Lots of Commercial Dose Radiation Dose Informatics: Using the …amos3.aapm.org/abstracts/pdf/72-20318-285394-90775.pdf · 2013-02-28 · ©2011 MFMER | slide-7 “Inventor” of

©2011 MFMER | slide-1

3

• William Pavlicek • AAPM Clinical Meeting• Monday March 18, 2013• Phoenix AZ

Radiation Dose Informatics: Using the Tools of Six Sigma to Improve Radiation Exposure

Prescription


Lots of Commercial DoseManagement Software!


Image Quality and Dose

Increasing Radiation Dose

IncreasingImage Quality


Problem: Good IQ with high (or even too high) dose




Too high dose prescription must be avoided




Narrow the range of x-ray prescriptions!




“Inventor” of Six SigmaW. Edwards Deming

• Physicist PhD (Yale, 28)

• Taught engineering, physics in the 1920s

• Long career in government statistics, USDA, Bureau of the Census

• Worked with Japan post war.

7

W. Edwards Deming, 1900 – 1993


* From Montgomery, D. C. (2009), Introduction to Statistical Quality Control 6th edition, Wiley, New York

Toyota reduced Variation ..Improved Quality 1960s

8


The Motorola Six-Sigma Concept 1980 - pagers

• Motoroloa found process subject to disturbances that could cause it to shift by as much as 1.5 standard deviations off target.

• No process or system is ever truly stable!

• Thus 3.4 parts per million for this system

9

* From Montgomery


“The mean never happens,” — a 4-day delivery time on one order, with an awful 20-day delay on another, and no real consistency! This customers in this chart feel nothing. Their life experience hasn’t changed; one bit. The customer only feels the variance that we have not yet removed. … Variation is evil in any customer-touching process. *

*Jack Welch, General Electric Company 1998 Annual Report

10

X-ray

Tubes

1990s


What diagnostic medical physicists measure?


Repeatability of device? eg. New ACR Daily CT Phantom


These devices were repeatable with accurate output,…..!!!


Reproducibility!!!

DifferentProcedures/Protocols, Operator Training,

Patients!


Six Sigma Process Improvement with an Emphasis

on Achieving Significant Impact!

• All work is performed in (interconnected) processes• Easy to see in some situations (manufacturing)• Harder in others

• Any process can be improved• An organized approach to improvement is necessary

15


Reduce System Variability

DMAIC (duy–may–ick)


Basic Tenets of Quality

• It is the process that creates variability

• Belief that things can be improved

• A blameless environment is needed for team solutions

• People closest to the product are most able to affect quality

• Everyone has responsibility for quality


The Primary Six Sigma Tools

• Process map

• Cause and effect analysis• Measurement systems analysis*

• Capability study*

• Failure mode and effects analysis

• Observational study (regression)*• Designed experiments*

• Control charts and out-of-control-action-plans*

* Statistical Tools 19

Measure, Analyze and Improve!Reduce Exam Prescription Variability….

Measure, Analyze and Improve!Reduce Exam Prescription Variability….

Executive SummaryExams are grouped by the dose range in 1 Gy increments,

The number of cases exceeding the Dose Threshold (set by the customer) are shown to the right.

5/3/2011

22

GE ‘Dose Watch’

Six Sigma/Informatics and Fluoroscopy


Three ‘steps’to improve fluoroscopy x-ray

prescriptions

1. Adding Copper filtration2. Lower pulse and frame rates3. Table-Detector positioning


ANALYZE: Lucite 15 cm to 35 cm20% Iodine solution for CNR measures


Slightly (~8-10%) reduced CNR 0.1 mm copper

CNR vs mm Cu Acquisition

0

20

40

60

80

100

120

140

160

0 0.2 0.4 0.6 0.8 1

mm Cu

CN

R

15cm Lucite20cm Lucite25cm Lucite27.5cm Lucite30cm Lucite35cm Lucite


BUT ~40% Lower Skin Dose 0.1mm Cu!

