Tài liệu Grid Computing P41 ppt

Section 41.3 describes what this means, and why it is a fundamental requirement for numerous gridapplications, particularly in medical image analysis, and especially in mammography.. In

eDiamond: a Grid-enabled federated database of annotated

mammograms

Michael Brady,1 David Gavaghan,2 Andrew Simpson,3

Miguel Mulet Parada,3 and Ralph Highnam3

1Oxford University, Oxford, United Kingdom,2Computing Laboratory, Oxford, United

Kingdom,3Oxford Centre for Innovation, Oxford, United Kingdom

41.1 INTRODUCTION

This chapter introduces a project named eDiamond, which aims to develop a Grid-enabled

federated database of annotated mammograms, built at a number of sites (initially in theUnited Kingdom), and which ensures database consistency and reliable image processing

A key feature of eDiamond is that images are ‘standardised’ prior to storage Section 41.3

describes what this means, and why it is a fundamental requirement for numerous gridapplications, particularly in medical image analysis, and especially in mammography The

eDiamond database will be developed with two particular applications in mind:

teach-ing and supportteach-ing diagnosis There are several other applications for such a database,

as Section 41.4 discusses, which are the subject of related projects The remainder of

Grid Computing – Making the Global Infrastructure a Reality. Edited by F Berman, A Hey and G Fox

 2003 John Wiley & Sons, Ltd ISBN: 0-470-85319-0

this section discusses the ways in which information technology (IT) is impacting on

the provision of health care – a subject that in Europe is called Healthcare Informatics.

Section 41.2 outlines some of the issues concerning medical images, and then Section 41.3describes mammography as an important special case Section 41.4 is concerned with

medical image databases, as a prelude to the description in Section 41.5 of the

eDia-mond e-Science project Section 41.6 relates the eDiaeDia-mond project to a number of other

efforts currently under way, most notably the US NDMA project Finally, we draw some

conclusions in Section 41.7

All Western societies are confronting similar problems in providing effective healthcare

at an affordable cost, particularly as the baby boomer generation nears retirement, as thecost of litigation spirals, and as there is a continuing surge of developments in oftenexpensive pharmaceuticals and medical technologies Interestingly, IT is now regarded asthe key to meeting this challenge, unlike the situation as little as a decade ago when ITwas regarded as a part of the problem Of course, some of the reasons for this change inattitude to IT are generic, rather than being specific to healthcare:

• The massive and continuing increase in the power of affordable computing, and the sequent widespread use of PCs in the home, so that much of the population now regardcomputers and the Internet as aspects of modern living that are equally indispensable

con-as owning a car or a telephone;

• The miniaturisation of electronics, which have made computing devices ubiquitous, inphones and personal organisers;

• The rapid deployment of high-bandwidth communications, key for transmitting largeimages and other patient data between centres quickly;

• The development of the global network, increasingly transitioning from the Internet tothe Grid; and

• The design of methodologies that enable large, robust software systems to be developed,maintained and updated

In addition, there are a number of factors that contribute to the changed attitude to ITwhich are specific to healthcare:

• The increasing number of implementations of hospital information systems, includingelectronic medical records;

• The rapid uptake of Picture Archiving and Communication Systems (PACS) whichenable images and signals to be communicated and accessed at high bandwidth around

a hospital, enabling clinicians to store images and signals in databases and then to viewthem at whichever networked workstation that is most appropriate;

• Growing evidence that advanced decision support systems can have a dramatic impact

on the consistency and quality of care;

• Novel imaging and signalling systems (see Section 41.2), which provide new ways tosee inside the body, and to monitor disease processes non-invasively;

• Miniaturisation of mechatronic systems, which enable minimally invasive surgery, andwhich in turn benefits the patient by reducing recovery time and the risk of complica-tions, at the same time massively driving down costs for the health service provider;

• Digitisation of information, which means that the sites at which signals, images andother patient data are generated, analysed, and stored need not be the same, as increas-ingly they are not;1 and, by no means least;

• The increased familiarity with, and utilisation of, PCs by clinicians As little as fiveyears ago, few consultant physicians would use a PC in their normal workflow, nowalmost all do

Governments have recognised these benefits and have launched a succession of tiatives, for example, the UK Government’s widely publicised commitment to elec-tronic delivery of healthcare by 2008, and its National Cancer Plan, in which IT fea-tures strongly

ini-However, these technological developments have also highlighted a number of majorchallenges First, the increasing range of imaging modalities allied to fear of litigation,2

mean that clinicians are drowning in data We return to this point in Section 41.2 Second,

in some areas of medicine – most notably mammography – there are far fewer skilledclinicians than there is a need for As we point out in Section 41.3, this offers an opportu-nity for the Grid to contribute significantly to developing teleradiology in order to allowthe geographic separation of the skilled clinician from his/her less-skilled colleague andthat clinician’s patient whilst improving diagnostic capability

