real gut wrenching laugh was when I read the read-me on the CD that said that it needed IE 5 !!!
I mean ... excuse me?
Someone in the system needs shooting. I still have the CD, and the pictures of my MRI that I took off the consultants screen with my camera phone.
CD's never contain "images themselves" they contain files in a format that requires software to read them and generate images from them. In this instance, the NHS recognised that you probably wouldn't have the right software on hand, and supplied it without being asked, and ensured that it would run on a very high percentage of home computers. But because you run an OS that is in the vast minority, and couldn't find a way of getting to a computer running any other OS (really?) , you think someone in the system is failing...
Sorry, this storyis an example of the NHS getting it right, no need to turn your own failings on them.
Someone needs shooting, because they only support ~98% of the workstation OSes used? Hyperbole, much?
As it happens, I know someone who has recently had an MRI scan and read the file on a Linux machine with standard FOSS software, then generated a 3D print of his brain, all using Linux. Did you even bother to look for the software to read the files, or just thought it was worth a good bitch?
As it happens, this came up as a link from THE FIRST HIT of a Google search for "foss to read mri files"
Well, as I didn't know anything about these types of files, I thoguht it was definitely worth a good bitch.
I mean, I paid for a CD of the images and I expected a CD of images. Even the hospital couldn't argue on that point.
As for Linux being a small time operating system ... that is a tired argument used by desperate people.
The whole point of this is proprietary lock in, as the hospital couldn't export the files to anyting like jpeg or tiff either. They were stuck. As operating systems move on, they're going to be eventually as screwed as the person in the original article.
So yes, as far as I can see it, it is an epic NHS fail and no messing.
Interesting education about the file types though. I'll see if I can get that working, although by the file numberings and sizes I doubt that there is enough to generate a 3d image of my brain.
"...As for Linux being a small time operating system ... that is a tired argument used by desperate people..."
Actually I'm a linux user, I use it at home and work, desktop and server at both locations. I don't have much sympathy with the "pity me I'm a linux user" attitude that you display, especially when it took me less than ten seconds to find information about the software you need on Google.
Like it or not, linux is small time on the desktop, sure there are lots of servers, but no-one can really argue that it's a significant player in the desktop market. I wouldn't expect every piece of software to be available for it, but I don't go around bitching, I do expect a little more footwork on some occasions. However your complaint seems to come from someone who just wants to complain, rather than bother to sort the problem out for themself.
Michelle, this probably has stuff all to do with the NHS - when they make these CD images they just press "Export" in whichever manufacturer's PACS system they use (and there are enough) and you're at the mercy of whatever Philips/Agfa/Siemens/etc/etc decided to write.
That said, there is the odd trust who likes to build their own CD frontends - these usually come written in VB :-(
We receive patient discs from a very wide geographical area when people visit us for healthcare. I'll agree - the general standard is dire. It's a nightmare trying to actually get the images out of some of them.
Oh, and don't get me started on the trusts which don't encrypt their discs...!
@AC - What, "pity me I'm a linux user" it's "I'm a Linux user, you're going to take notice of me because I'm a paying customer and I'm proud to be a penguin." (and in case someone decides to repeat the same tired argument that I'm not a customer, yes I'm a customer because I have donated to Canonical and will likely do so again, even though I HATE unity with a passion) I also pay my taxes, and I paid for a "cd contining my images"
@Hyphen, It has everything to do with the NHS, indeed all public/government services that risk putting themselves in a position where they could have to spend public money because they made the wrong decisions; for which there is PLENTY of prior experience to warn them about the path they are going down.
Proprietary lock in of data has been a tool used by the IT industry in order to stop their customers from jumping ship, for decades.
And after all these decades, no one in authority seems to have learned or passed on these messages.
If managers are taught things these days like the likelyhood of a business going bust if it has a data disaster, then how much of a disaster can you get if you're data is held hostage by an IT company who have a gun to your head.
So you folks can down thumb me to hell and back if you like.
