Twitter Updates

    follow me on Twitter

    26 November 2005

    Study shows PC monitors equivalent to PACS workstation for detecting wrist fractures

    A recent study reports equivalent performance for detection of wrist fractures between a 17" PC monitor with a web browser, and a 2 MP diagnostic quality Barco monitor with a PACS workstation.

    The authors concluded that:
    there is no difference in the accuracy of observer performance in the detection of wrist fractures on digital radiographs with a standard PC compared with that with a high-quality workstation. In our opinion, important factors in achievement of this result are the use of image magnification to employ the maximum spatial resolution of the original digital image and the use of an environment with low ambient light.

    In the UK a 17" PC monitor of the type they describe costs about 300 GBP, compared to about 2000 GBP for the 2MP Barco. The PACS workstation probably costs another 10 000 GBP at least, whereas web browser displays run off standard PCs.

    Of course for large volume reporting, high resolution monitors make sense, otherwise you'd have to zoom everything right up and pan around a lot, but in other situations, e.g. casualty departments, PC monitors are probably fine (provided the viewing area is dark, which is another story)

    Personal Computer versus Workstation Display: Observer Performance in Detection of Wrist Fractures on Digital Radiographs
    Anthony J. Doyle,, James Le Fevre, and Graeme D. Anderson
    Radiology Dec 2005;237:872-877

    12 November 2005

    Ultrasound poor (herniography good) at detecting clinically occult hernias

    In this study, ultrasound had a sensitivity of 29% and specificity of 90% compared with the herniography. Correlation with surgical findings showed ultrasound to have a sensitivity of 33% and a specificity of 100%.

    I'm not surprised about the ultrasound results. I can't comment on the herniography part, since I've only ever done one (many years ago), but maybe I'll need to start practising.

    Adeeb Alam, Colin Nice and Raman Uberoi
    The accuracy of ultrasound in the diagnosis of clinically occult groin hernias in adults
    European Radiology 2005 (Dec) 15 (12): 2457 - 2461

    26 October 2005

    Fantasy Health Minister: I'm on the high scorers list

    As described in a previous post, it's an online game where you get to be the Health Minister and select from a range of policy decisions.

    I scored 8851 on my last attempt, which made me the seventh highest scorer. Note that I got the health of the nation to 96%, but some of the higher scorers did not do as well. This is because the score also depends on popularity, and you have to please everyone, including the private sector.

    At present, I'm the only Authoritarian on the high scorers' list.
    (I'm "dr_lee_xray"' in case you were wondering).
    Most of the others are Idealists...Pah!!

    Fantasy Health Minister

    16 October 2005

    CT vs MR cholangiography for biliary calculi

    In our hospital we tend to use CT cholangiography (CTCh) more often than than MR cholangiopancreatography (MRCP) for the non-invasive detection of common bile duct stones because CT is more readily available, and we find CTCh a simpler and more robust test. Of course MRCP has no radiation and does not need any injections, but how do the two tests compare in terms of diagnostic accuracy?

    A recent paper addresses this question. 40 patients had both CTCh and MRCP, with the following results:

    For detection of choledochal stones the two tests were comparable:
    CTCh: sens 87%, spec 96%
    MRCP: sens 80%, spec 88%

    For detection of gallstones MRCP was a bit better,
    CTCh: sens 78%, spec 100%
    MRCP: sens 94%, spec 88%
    (The reduced sensitivity of CTCh was due to non-opacification of the gall bladder due a stone being stuck in the gall bladder neck)

    CTCh demonstrated bile leakage in one patient, and air in the biliary tree in two patients, which were not detected on MRCP.
    CTCh also has the advantage over MRCP in demonstrating anastomotic patency, since opacification in CTCh reflects flow, whereas MRCP only demonstrates the presence of fluid.

    I've never been able to figure out why are there so few papers on CTCh compared to MRCP. According to the article, Biliscopin (the contrast agent used for CTCh) is not approved for use in the USA. I didn't realise this but perhaps that's part of the reason. (As an aside, it's odd how some substances are permitted in one country but not another, for no obvious reason. Buscopan, for instance, is not licensed for use in the USA either).