41.2 MEDICAL IMAGES

R¨ontgen’s discovery of X rays in the last decade of the Nineteenth Century was the first

of a continuing stream of technologies that enabled clinicians to see inside the body,without first opening the body up Since bones are calcium-rich, and since calcium atten-uates X rays about 26 times more strongly than soft tissues, X-radiographs were quicklyused to reveal the skeleton, in particular, to show fractures X rays are normally used

in transmission mode – the two-dimensional spatial distribution is recorded for a given(known) source flux A variety of reconstruction techniques, for example, based on theRadon transform, have been developed to combine a series of two-dimensional projectionimages taken from different directions (normally on a circular orbit) to form a three-dimensional ‘tomographic’ volume Computed Tomography (CT) is nowadays one of thetools most widely used in medicine Of course, X rays are intrinsically ionising radiation,

so in many applications the energy has to be very carefully controlled, kept as low aspossible, and passed through the body for as short a time as possible, with the inevitableresult that the signal-to-noise (SNR) of the image/volume is greatly reduced X rays ofthe appropriate energies were used increasingly from the 1930s to reveal the properties

of soft tissues, and from the 1960s onwards to discover small, non-palpable tumours forwhich the prognosis is very good This is most highly developed for mammography, to

1 This technological change, together with the spread of PACS systems, has provoked turf battles between different groups of medical specialists as to who ‘owns’ the patient at which stage of diagnosis and treatment The emergence of the Grid will further this restructuring.

2 It is estimated that fully 12% of malpractice suits filed in the USA concern mammography, with radiologists overwhelmingly

which we return in the next section; but it remains the case that X rays are inappropriatefor distinguishing many important classes of soft tissues, for example, white and greymatter in the brain.

The most exquisite images of soft tissue are currently produced using magnetic onance imaging (MRI), see Westbrook and Kaut [1] for a good introduction to MRI.However, to date, no pulse sequence is capable of distinguishing cancerous tissue fromnormal tissue, except when using a contrast agent such as the paramagnetic chelate ofGadolinium, gadopentetate dimeglumine, abbreviated as DTPA In contrast-enhanced MRI

res-to detect breast cancer, the patient lies on her front with the breasts pendulous in a cial radio frequency (RF) receiver coil; one or more image volumes are taken prior

spe-to bolus injection of DTPA and then image volumes are taken as fast as possible, for

up to ten minutes In a typical clinical setting, this generates 12 image volumes, eachcomprising 24 slice images, each 256× 256pixels, a total of 18 MB per patient pervisit This is not large by medical imaging standards, certainly it is small compared

to mammography Contrast-enhanced MRI is important for detecting cancer because

it highlights the neoangeogenesis, a tangled mass of millions of micron-thick leakyblood vessels, grown by a tumour to feed its growth This is essentially physiolog-ical – functional – rather than anatomical – information [2] Nuclear medicine modali-ties such as positron-emission tomography (PET) and single photon emission computedtomography (SPECT) currently have the highest sensitivity and specificity for cancer,though PET remains relatively scarce, because of the associated capital and recurrentcosts, not least of which involve a cyclotron to produce the necessary quantities of radio-pharmaceuticals

Finally, in this very brief tour (see [3, 4] for more details about medical imaging), sound image analysis has seen major developments over the past decade, with Doppler,second harmonic, contrast agents, three-dimensional probes, and so on; but image quality,particularly for cancer, remains sufficiently poor to offset its price advantages

ultra-Generally, medical images are large and depict anatomical and pathophysiologicalinformation of staggering variety both within a single image and across a population ofimages Worse, it is usually the case that clinically significant information is quite subtle.For example, Figure 41.1 shows a particularly straightforward example of a mammogram.Microcalcifications, the small white spots shown in Figure 41.1, are deposits of calcium

or magnesium salts that are smaller than 1 mm Clusters of microcalcifications are often theearliest sign of non-palpable breast cancer, though it must be stressed that benign clustersare often found, and that many small white dots do not correspond to microcalcifications(see Highnam and Brady [5] for an introduction to the physics of mammography and tomicrocalcifications) In order to retain the microcalcifications that a skilled radiologist candetect, it is usual to digitise mammograms to a resolution of 50 to 100µ It has beenfound that the densities in a mammogram need to be digitised to a resolution of 14 to