I want data that I can access - in an open format. I want managers who buy services and don't think about data lock-in, open document standards and future access, to be taken out and shot at dawn.
Flame suit donned - let me have it.
And by the way, when it comes to Jo public, and I'm included in that, being outside the health sphere; how the heck do you expect me to find google articles on common scanner file formats if I don't know the terminology that I'm searching for? What the heck else am I going to do, if not go back to the hospital and say, "oi, these aren't the pictures I paid for, and the instructions on the CD say I require IE 5, what kind of game do you think you're playing?" to then get a reply something along the lines of, "Oh, that's not a problem. They're in DICOM format and you need to web search for 'fos mri' " (in fact, I don't think they were in DICOM format, which is why they couldn't point me to any reading software that would work under Linux. I'd have to check my e-mails tonight.)
I mean, I know I'm a techie, but do you REALLY expect me to have known MRI format terminology? Come on.
Actually I was worried when the NHS sent me a DVD of my right knee MRI scan as they refered to it as a disc of images.... being sent in the post? Surely they should be password protected as it contains my medical details...?
I got the DVD 2 working days after I posted the cheque to the PACS people, and on that morning they called me on the number on file to advise of the password. Put in the DVD, ran the app on the root of the disc and entered in the password provided. Voila - images, slices, metadata, consultant reports etc. all in this non-standard app. If I want to share the contents I can press export and dump into PDF, JPG, and even a MHTML flat file.
For 95% of computer users, if they have the same experience as me then they will be very, very satisfied. The default action of the NHS when you ask for a copy of data is to:
Process the request within 2 working days
Provide the data required AND associated reports (if you asked or not!)
Encrypt said data
Post media to details on the request form
Contact patient the following day in the morning with the password
95% of people can put in the disc, run the app and go
Should the patient want a standard / open output they can if they wish
And this is bad? Honestly, some people are never satisfied.... Put on a trial VM of Win7 on your linux box and spit your pics out in a nice JPEG (without the password, or medical metadata). Or ask the nice people at PACS if they can export the pics to you - although I'd imagine they'll make you sign a disclaimer acknowleding the risks of unencrypted data etc....
Hey MIchelle I have an old 12" Powerbook G4 with OSX 10.4 with IE 5.2.3 for Mac onboard if you really want to look at your old scans and a lots less than the £2K cost of the main article. It also has a functioning DVD drive.
I still have it as it has a POTS modem built in which can be useful down here on the Dark Continent (and in some parts of the UK). It will be in the UK in two weeks time (me as well). Love to sell it as I need funds for a 27" iMac!
If you don't want to own legacy stuff I could e-mail the IE 5 pkg which was a free download back in 2001 but you need access to a Mac running OSX 10.1 to 10.6 for it to work. I'm sure Vulture central will forward me your address if you ask them nicely.
Speaking of legacy stuff any takers for a plastic IBM flowchart stencil COBOL vintage?
@The Original Steve
I wish I had the same experience as you. Mine wasn't even password protected.
Thanks for the offer. I did eventually get an old virtual machine image running WIndows which I was able to access the images with. There were very few slices and the thing wasn't even encrypted.
I was hoping that my complaint might make the management wake up and think; I spelled some things out for them, but they just didn't seem to care. They more or less laughed at open standards and I can't help but shudder at what will happen in the future; the machinery was already old at that point.
They couldn't even export the images out of the format that they were stored in. Fortunately for me, I'm in reasonably good health and my records to date don't contain that much. People whos future treatment relies on past scans, however, are in a very different position.
Good luck selling the Powerbook. I've already got a tablet, a netbook and a Tecra laptop that I don't use to their full capacity; I can't justify spend on another machine I don't use! So very different from The Original Steve's experience.
As fro the stencil, thanks for the offer but I already have one :-)
Why on earth does he think he has the "right" to view his long archived medical scan's?
If he doesnt like the service provided by the NHS then he can bog off to Private Healthcare land. Especially if anything medically relevant was noted in the written transcript.