    Okada M, Fukada J, Toya K, Ito R, Ohashi T, Yorozu A.
    The value of drip infusion cholangiography using multidetector-row helical CT in patients with choledocholithiasis.
    Eur Radiol. 2005 Oct;15(10):2140-5

    9 October 2005

    Want to run the National Health Service?

    Here's an online game where you get to be the Health Minister and select from a range of policy decisions, which in turn have an effect on the health of the Nation, your budget, and your popularity.

    On my last attempt I got the health of the nation up to 96% (you start at 43%), kept within budget, and increased my popularity. You also get an indication of your political complexion. Mine varies but so far it's usually 'authoritarian', occasionally 'idealist' or 'realist', but never 'liberal'.

    Fantasy Health Minister

    27 September 2005

    Hospital to install OsiriX: open source diagnostic workstation

    A small private hospital where I do a bit of work will be buying an iMac to run OsiriX, an open source diagnostic workstation developed at UCLA and the University of Geneva.

    It's not as sophisticated or as fast as the expensive Vitrea workstations we use in the big NHS hospital, but those cost about £35000, compared to £2000 for the iMac, and would not be financially viable for a small hospital with a relatively low volume of work.

    An iMac with 2G of RAM will be adequate for our requirements and the software is downloaded for free from the OsiriX website, so it's excellent value for money.

    Incidentally, one commercial company did offer us an outrageously expensive workstation with no multiplanar or 3D capability, that could not even read the discs produced by a CT scanner from their own company.

    Upgrades for OsiriX are frequent and free, the instruction manuals are on wiki, and excellent applications support is provided by the online community on a yahoo discussion group.

    As an open source product, anyone with the skills and interest can contribute to its development, and many people do.

    You can also plug your iPod straight into your Mac workstation and save files and pictures to it. (I'm not sure how radiology images on an iPod photo are of any real use, but I guess some people might think they look cute).
    Or download music from iTunes more easily than from your home PC .... oooops!

    For a lot more money, you can buy the most expensive available PowerMac, with lots and lots of RAM, and install a VolumePro board from TeraRecon, for which a special version of Osirix is available. This is meant to provide much better 3D performance but I haven't seen it myself. It's still cheaper than the commercial products though.

    The public healthcare sector might like to keep an eye on this, especially now that many state hospitals are feeling the squeeze financially. OsiriX is NOT a replacement for commercial 3D workstations, but it does have a lot of functionality at a low price, and is in continuous development.

    I am a PC user. I do not own a Mac. My mother has a Mac: I think it's fine, and I have nothing against Macs. I do not have an iPod either: I have a Zen Neeon, which does not work with Macs.

    Definition of Open Source
    Vitrea Diagnostic Workstation from Vital Images Inc.
    Zen Neeon Cheaper and better than the iPod
    Hospital bosses ordered to cut debts The Guardian 16 Sep 2005
    One third of the population mistakenly believes that an anaesthetist is not a doctor, according to a BBC Radio 4 programme I am listening to right now.

    I bet that's not the case with radiologists :-)

    24 September 2005

    Ultrasound instead of chest x-rays for nasogastric tube localisation

    I was giving a talk to our staff on "looking for medical information on the internet". There was a live demo on using PubMed, and the topic of nasogastric tubes was suggested by someone who clearly found the whole business of doing chest x-rays for localisation a bit tedious.

    This is an article we found:

    Sonography as an alternative to radiography for nasogastric feeding tube location.
    Vigneau C, Baudel JL, Guidet B, Offenstadt G, Maury E
    Intensive Care Med. 2005 Sep 20; [Epub ahead of print]
    Bedside sonography performed by nonradiologists is a sensitive method for confirming the position of weighted-tip feeding nasogastric feeding tubes. It is more rapid than conventional radiography and can easily be taught to ICU physicians. Conventional radiography could be reserved for cases in which sonography is inconclusive.

    Whoopee! Our ICU physicians have said they want to get their own ultrasound machine.

    We'll see.

    Making Gadolinium DTPA in your kitchen

    According to a lecturer (a chemist) at recent MRI course I attended, Gadolinium DTPA (the fabulously expensive contrast agent used in MRI scanning) is a very simple compound that any first year chemistry student could make at home.