16 bits, yielding 2 bytes per pixel An A4-sized mammogram digitised at the appropriateresolution gives an image that is typically4000× 4000pixels, that is 32 MB Generally,two views – craniocaudal (CC, head to toe) and mediolateral oblique (MLO, shoulder

to opposite hip) – are taken of each of the breasts, giving 128 MB per patient per visit,approximately an order of magnitude greater than that from a contrast-enhanced MRI

Figure 41.1 A patient aged 61 years presented with a breast lump Mammography reveals a

2 cm tumour and extensive microcalcifications, as indicated by the arrows Diagnostically, this

in the MRI receiver coil Such a degradation of an image may appear subtle, and may bediscounted by the (expert) human eye; but it can distort massively the results of automatictissue classification and segmentation algorithms, and give wildly erroneous results foralgorithms attempting quantitative analysis of an image

Over the past fifteen years there has been substantial effort aimed at medical imageanalysis – the interested reader is referred to journals such as IEEE Transactions on Med-ical Imaging or Medical Image Analysis, as well as conference proceedings such asMICCAI (Medical Image Computation and Computer-Assisted Intervention) There hasbeen particular effort expended upon image segmentation to detect regions-of-interest:shape analysis, motion analysis, and non-rigid registration of data, for example, fromdifferent patients To be deployed in clinical practice, an algorithm has to work 24/7 withextremely high sensitivity and specificity This is a tough specification to achieve even forimages of relatively simple shapes and in cases for which the lighting and camera-subjectpose can be controlled; it is doubly difficult for medical images, for which none of thesesimplifying considerations apply There is, in fact, a significant difference between imageanalysis that uses medical images to illustrate the performance of an algorithm, and med-ical image analysis, in which application-specific information is embedded in algorithms

in order to meet the demanding performance specifications

We noted in the previous section that clinicians often find themselves drowning indata One potential solution is data fusion – the integration of diverse data sets in a singlecohesive framework – which provides the clinician with information rather than data Forexample, as we noted above, PET, and SPECT can help identify the microvasculaturegrown by a tumour However, the spatial resolution of PET is currently relatively poor(e.g 3 to 8 mm voxels), too poor to be the basis for planning (say) radiotherapy On theother hand, CT has excellent spatial resolution; but it does not show soft tissues such

as grey matter, white matter, or a brain tumour Data fusion relates information in the

CT with that in the PET image, so that the clinician not only knows that there is a tumour but where it is Examples of data fusion can be found by visiting the Website:

http://www.mirada-solutions.com

PACS systems have encouraged the adoption of standards in file format, particularly theDICOM standards – digital communication in medicine In principle, apart from the rawimage data, DICOM specifies the patient identity, the time and place at which the imagewas taken, gives certain technical information (e.g pulse sequence, acquisition time),specifies out the region imaged, and gives information such as the number of slices, and

so on

Such is the variety of imaging types and the rate of progress in the field that DICOM

is currently an often frustrating, emerging set of standards

41.3 MAMMOGRAPHY

41.3.1 Breast cancer facts

Breast cancer is a major problem for public health in the Western world, where it is themost common cancer among women In the European Community, for example, breastcancer represents 19% of cancer deaths and fully 24% of all cancer cases It is diagnosed

in a total of 348 000 cases annually in the United States and the European Communityand kills almost 115 000 annually Approximately 1 in 8 of women will develop breastcancer during the course of their lives, and 1 in 28 will die of the disease According tothe World Health Organization, there were 900 000 new cases worldwide in 1997 Suchgrim statistics are now being replicated in eastern countries as diets and environmentbecome more like their western counterparts

During the past sixty years, female death rates in the United States from breast cer stayed remarkably constant while those from almost all other causes declined Thesole exception is lung cancer death rates, which increased sharply from 5 to 26 per

can-100 000 It is interesting to compare the figures for breast cancer with those from vical cancer, for which mortality rates declined by 70% after the cervical smear gainedwidespread acceptance

cer-The earlier a tumour is detected the better the prognosis A tumour that is detectedwhen its size is just 0.5 cm has a favourable prognosis in about 99% of cases, since it

is highly unlikely to have metastasized Few women can detect a tumour by palpation(breast self-examination) when it is smaller than 1 cm, by which time (on average) thetumour will have been in the breast for up to 6 to 8 years The five-year survival rate

for localized breast cancer is 97%; this drops to 77% if the cancer has spread by thetime of diagnosis and to 22% if distant metastases are found (Journal of the NationalCancer Institute).