Why should those of us who fund the NHS have our taxes pissed up the wall by this joker?
Its amazing how many rights people want to invent these days. There are few if any "unalienable" rights - US constitution not withstanding.
You forget the basic principle of the DPA, records should not be kept longer than necessary. The records in this case are - for medical purposes at least - useless so really the health authority should have deleted them once a newer scan became available and the scans on record became obsolete.
This rule, like the rest of the DPA, only applies to personal/identifiable data. Strictly speaking an obsolete scan is not necessarily personal data, and may well have further use in terms of research etc. As long as it's anonymised, the trust can do whatever they want with it.
Tough, obviously, in this case, there's been a DPA breach regardless as the scan is obviously identifiable (they know it's his) but it has not been appropriately maintained (it's inaccessible). Naughty trust.
AC due to being at work.
All of us have a right to copies of any of our own data less than ten years old held by anybody.
And the NHS have an obligation to maintain records for a certain period too.
Claiming that they DID maintain the records, but not a way to read them, is just finding a way around the law.
It may be "right" of him to let them off in this instance, ratehr than waste public funds, but he does have a "right" to the data too.
Of course he has the right to see his medical records, they are HIS medical records, data protection act say they have to hand them over...
That they don't store them on a standard drive is their stupid fault... and I would suggest they buy the drive, get some software and start copying over to some other form of backup medium that will NOT go out of production, i.e. a tape drive of some kind!
> Why on earth does he think he has the "right" to view his long archived medical scan's?
As others have said, they are "his* records.
I would actually go even further: since he is presumably a tax payer, the records really do belong to him. He paid for them to be produced through his taxes. The hospital is being paid through his taxes, he is the customer.
I think a lot of people seem to be operating under the misconception that this hospital is being forced to do him a favour funded from their own pockets.
To be fair though, it sounds like the hospital is not outright refusing to cough the images up. They are just raising the cost issue. They are probably to blame for the mess in the first place by using proprietary formats and hardware, many others are in the same boat.
This is a bit of a non-story.
I worked in medical imaging at a London teaching hospital several years ago.
I was responsible for storign and retrieving images om magneto-optical disk.
Remember folks - on those days a gigabit hard drive was considered big, and medical images contain a lot of data.
I haven't seen a drive like that in years, and I doubt they are still available.
Actually, that is an argument for X-ray film - film is an acquisition medium (ie you stick it behind the patient when you do the X-ray), it is a viewing medium - all you need is a bright light to view it and it is an archive medium - stick it in an envelope and you still have it eyars later.
Having sadi that, I am not doubting that PACS is the best way to do things (and I worked on a pioneering UK PACS system at the Hammersmith too).
Film is bulky and expensive because of the use of silver, and there is only one copy of it, so has to be physically shifted around for more than one doctor to have a look at it.
I'm sure copying X-rays is possible. Remember however that you would need a high resolution scanner, with a good dynamic range - these things are used to look for subtle changes for diagnostic purposes. Such as spotting breast cancer.
However, having worked in a hospital in medical imaging I have never seen X-rays being copied.
I could be wrong - this may be routine these days.
This is one of the arguments for PACS - in a hospital if you are running a clinic a request is made for patients notes and X-ray films which are brought up on a trolley.
Notes and X-rays quite often go missing, and what happens if a patient is attending two clinics on the same day?
with a digital PACS system the images are available in multiple locations.
Regarding your query about patients attending two clinics on the same day, I have a friend who is a cancer patient, with the NHS, she has to visit three diferent hospitals for seperate scans and treatment, often on the same day and none of them talk to each other, so when she mentions results or medication given by one hospital, that is usually the first the other knows about them. And this with a supposedly central NHS database, that all can access for patient info..