    In China, they are not too concerned with copyrights and patents, I was told, so the pharmaceutical companies have started factories there to protect themselves. Enough said.

    My sister studied chemistry at university, but unfortunately that was some time ago and she is unable to assist our department in obtaining cheap supplies.

    15 April 2005

    Learning from ER: embolectomy for stroke

    Last night on ER: Season 11 Episode 15: Alone in a Crowd
    A young mother of three (Cynthia Nixon of Sex in the City) is brought in with symptoms of a potentially fatal stroke .....

    Its too late for thrombolysis but the interventional radiologist carries out an embolectomy using a new technique ...

    I'd not heard of this one before, so I turned to Google.
    Here's the result of one Google search.

    Thrombolysis nust be initiated within three hours of onset of an acute stroke, but embolectomy with the Merci retrieval system is effective up to 8 hours after onset of stroke symptoms.

    A description of the technique can be found here.
    For a video, click here.
    The vendor's site is here.


    Mechanical thrombectomy of the internal carotid artery and middle cerebral arteries for acute stroke by using the retriever device.
    Martinez H, Zoarski GH, Obuchowski AM, Stallmayer MJ, Papangelou A, Airan-Javia S
    AJNR Am J Neuroradiol. 2004 Nov-Dec;25(10):1812-5

    MERCI 1: a phase 1 study of Mechanical Embolus Removal in Cerebral Ischemia.
    Gobin YP, Starkman S, Duckwiler GR, Grobelny T, Kidwell CS, Jahan R, Pile-Spellman J, Segal A, Vinuela F, Saver JL.
    Stroke. 2004 Dec;35(12):2848-54. Epub 2004 Oct 28

    And a case report published in 2000 claiming to be 'the first reported use of a snare to remove clot in the setting of thromboembolic stroke'
    Transcatheter snare removal of acute middle cerebral artery thromboembolism: technical case report.
    Chopko BW, Kerber C, Wong W, Georgy B.
    Neurosurgery. 2000 Jun;46(6):1529-31

    Other things I've learnt from ER:

    Different ways of holding and positioning a Sonosite portable ultrasound machine.

    How to rule out a mandibular fracture and avoid doing an x-ray (get the patient to bite a tongue depressor, then twang it; if it doesn't hurt, there's no fracture).

    I'll be submitting my ER viewing for CME credits.

    10 April 2005

    Farewell to intravenous urography

    We've now stopped doing IVUs in my department!

    In place of IVUs, we do noncontrast CT for investigation of urinary stones, and CT urography, where indicated, for haematuria. (Indications for ultrasound are, for now, unchanged).

    CT urography is a better test, and more sensitive for detecting stones, urothelial lesions and tumours. The main drawback is radiation dose. However, for stone disease, the dose from a noncontrast CT can be brought to levels similar to that of an IVU using a low-dose technique. For patients with haematuria at high risk for cancer, many of them would have ended up with a CT anyway, and if not I think the extra dose is justified by improved diagnostic confidence and accuracy. As they say, if it was me or one of my relatives .....

    The technique is relatively simple, and it is easier to get consistent results than with a conventional IVU. It takes less time, and a morning's work in the IVU room can be completed in the CT scanner in about an hour. One extra radiographer has now been allocated to the CT scanner (in place of the one who used to do IVUs), which has helped to resolve some of our staffing problems. The radiographers are delighted because they didn't like doing IVUs. The urologists are pleased as well.

    Our CT urography protocol, on a Toshiba Aquilion 16, consists of a low dose noncontrast run to detect stones, followed by post-contrast imaging using a split blous technique which produces a combined nephrographic and excretory phase image. Imaging is done supine, and repeated in the prone position if ureteric opacification is inadequate.

    9 April 2005

    Whole-body CT screening

    A whole-body CT screening centre has been in operation at a town near mine, part of a chain called Lifescan. Patients pay between £300 (for a lung scan) and £750 (whole body CT with virtual colonoscopy) for CT scans to screen for unsuspected disease. Anything that is found that might possibly be significant then needs to be followed up. Sometimes early, unsuspected cancer is found . But often, the abnormalities turn out to be insignificant and of no consequence.

