Radiologists will love the A ustrian P ACS P rocedure C ode Franz - - PDF document

radiologists will love the
SMART_READER_LITE
LIVE PREVIEW

Radiologists will love the A ustrian P ACS P rocedure C ode Franz - - PDF document

BURA Radiologists will love the A ustrian P ACS P rocedure C ode Franz Frhwald APPC What is the APPC? A coding system for radiologic anatomy, modalities and procedures to support automated retrieval of relevant prior studies out of


slide-1
SLIDE 1

BURA

Radiologists will love the APPC

Austrian PACS Procedure Code

Franz Frühwald

slide-2
SLIDE 2

BURA

What is the APPC?

  • A coding system for radiologic anatomy,

modalities and procedures to support automated retrieval of relevant prior studies out of your own or “foreign” PACS

  • What problem is solved
  • How is coding done
  • What will be the further development
slide-3
SLIDE 3

BURA

Where are we now ?

  • Different imaging machinery feeds data into

PACS

  • We read everything on our workstation
  • We exchange data inside the hospital, the

hospital chain, inside OUR own organization

  • We do not access someone else's PACS
  • Why?? (privacy, legal stuff etc.)
  • Technically: OUR PACS can not “intelligently”

interpret the data of a “foreign” PACS – no “language interpreter” available

slide-4
SLIDE 4

BURA

What link is missing?

  • Everything works fine inside YOUR OWN
  • rganization
  • Even machinery of different vendors cooperate

smoothly in YOUR hospital, where YOU have defined everything

  • Till now nobody cared finding stuff in someone

else's PACS

  • Missing link: trans organizational search for

imaging data – till now not acknowledged as problem and therefore not addressed

slide-5
SLIDE 5

BURA

Why a new Code??

  • Austria is beginning to connect all existing

PACS

  • all vendors
  • all different hospital-chains
  • all radiologists offices
  • All stored images of all citizens should

become accessible

  • Working group “Radiology and ELGA”
  • Seems we only stumbled into the problem

first

slide-6
SLIDE 6

BURA

What's the problem??

  • Proper privacy of medical data
  • Extensive existing hard and software with

different structure, wording and coding

  • No commonly accepted terminology for

radiologic procedures

  • No defined wording for anatomical regions

and structures needed in Radiology

  • (RadLex !!!)
slide-7
SLIDE 7

BURA

Problems of existing coding systems

  • In most cases vendor-specific
  • Most coding systems developed for

accounting or reimbursement purposes

  • Too little or too much granularity
  • Virtually impossible to get vendors to use

another vendors codes (license fees)

  • Without a number-based generally accepted

exchange-code no “intelligent” image exchange between different vendors PACS

slide-8
SLIDE 8

BURA

Solution A

  • New standard
  • means all existing stored imaging data

get obsolete,

  • all Hard- and Software to be renewed.
  • Utopic, unpractical because of costs

involved

  • Industry would not cooperate
slide-9
SLIDE 9

BURA

Solution B

  • New Standard with “translation” to

existing coding systems

  • Free of costs (licenses)
  • Global (language independent)
  • Existing hard and software and all

stored imaging data saved from

  • bsolesce
slide-10
SLIDE 10

BURA

Chest

  • Chest pa and lateral
  • Chest
  • Heart and Lung
  • Thorax
  • Lung
  • If all those names were connected to

the same number they cold be found und retrieve automatically

slide-11
SLIDE 11

BURA

We need old images of the chest of Mr. John Doe

  • “solitary lung nodule”

– Chest films 10 years back – CT of the chest if available

  • Result:

Hospital 1 Hospital 2 Hospital 3 private office A private office B “Data mining”

  • Feb. 15 2009
slide-12
SLIDE 12

BURA

Screen layout of different vendors differs enormously

slide-13
SLIDE 13

BURA

Screen layout of different vendors differs enormously

slide-14
SLIDE 14

BURA

Screen layout of different vendors differs enormously

slide-15
SLIDE 15

BURA

Screen layout of different vendors differs enormously

slide-16
SLIDE 16

BURA

Hospital 1 Doe John 30.12.1966 chest 15.02.2009 barium enema 15.01.2008 ribs 22.08.2007 CT thorax 11.09.2006 chest 05.03.2006