At least you have patient confidentiality, after all even your doctor can't see your records! That is superb data protection. And FYI it has been like that forever as far as she knows, ( she used to be a nurse) each hospital used to store documents in their own formats. Obviously that doesn't happen now (must remove tongue from cheek later)
You must remember that the NHS has been split into "commissioning" and "provider" organisations. The commissioners (CCGs) have the money, the providers (hospital trusts etc) compete with each other to get the commissioner's money. Under the brave new 'lets apply competition law to the NHS' system that is like recommending Ford ask to borrow some kit from Renault to get them out of a hole. It would be better for Renault to watch them pay or get fined by the ICO.
They did, no one has the obsolete gear anymore. Personally I think the guy is being unreasonable, there is no medical reason for him requiring the image other than for his own personal records, doctors have told him the image is useless and the only useful ones are current ones, but he'd rather take away money from patient care to get this image or take them to court taking MORE money from patient care.
Never mind that the cardiology technicians and consultants have told him the data is of no medical use, he is a PATIENT, godsdamnit, and he knows more about the finer points of cardiac function than some asshole that spent best part of a decade qualifying as a doctor and he says "fetch".
If "clinical decisions and treatment are based on diagnostic tests undertaken at the time of admittance" then why are they storing this guys heart scan at all, escpecially since they can't view it anyway, and why do they bother to keep patient records that go back to birth?
You are slightly confusing patient records and imaging data.
I'm not sure about patient records, but there is no requirement to keep medical images since birth.
If I recall correctly, for children images have to be kept till they are eighteen.
For adults it is something like ten years.
I'm not sure of the exact number of years - as I said it is some time sonce I worked in medical imaging.
Think about it - do you REALLY expect hospitals to maintain warehouses with the X-ray films your eighty year old granny had taken when she was a child?
Conventional X-ray films were reoutinely recycled for their silver content.
Medical images are routinely stored digitally in systems called PACS. You can take X-rays directly on digital plates, and CT and MRI data are of course digital anyway.
HOWEVER if we restrict ourselves to patient notes for this discussion, have you ever SEEN a patient records department? I have. Do you know what a Lloyd-George envelope is?
When I worked at a leadign London teaching hospital well over ten eyars ago, the head of medical physics asked me to get involved in a project some bright spark had to scan int he entire hospitals medical records.
I ran a mile.
I doubt a project like that has ever successfully been done.
1) This is why one should never use proprietary formats for data.
2) Why wasn't the data moved to a different drive/system? Surely there's a good chance a doctor might want to see it? And what about other patient's records? Did they learn nothing from NASA?
Find the manager's responsible for allowing this to happen and fire them. Summary dismissal, grounds of professional negligence and bringing the NHS into disrepute.
Easy for you to say that Anonymopus coward.
" This is why one should never use proprietary formats for data."
Have you ever read data off a brain scanner?
I have - I have worked in medical imaging departments, and for a nuclear medicine company which maintained scanners.
I ahve worked with Siemens and Philips gear, and with GE probably but I forget that.
So you buy a million pound brain scanner, dig out your basement, install it, install a cyclotron next door.
So you're then going to say to the doctors that they can't see the data off it because CTI Siemens provide the data in their own format?
Of course, hat you do is translate that data into other formats.
there is a standard for nuclear medicine images, which I worked with.
Similarly DICOM for other imaging modalities.
"So you're then going to say to the doctors that they can't see the data off it because CTI Siemens provide the data in their own format?"
Well, one would have required the Siemens kit to spit things out in an open format before one bought it. Not requiring this was the first phase of the problem. But it could still be rectified after the fact...
"Of course, [what] you do is translate that data into other formats."
Exactly. So why didn't these idiots do so?
"[There] is a standard for nuclear medicine images, which I worked with.
Similarly DICOM for other imaging modalities."
So you agree that these open formats exist, are you also agreeing that the NHS were negligent in not storing this data in one of these formats (and on media they can read)?
It might not such a big deal that whilst the data is being processed/used it is proprietary, but when it goes to archive is should be in an open format. Why? Go ask NASA.
All you have done is spout about your nuclear imaging (a claim that cannot be verified) and not really added anything.