    This ends up being very expensive, and a team from the Massachusets General Hospital concluded that CT screening was not cost-effective.
    Cost-Effectiveness of Whole-Body CT Screening
    Beinfield, Wittenberg, Gazelle, Radiology 2005 Feb;234:415-422

    If the patient pays directly, this is not too much of a problem (at least for the health care providers), but medical insurers do not cover CT screening, and in many countries, the state ends up bearing the cost.

    Here is what others have said:

    Prof Adrian Dixon of Cambridge University, Warden of the Faculty of Clinical Radiology (the Royal College of Radiologists)
    Another topical issue is the question of walk-in screening centres offering to "save lives" by whole-body CT screening, sometimes for profit. In the UK, the National Screening Committee has suggested that the National Health Service should not offer such screening at present (Muir Gray, personal communication). Many sources suggest that a radiologist's private practice should mirror their NHS practice. The current radiological and radiographic workforce can barely cope with the existing workload, let alone such new ventures. It will be interesting to hear what other countries recommend. Several observers have noted that such walk-in centres are usually located in relation to educated health-conscious consumers who can afford such procedures! ....... Hard facts from primary research are urgently needed. But I cannot see too many grant giving bodies rushing to assess what is currently viewed as a somewhat opportunistic and entrepreneurial endeavour.
    Whole-body CT health screening.
    Dixon AK. Br J Radiol. 2004 May; 77(917):370-1

    A group from the Mayo Clinic
    CT screening for lung cancer offers the possibility of reducing mortality from lung cancer. Our preliminary results do not support this possibility and may raise concerns that false-positive results and overdiagnosis could actually result in more harm than good .....
    Our data do not suggest a mortality benefit; whether CT for lung cancer meets the criteria for an effective screening test remains to be proved .....
    Our preliminary mortality results should cause physicians to pause and reexamine their positions if they are performing routine CT screening outside of a clinical trial.
    CT Screening for Lung Cancer:Five-year Prospective Experience
    Swensen et al. Radiology 2005 April ;235:259-265.

    A group from Stanford (with regard to the situation in the United States)
    Direct-to-consumer marketing of self-referred imaging services, in both print advertisements and informational brochures, fails to provide prospective consumers with comprehensive balanced information vital to informed autonomous decision making. Professional guidelines and oversight for advertising and promotion of these services are needed.
    Advertising, Patient Decision Making, and Self-referral for Computed Tomographic and Magnetic Resonance Imaging
    Illes et al. Archives of Internal Medicine, December 13, 2004; 164(22): 2415 - 2419.

    15 March 2005

    Could a computer report rheumatology clinic films, please?

    We radiologists just love reporting rheumatology clinic films, meticulously comparing one hand with another to see if there are new erosions..... yeah sure.

    Now that radiographers are reporting skeletal films, perhaps they might like to expand their role to include the rheumatology clinics. And when they get bored, perhaps this product will be ready for use:

    Computer aided diagnosis (CAD) in rheumatoid arthritis
    P. Peloschek, G. Langs, F. Kainberger, H. Bischof, W.G. Kropatsch, H. Imhof ECR 2005

    13 March 2005

    Digital Bone Age Atlas

    Our departmental copy of the Greulich and Pyle atlas always seems to go missing, and I had been wondering if a digital version was available. Perhaps someone, somewhere might have scanned it and saved it as a pdf (probably highly illegal).

    Then I saw this pocket-sized paperback at the ECR.

    Hand Bone Age A Digital Atlas of Skeletal Maturity
    V Gilsanz and O Ratib
    Springer 2005. Approx 110 pp. 90 illus. With CD-ROM formatted for PC, PDA version for PALM and Pocket PC. Softcover $79.95 ISBN 3-540-20951-4
    "This atlas integrates the key morphological features of ossification in the bones of the hand and wrist and provides idealized, sex- and age-specific images of skeletal development. This computer-generated set of images should serve as a reasonable alternative to the reference books currently available".

    It is pretty cheap, as far as illustrated medical books go, looks easier to use than the G&P atlas (less text), and I can't see any reason why we shouldn't use it. It does not seem to have appeared on the Sprnger website yet, but a Google search brought out an abstract on this project presented by the authors at RSNA 2003.