Relevant informations extracted

slide-17
SLIDE 17

BURA

Hospital 2 Doe John 30.12.1966 upper GI 15.02.2008 knee 01.02.2008 hip 05.06.2005 hands 05.06.2005 feet 08.04.2004 chest 04.04.2004

Relevant informations extracted

slide-18
SLIDE 18

BURA

Hospital 3 Doe John 30.12.1966 Finger 15.05.2006

Relevant informations extracted

slide-19
SLIDE 19

BURA

private office A Doe John 30.12.1966 15.01.2007 barium enema 22.08.2006 ribs 11.09.2005 CT Abdomen 05.03.2002 chest 15.02.2003 chest

Relevant informations extracted

slide-20
SLIDE 20

BURA

private office B 30.12.1966 John Doe US abdomen 03.03.2008 nuk Thyroid 28.01.2005 chest film 14.01.1999 MRI Knee left 15.08.2001

Relevant informations extracted

slide-21
SLIDE 21

BURA

Desperately needed

  • Overview
  • Source
  • Chronologic
  • Highlighted
  • Intelligently compressed
slide-22
SLIDE 22

BURA

Synopsis of all relevant information's

slide-23
SLIDE 23

BURA

Chronological display with source indicated

slide-24
SLIDE 24

BURA

Relevant prior examinations highlighted

slide-25
SLIDE 25

BURA

Compressed view with relevant information's only

slide-26
SLIDE 26

BURA

  • Feb. 15 2009

March 5 2003

  • Sept. 11 2006

John DOE John Doe´s pulmonary nodule turns out to be a harmless bony island of a rib. No further examinations required!

slide-27
SLIDE 27

BURA

Solution B

  • New Standard with “translation” to

existing coding systems

  • Free of costs (licenses)
  • Global (Language independent)
  • Existing Hard and software and all stored imaging data saved from obsolesce
  • Must bee widely accepted to become a (global)

standard

  • Should be somehow logical (at least from a

physicians point of view)

  • Should support search for images in

“foreign PACS” (where we have no influence on the structure)

slide-28
SLIDE 28

BURA

APPC translates between different languages, typewriting as well as different internal number coding

slide-29
SLIDE 29

BURA

Requirements for a good radiologic exchange code

  • Abstract (numbers, no words)
  • Logic, hierarchic
  • supporting search for “regions”, not

diseases

  • Enough / not too much granularity
  • Unlimited expandability
  • Open to be connected to accounting

and referral systems

slide-30
SLIDE 30

BURA

Tasks for workstation developers

  • Who gives us the smartest searching

tools?

  • Who supports us best in (preferably

automated) finding all relevant studies to compare them to our recent study?

  • Who is best in applying my favorite

hanging protocols to foreign image data?

slide-31
SLIDE 31

BURA

What support do Radiologists need when all patients images are accessible?

  • Automated search for relevant studies / images

we need to read the recent study

  • “seamless integration” of imagery of other

houses

  • Display arranged using our own DICOM viewer

and the hanging protocols we are used to

  • (No more CD-ROM or DVD material with

viewers we are not used to)

slide-32
SLIDE 32

BURA

Hanging protocols

On viewing boxes

slide-33
SLIDE 33

BURA

Customarized screen layouts

slide-34
SLIDE 34

BURA

Every radiologist likes his personal layout

slide-35
SLIDE 35

BURA

Working speed is increased with perfect hanging protocols

slide-36
SLIDE 36

BURA

Enhanced reading speed with perfect identical arrangement of old and new images new

  • ld
  • ld
  • ld
  • ld

new new new

slide-37
SLIDE 37

BURA

Standards

  • DICOM defines images
  • HL 7 defines documents
  • ICD 10 defines clinical situations and

illnesses

  • RadLex defines radiologic vocabulary
  • Presently no generally accepted standard

number code for radiologic procedures, modalities

  • Presently no ideal coding system for regions

and anatomy necessary in radiology

slide-38
SLIDE 38

BURA

How will trans organizational PACS access change radiologic work?