@AC - if you'd bothered to check you would have found out that the images were stored in DICOM Format, there is no question about that. The issue is that the machine to read the 9Gb Optical Discs is no longer available to read the discs to obtain the images. The machine in question being a Sonos 550 Phillips Echocardiography machine.
It's rather akin to storing stuff on a 1.44Mb floppy and trying to read the data on a modern Mac, you can't do it unless you get a bit of 3rd party hardware which can read the 1.44Mb floppy drive. Heck, I got a box full of 1.44's I can't use because I have no FDD but am I willing to spend money on buying an FDD that may only get 1 use? Nope.
"All you have done is spout about your nuclear imaging (a claim that cannot be verified)"
Errr... it is a claim that certainly can be verified.
Damned if I'm going to bother to name the company on here, but its on my CV.
And I worked in the medical physics department at a London teaching hospital for several years, and was the computer manager at a brain imaging unit it the same hospital.
If you think thats unverifaible, so be it.
I'm sure someone did but it was probably seen as too expensive.
I would have thought Phillips could direct them to one of their customers who still have a drive under maintenance and possibly a data restoration company who invested in such a drive?
it does however seem the trust have shot themselves in the foot and the patient is busily passing them more cartridges.
If there was no medical reason to update old images, why would they do this? Why go to the time and expense of converting something that is of no use?
If this is the first time someone has come forward to request a copy of an image of this type in however many years since the kit stopped being used, and isn't really doing so with a valid medical reason, then hasn't the decision not to maintain this archive in a currently readable format been the correct one?
(proprietery formats still suck, but a mistake from the past that doesn't need correcting doesn't really matter)
Eight years compared to what? Installing Windows XP today and expecting 8 years of support isn't going to happen. It's likely that the equipment the image was stored on should have been binned far more then eight years ago, but due to things like hospital bureaucracy and budgets was being used far longer then it should have.
It's highly likely that the DSR-TIFF format software is far older then 8 years too, as I don't think new software to write it has been developed (by the manufacture) for at least 12 years.
Lastly, conversion is expensive, doubly so when you're switching out lots of MO disks. It's not like a huge SAN where you can script it and let it run, lots of manual labor.
Most likely because they have to keep it (at least in the US). I'd guess a catch-22 situation where they have to have the records, but the costs of conversion are higher then just keeping them in records storage. The vast majority of these records will never be touched again, so after some long amount of time they'll just be shredded.
'The Red Shoes' was recently fully digitally restored for posterity - having spent a shedload of cash, they wanted it to be available in the future too. Digital formats have this horrible habit of being unreadable every 20 years or so, so they archived back to celluloid! You can pick up a reel of film, see it contains a sequence of images, and fudge a way to play them back. Keep the reel cool and dry enough and it'll last for pretty much ever - another civilization might get to appreciate The Red Shoes where they will never see The Phantom Menace if the chain of data and the ability to process it is ever broken - especially likely with DRM on everything toboot.
...as far as digital data preservation goes.
The US Library of Congress has some good guidance at http://www.digitalpreservation.gov/, but the proliferation of propriatary data formats across multiple industires doesn't bode well for future accessibility of data.
First of all, it might be the case that he thinks that doctors have misdiagnosed his heart problems. Perhaps, for several years, they told him he was suffering from one thing and then, suddenly, said he was suffering from another thing. Perhaps he wants to take the old scan to an expert to see what he actually is suffering from and whether whatever it is could be spotted way back when.
It might also be the case that doctors have been moving goalposts, as they have done with blood pressure. Perhaps there is nothing much wrong with him, but doctors thought there was and have been treating him accordingly, but now the doctors say that things are worse. He is wondering about this and wants the old scan so that it can be compared with recent scans to see whether his condition has indeed deteriorated or whether the doctors are simply moving the goalposts in their quest to 'perfect' his body.