    Viennese film bags

    A selection of film bags from Viennese clinics. Click on thumbnails for larger images.

    9 March 2005

    Radiographers reporting CT colonography?

    In a multicentre European trial, after training with 50 cases, non-experts were pretty good compared to experts at polyp measurement and size categorisation on CT colonography.

    The non-experts, who had been trained on 50 cases, were radiologists and radiographers. The performance of non-expert radiologists and radiographers was about the same .

    ECR 3/7/2005, B-696 Polyp measurement and size categorisation by CT colonography: Agreement with colonoscopy and effect of observer experience
    D. Burling, &. ESGAR CT colonography study GROUP investigators; St. Mark's Hospital, London/GB

    Here is the abstract

    In Britain, we have radiographers trained to perform and report barium enemas, which they do to a high standard. Barium enemas will be obsolete in a few years' time.


    8 March 2005

    Spinzoo: Animations of MRI spins

    Do you find the drawings of MR spins hard to figure out? Have a look at these animations.

    Some people believe that spins are just a quantum mechanical property of nuclei. However, spins are very small but lovely animals. And they like magnetic fields.We have collected some spins that used to live in our MR scanner, and we put them into our spin zoo.During the spin zoo tour you can see different aspects of the spin life, and how they react on MR sequences.
    MR-Physik , Departement Medizinische RadiologieUniversitätsspital / Universität Basel

    5 March 2005

    ECR exhibit: CPU vs GPU for workstation graphics processing (????)

    3mensio is a workstation company founded by Frank Wessels, who is one of the co-founders of Applicare (who originally made RadWorks, which we still use for teleradiology to our neurosurgery centre).

    According to what he told me, current 3D workstations (Vitrea, Voxar, Leonardo, etc) utilise the computer's central processing unit, or CPU (e.g. Pentium Processor) for graphics processing. The 3mensio workstation, 3viseon (sic), utilises the graphics processing unit, or GPU, which I understand to be the graphics card. These need to be cheap because they are made for kids to play games with, so that the computer used for the workstation can also be very cheap (about $1000) . It then becomes feasible for orthopaedic and vascular surgeons to have a 3D workstations in their offices (and radiologists too, of course).

    You read the abstract from the ECR presentation here

    I really am not in a position to assess the technical side of things, but I had a play with the workstation, running on a standard PC with about 1G of RAM, and it was pretty good, and fast.

    It makes you wonder
    (a) why the other manufacturers have not adopted the same approach, if it is feasible
    (b) why Nintendo have not diversified into the medical imaging sector

    Worth keeping an eye on, I think.

    4 March 2005

    Cerefy Brain Atlas

    Something else I came across at the ECR. I just picked up a leaflet, as there was nobody around at the time to explain the product.

    If I have understood correctly, it is an electronic brain atlas that allows you to place anatomical labels on your CT or MRI scans automatically.

    Here's an article about it I found via Google.

    Some of the products are free and can be downloaded from the website .

    I'l have to try it out when I get back.

    Barium enemas are on the way out...

    It was only about 6 years ago at the Leeds GI Radiology course, that I attended a lecture by Frans-Thomas Fork, who is professor of radiology and endoscopy at Malmö, where he showed how you could pick up small polyps and other subtle lesions on barium enemas by a combination of good technique and attention to detail. It was pretty impressive stuff from a guy who was clearly very good.

    Today (Friday) at the ECR, in a lecture on imaging rectal cancer, he said that the barium enema should probably be discarded... It's just not sensitive enough for polyp detection compared to endoscopy or CT colonography.

    It probably has another five or ten years left, before being completely superseded by CT colonography (or even MR, who knows?).

    Something for the barium radiographers to consider. Enjoy it while it lasts.

    3 March 2005

    Off to the ECR

    I'm leaving for Vienna tonight, for the European Congress of Radiology.

    I'll be staying with my friend Danielle, who tells me that Vienna is colder than usual, with the wind chill factor making the effective temperature about -15 degrees Celsius. Recent weather forecasts seem a bit more optimistic (but still sub zero). I'm reminded about a story I heard about new arrivals in Siberia being supplied with mink underwear.