  • Increase in workload (everything stored

is accessible !!)

  • Massive slowdown of working speed

when all existing imaging data need to be at least shortly acknowledged

  • Increased liability if clues from old data

are missed because not viewed

slide-39
SLIDE 39

BURA

How do Radiologists search for relevant prior studies ?

  • They go for the region

(e.g. abdomen, chest, head, neck etc.)

  • Not so much very specific
  • They like all existing modalities
  • They like to filter the time period
  • Sometimes they want it super specific

when it is a follow up examination

slide-40
SLIDE 40

BURA

APPC

Austrian PACS Procedure Code

  • Developed to support trans -
  • rganizational image data research
  • Supported by all scientific and

professional radiologic organizations in Austria

  • En route to become standardized in the

HL 7

  • Matched with the vocabulary of RadLex
slide-41
SLIDE 41

BURA

APPC

  • 4 Axes:

–Modality –Laterality –Procedure –Anatomy

slide-42
SLIDE 42

BURA

APPC

  • Modality

0 undefined 1 X-Ray 2 CT 3 MRI 4 Ultrasound 5 Nuclear Medicine 6 PET

  • Laterality

undefined 1 right 2 left 3 bilateral 4 unpaired organ 5 atypical Situation (transplant etc.)

slide-43
SLIDE 43

BURA

APPC

  • Procedure

0 undefined 1 Imaging of preformed ducts 2 Quantitative Analysis / Reconstructions 3 Documentation of Interventions (via artificial access) 4 Imaging with open radioactive substances

slide-44
SLIDE 44

BURA

APPC

  • Anatomy

0 undefined 1 Head 2 Neck 3 Thorax 4 Abdomen 5 Musculoskeletal 6 Vascular System 7 Breast

slide-45
SLIDE 45

BURA

Example of APPC coding

slide-46
SLIDE 46

BURA

Example of APPC coding

slide-47
SLIDE 47

BURA

Example of APPC coding

slide-48
SLIDE 48

BURA

Full extent of APPC – not very huge!

slide-49
SLIDE 49

BURA

Coding

  • Important for search in foreign PACS:

– Region – Modality

  • Important for own PACS

– Maybe everything, depending on how you use your system

slide-50
SLIDE 50

BURA

Coding example

Modality . Laterality . Procedure . Anatomy Barium enema:

  • 1. 0 . 1. 4-2-3

X-ray . undefined . preformed ducts . Colon CT colonography

  • 2. 0 . 2-3-7. 4-2-3

CT . undefined . virtual endoscopy . Colon

slide-51
SLIDE 51

BURA

Filtering for studies

  • All abdominal Studies 0.0.0.4
  • All bowel images

0.0.0.4-2-3

  • All CT-studies

2.0.0.0

  • All MRI arthrographies
  • f the right knee

3.1.1-5.5-2-4

  • All existing material

0.0.0.0

slide-52
SLIDE 52

BURA

Implementation

  • 1 additional field in your RIS or PACS

for the APPC – number of any examination or procedure you perform in your organization

  • Predefine what should be searched for

(a task for workstation developers)

  • Everything preprogrammed – the direct

user is not forced to code anything

slide-53
SLIDE 53

BURA

Further developement of the APPC

  • Maintenance: by the professional radiologic
  • rganizations of Austria
  • Yearly updates planned
  • Feedback and propositions for new entry's

welcome

  • New language translations for all interested

(presently English and German versions are available)

  • Lobbying to get connected to DICOM and

HL7 standards

slide-54
SLIDE 54

BURA

APPC website for download

  • www.BURA.at
  • Comments welcome
  • Mailing list for all who want to be

informed of new releases

  • E-mail account for feedback and

proposals of new entry's

slide-55
SLIDE 55

BURA

Summary

  • What problems exist in work with trans
  • rganizational PACS
  • How the APPC could help to solve those
  • How the APPC is used and coding is

performed

  • How the APPC will be developed in the future
  • Radiologists will love the support in their daily

work

slide-56
SLIDE 56

BURA

Thank you