This is the clear rationale for screening, which is currently based entirely on X raymammography (though see below) The United Kingdom was the first country to develop

a national screening programme, though several other countries have established such grammes: Sweden, Finland, The Netherlands, Australia, and Ireland; France, Germany andJapan are now following suit The first national screening programme was the UK BreastScreening Programme (BSP), which began in 1987 Currently, the BSP invites womenbetween the ages of 50 and 64 for breast screening every three years If a mammogramdisplays any suspicious signs, the woman is invited back to an assessment clinic whereother views and other imaging modalities are utilized Currently, 1.3 million women arescreened annually in the United Kingdom There are 92 screening centres with 230 radi-ologists, each radiologist reading on average 5000 cases per year, but some read up to

of post-menopause women In essence, the BSP defines the menopause to be substantiallycomplete by age 50!

The UK programme resulted from the Government’s acceptance of the report of thecommittee chaired by Sir Patrick Forrest The report was quite bullish about the effects

of a screening programme:

by the year 2000 the screening programme is expected to prevent about 25% of deaths from breast cancer in the population of women invited for screening On average each of the women in whom breast cancer is prevented will live about 20 years more Thus by the year 2000 the screening programme is expected to result in about 25 000 extra years of life gained annually in the UK.

To date, the BSP has screened more than eleven million women and has detected over

65 000 cancers Research published in the BMJ in September 2000 demonstrated thatthe National Health Service (NHS) Breast Screening Programme is saving at least 300lives per year The figure is set to rise to 1250 by 2010 More precisely, Moss (BritishMedical Journal 16/9/2000), demonstrated that the NHS breast screening program, begun

in 1987, resulted in substantial reductions in mortality from breast cancer by 1998 In

1998, mortality was reduced by an average of 14.9% in those aged 50 to 54 and 75 to 79,which would be attributed to treatment improvements In the age groups also affected byscreening (55 to 69), the reduction in mortality was 21.3% Hence, the estimated directcontribution from screening was 6.4%

Recent studies suggest that the rate of interval at which cancers appear between cessive screening rounds is turning out to be considerably larger than predicted in theForrest Report Increasingly, there are calls for mammograms to be taken every two yearsand for both a CC and MLO image to be taken of each breast

suc-Currently, some 26 million women are screened in the United States annually imately 55 million worldwide) In the United States there are 10 000 mammography-accredited units Of these, 39% are community and/or public hospitals, 26% are privateradiology practices, and 13% are private hospitals Though there are 10 000 mammogra-phy centres, there are only 2500 mammography specific radiologists – there is a world-wide shortage of radiologists and radiologic technologists (the term in the United Kingdom

(approx-is radiographers) Huge numbers of mammograms are still read by non-special(approx-ists, travening recommended practice, nevertheless continuing with average throughput ratesbetween 5 and 100 per hour Whereas expert radiologists have cancer detection rates of

con-76 to 84%, generalists have rates that vary from between 8 to 98% (with varying numbers

of false-positives) The number of cancers that are deemed to be visible in retrospect, that

is, when the outcome is known, approaches 70% (American Journal of Roentgenology1993) Staff shortages in mammography seem to stem from the perception that it is ‘bor-ing but risky’: as we noted earlier, 12% of all malpractice lawsuits in the United Statesare against radiologists, with the failure to diagnose breast cancer becoming one of theleading reasons for malpractice litigation (AJR 1997 and Clark 1992) The shortage ofradiologists is driving the development of specialist centres and technologies (computers)that aspire to replicate their skills Screening environments are ideally suited to computers,

as they are repetitive and require objective measurements

As we have noted, screening has already produced encouraging results However, there

is much room for improvement For example, it is estimated that a staggering 25% ofcancers are missed at screening It has been demonstrated empirically that double readinggreatly improves screening results; but this is too expensive and in any case there aretoo few screening radiologists Indeed, recall rates drop by 15% when using 2 views ofeach breast (British Medical Journal, 1999) Double reading of screening mammogramshas been shown to half the number of cancers missed However, a study at Yale of boardcertified, radiologists showed that they disagreed 25% of the times about whether a biopsywas warranted and 19% of the time in assigning patients to 1 of 5 diagnostic categories.Recently, it has been demonstrated that single screening plus the use of computer-aideddiagnosis (CAD) tools – image analysis algorithms that aim to detect microcalcificationsand small tumours – also greatly improve screening effectiveness, perhaps by as much

as 20%

Post-screening, the patient may be assessed by other modalities such as palpation,ultrasound and increasingly, by MRI 5 to 10% of those screened have these extended