These are just two reasons why the scan could be of use. Of course, it would be very convenient for the NHS to just lose old scans and other information so that no-one could ever prove that doctors had got it wrong. It would be very convenient if no-one had access to their old records in case the NHS came out looking something less than the paragon of perfection that it portrays itself as. He who controls the past, you know, controls the future, and that includes those who can 'disappear' the past as well as those who re-write it.
Has anyone out there (apart from JLH) ever looked at medical equipment? All of it is overpriced and proprietary. You think the Microsoft Office file formats offer only lock-in? You've obviously never looked at data storage and retrieval in the healthcare field!
And you don't tell these guys what format you want your data in, they tell you what you're going to get. (as an aside, the UK healthcare requirements are pretty much ignored by the Big Boys anyway)
As for data migration, who's going to migrate the data? If you've changed suppliers, the outgoing one has no interest in you, and the incoming one doesn't need the hassle. (They usually agree to do it, then come back later and tell you that "despite their best efforts the migration has failed" They generally blame the outgoing guys for not being transparent about their formats.)
As regards test results (ie images etc). These are transient because it's generally not the result that's important, it's the interpretation of the result, in conjunction with other diagnostic information that gives your doctor the information he/she needs to direct your healthcare. I can think of no situation where a 4 year-old image (raw data in other words) can inform a patient's current treatment.
"As regards test results (ie images etc). These are transient because it's generally not the result that's important, it's the interpretation of the result, in conjunction with other diagnostic information that gives your doctor the information he/she needs to direct your healthcare."
Absolutely agree. Very well put.
What we're missing here is the process - a scan is ordered by a clinician, patient turns up at department, and a lot of digital data is gathered. That dat is kept on disk (in this case on a MO disk, if I remember the capacity being a whopping 80 Megabytes).
The data is kept in digital fomat so that a consultant can view it, and interpret it. The consultant will dictate a report with his/her findings. And remember it takes years to train as a radiologist.
It is the report which is the important part, and this should be what you think of as your "medical records" - and should be kept.
I don't see what the fuss is about keeping all that raw data - there isn't the storage room for it, or the need.
The data IS kept for a reasonable amount of time - after all you might be in hospital for months, or be undergoing (say) a cancer treatment. It is useful to compare the tumour size pre-and-post treatment (I'm sure there are many more examples). But eight years later? No.
>Has anyone out there (apart from JLH) ever looked at medical equipment?
Nah, except for when I worked in a broom cupboard (ostensibly the mail room) of an NHS department, there was a massive horizontally mounted tape, probably 1/2", machine of some description, competing for space with my mail sacks. No idea what it was, but the top was cast from a fair bit of alu/mag alloy. They hadn't used it for a long time, nor had they slung it out.
Though yesterday, when looking at the 'Pixar names building after Steve Jobs' article, I read up again on what Jobs bought Pixar for- their hardware division. Apparently they only sold 300 machines at around $150,000 a piece (not including a required $30,000 SGI workstation), aimed at doctors with $2million dollar medical scanners. However, doctors were trained to analyse 2D images and could be sued if they didn't follow this procedure... I got the impression that its storage and formats would be proprietary.
Medical equipment is where you take a normal device or software and randomly assign X10 to X100 value to it. I will say some of the latest software a clinic I admin uses runs on regular MSSQL, exports to PDF and keeps image records in regular TIFF format. I do believe that I could even migrate data off of it.
I think the biggest reason this data is being kept for long periods of time, even though there is little medical use for it, is lawsuits.
"I can think of no situation where a 4 year-old image (raw data in other words) can inform a patient's current treatment"
OK case study here for your education. A friend of mine had a problem with her spine that needed surgery about 10 years ago. About 2 years ago she started to have further issues with her back, and of course they took more scans which showed the original operation details but they needed to check whether there had been significant changes to her spine & surrounding areas in the intervening 8 years - this could help determine whether her new problems really were new, or related to the earlier problem (e.g. scar tissue that had developed over time).