‘work-up’ Post work-up, around 5% of patients have a biopsy In light of the number

of tumours that are missed at screening (which reflects the complexity of diagnosingthe disease from a mammogram), it is not surprising that clinicians err on the side ofcaution and order a large number of biopsies In the United States, for example, there areover one million biopsies performed each year: a staggering 80% of these reveal benign(non-cancerous) disease

It has been reported that between screenings 22% of previously taken mammogramsare unavailable or are difficult to find, mostly because of the fact that they have beenmisfiled in large film archives – lost films are a daily headache for radiologists around

the world, 50% were obtained only after major effort, Bassett et al (American Journal of

Roentgenology, 1997)

41.3.2 Mammographic images and standard mammogram form (SMF)

Figure 41.2 is a schematic of the formation of a (film-screen) mammogram A collimatedbeam of X rays passes through the breast and is compressed (typically to a force of

14 N) between two Lucite plates The X-ray photons that emerge from the lower platepass through the film before being converted to light photons, which then expose the film,which is subsequently scanned (i.e converted to electrons) at a resolution (typically) of

50µ In the case of full-field digital mammography, the X-ray photons are converteddirectly to electrons by an amorphous silicon sensor that replaces the film screen AsFigure 41.2 also shows, a part of the X-ray flux passes in a straight line through thebreast, losing a proportion of less energetic photons en route as they are attenuated by thetissue that is encountered The remaining X-ray photon flux is scattered and arrives at thesensor surface from many directions (which are, in practice, reduced by an anti-scattergrid, which has the side-effect of approximately doubling the exposure of the breast) Fulldetails of the physics of image acquisition, including many of the distorting effects, andthe way in which image analysis algorithms can be developed to undo these distortions,are presented in Highnam and Brady [5]

For the purposes of this article, it suffices to note that though radiologic gists are well trained, the control over image formation is intrinsically weak This is

technolo-illustrated in Figure 41.3, which shows the same breast imaged with two different

expo-sure times The images appear very different There are many parameters p that affect

the appearance of a mammogram, including: tube voltage, film type, exposure time, andplacement of an automatic exposure control If these were to vary freely for the samecompressed breast, there would be huge variation in image brightness and contrast Ofcourse, it would be ethically unacceptable to perform that experiment on a living breast:

the accumulated radiation dose would be far too high However, it is possible to develop

a mathematical model of the formation of a mammogram, for example, the Brady physics model With such a model in hand, the variation in image appearance

Highnam-can be simulated This is the basis of the teaching system VirtualMammo developed

Film screen cassette, anti-scatter grid, and intensifier

Figure 41.2 Schematic of the formation of a mammogram.

Figure 41.3 Both sets of images are of the same pair of breasts, but the left pair is scanned with

a shorter exposure time than the right pair – an event that can easily happen in mammography Image processing algorithms that search for ‘bright spots’ will be unable to deal with such changes.

by Mirada Solutions Limited in association with the American Society of RadiologicTechnologists (ASRT)

The relatively weak control on image formation, coupled with the huge change inimage appearance, at which Figure 41.3 can only hint, severely limits the usefulness ofthe (huge) databases that are being constructed – images submitted to the database maytell more about the competence of the technologists who took the image, or the state of theequipment on which the image was formed, than about the patient anatomy/physiology,which is the reason for constructing the database in the first place! It is precisely this

problem that the eDiamond project aims to address.

In the course of developing an algorithm to estimate, and correct for, the scatteredradiation shown in Figure 41.2, Highnam and Brady [5] made an unexpected discovery:

it is possible to estimate, accurately, the amount of non-fat tissue in each pixel column

of the mammogram More precisely, first note that the X-ray attenuation coefficients ofnormal, healthy tissue and cancerous tissue are very nearly equal, but are quite differentfrom that of fat Fat is clinically uninteresting, so normal healthy and cancerous tissuesare collectively referred to as ‘interesting’: Highnam and Brady’s method estimates – inmillimetres – the amount of interesting tissue in each pixel column, as is illustrated inFigure 41.4

The critical point to note is that the interesting tissue representation refers only to jected) anatomical structures – the algorithm has estimated and eliminated the particular

(pro-parameters p(I) that were used to form this image I In short, the image can be regarded

as standardised Images in standardised form can be included in a database without the

confounding effect of the (mostly irrelevant – see below) image formation parameters.This greatly increases the utility of that database Note also that the interesting tissue

representation is quantitative: measurements are in millimetres, not in arbitrary contrast

units that have no absolute meaning