There were NO electronic copies of scans from the earlier injury & operation, and eventually after weeks of phoning & chasing (which she had to do, as her local NHS trust reckoned they had no means of doing this) she eventually managed to source a rather blurry copy of her original scans which were emailed to her. This showed that what they had thought was the old problem deteriorating WASN'T the cause of her new problems, and saved her having a major operation on her spine that would have had a 20% chance of leaving her paralysed from the waist down.
So yes there are some very important situations where old images can be useful.
It seems that the trust has the information, but not a means of expressing it in intelligible form without reacquiring some dedicated kit, or else finding a trusted third party to perform the conversion. I would be surprised if the trust did not argue that this constituted disproportionate effort, meaning that it does not need to supply the data:
s8(2), Data Protection Act 1998:
The obligation imposed by section 7(1)(c)(i) [to have communicated to the data subject in an intelligible form the information constituting any personal data of which that individual is the data subject] must be complied with by supplying the data subject with a copy of the information in permanent form unless—
(a)the supply of such a copy is not possible or would involve disproportionate effort
The Information Commissioner's Office has a reasonably concise guide on applying this test: http://www.ico.gov.uk/~/media/documents/library/Data_Protection/Detailed_specialist_guides/disproportionate_effort.pdf
it does not need to supply the data
It could supply a copy of the data, perhaps, just not in an intelligible form — it perhaps depends on whether the storage medium has more than one patients' records on it, and whether it has any way of duplicating the disc without the specialist machine.
This is what our film archives looked like in 2004.
This is just one row of a dozen in the basement. <a href="http://imgur.com/egUoT"><img src="http://i.imgur.com/egUoT.jpg" title="Film Fileroom 2004" alt="" /></a> Those films have been digitized in needed and recycled.
There was another room that held mammography films. Those are now held off site. If they are needed, they are brought in, digitized and then recycled.
This was our tape jukebox that we used up until 2 years ago. <a href="http://imgur.com/QjTg9"><img src="http://i.imgur.com/QjTg9.jpg" title="AIT3 Jukebox 2004" alt="" /></a>
It held about 35 TB of data. That was migrated to the new system. (Philips iSite/Intellispace)
And how many thousand disks do they have? MO disks aren't very high capacity so a huge amount of manual labor would be needed.
Also, since this evidently is the only (or only person would made a fuss about) person that needed the data in recent history, evidently the data isn't that requested.
Hardly; the scans are worthless. They're medically irrelevant. They could digitise the rest of the records, but there's no point. This gentleman is basically requesting them as an exercise in Being An Arse, an activity anyone in customer-facing services should be well familiar with. If the records were to be digitised for research purposes or to settle a legal case (both extremely unlikely), then the £2000 expenditure for the gear would be worthwhile and a small fraction of the total costs of either of those situations.
In this case a trust would be paying £2,000 (that's a nurse for a month, for anyone keeping track), plus the secondary costs of verification/installation/etc., just to let one single customer see what his heart looked like eight years ago. Anyone who's had a kid or two in the last decade should be able to tell you how much ultrasound has come on in that time.
Is the NHS a happy world where everyone's on the ball and makes no mistakes
The Trust involved paid out millions last year in compensation
They SHOULD digitise the records. There are 200 disks with scans up to 2008. Whats more they should have done it before. NHS guidance in 2008 required planned migration and continued readability of data. Sometimes old medical records help unmask gross negligence sometimes affecting many patients, often many years later. If a patient wanted a 2nd opinion would you say the NHS should have to pay for the conversion at a cost of 2000 ?
What about the thing no one seems to have thought about.... intertnal budgets for repair of failed hardware (the hardware failed apparently in this case ) and also budgets for upgrading and migrating to platforms
What is more the £2000 is way more than the equipment would cost. It can be acquired for a few hundred. You can acquire a working ultrasound machine for less than 2 K
The funny thing is they are right... old records are not that important. You need to take a history, examine the patient and take new tests when needed. But the basis of the Connecting for Health was that old records are essential. Rubbish, all they can do is make you lazy and make repeat past errors. Umteen billion wasted on a false premise.
Biting the hand that feeds IT © 1998–2021