Differential Dose Acquisition

-90%

-80%

-70%

-60%

-50%

-40%

-30%

-20%

-10%

0%0.0 0.2 0.4 0.6 0.8 1.0

mm Cu

Perc

ent C

hang

e in

Dos

e fo

r A

dditi

on o

f Cop

per

15cm Lucite20cm Lucite25cm Lucite27.5cm Lucite30cm Lucite35cm Lucite


Step 1: EASY IMPROVEMENT!

• 0.1mm Copper minimum ADDED TO:

• 100% Stationary Fluoroscopy devices

• 100% protocols

• 100% Patients have 40% reduced exposure!


DMAI Control


Step 2:Operator Frame Rate Behaviors

• Lowering pulse and frame rates

• Important!: Can you see what is needed?

• Tool 1: Collect pilot examples

• Tool 2: Share results (data - Shine a light)

• Tool 3: Educate


(70-75 bpm, flat panel)(70-75 bpm, II)

7.5 FPS cine 15 FPS cineNew ½ Dose Old Dose


Shine a light!

Tool 2Data can Shine a light!

Names blocked out

Ufors RaySafe

Unfors RaySafe

Cumulative Dose (ESAK) Incidence Map

Shows Cumulative dose (ESAK) with respect to gantry angulations. The

horizontal axis indicates gantry positions with respect to LAO/RAO patient

positioning, and the vertical axis indicates cranial/caudal gantry positions. The dose (ESAK) is displayed on the incidence map indicating how much of the exam dose was distributed to the patient with respect to the

position of the gantry in 30 degree increments.

For purposes of reporting high skin dose events, add up the max total dose in any four contiguous blocks (four blocks with a corner in common). This represents the worst possible overlap of the collimated

fields, producing a "hot spot" that is potentially higher dose than any one of the

fields

5/3/2011

34

GE ‘Dose Watch’

©2011 MFMER | slide-35 ©2011 MFMER | slide-36

NEED TOOL!

• Was the 3.8 Gy skin exposure optimized?

• Analyze what happened!

• Educate to Improve behaviors

©2011 MFMER | slide-38©2011 MFMER | slide-38

Post Procedure(Informatics)

• Good Geometry?

• When was DSA

or Cine used?


DMAIC-Improve: Informatics ToolALARA Review

3.8 Gy Total


Informatics Tool 3ALARA Review

Good!Low detector!

High table

3.8 Gy Total



Table too low!Below IRP!

Detector too high!

3.8 Gy Total



3.8 Gy Total Max magneeded?

NOW! Six Sigma/Informatics and Computed Tomography

CT Value Stream (Process Map)

• Value is created via efforts of people as product created (knowledge).

• Relentless pursuit of waste elimination as competitive leverages

Scheduling Pt check in Pt prep CTScan

Report completed

Value Stream for CT Radiology

Protocoldetermined

Dose Informatics and Analytics

Clinician needs a CT

• Value to Patient/Physician customer is knowledge for optimal next action

ClinicianOrders

Pt check in Pt prep CTScan

Report completed

Protocoldetermined

Clinician Orders a CT

• Value to Patient/Physician customer is knowledge for optimal next action - ‘medical, surgical or interventional’

ClinicianOrders

Patientchecks in Pt prep CTScan

Report completed

Protocoldetermined

Are there useful Prior Exams?Eliminate Repeat?

Note Changes?

RSNA Image Share (NIBIB)

AWARE

AWARE

Clinician Reviews Prior Exams

Unfors RaySafe

Clinician Orders Exam

Unfors RaySafe

Unfors RaySafe

Clinician Orders NEW CT MEANWHILE…. Technologist manages the Protocol Book- the CT Exam ‘Scripts’

• Is this efficient?

ClinicianOrders


Report completed

Protocols

What set of acquisitions are on the scanner?

On ALL scanners?All the SAME?

Protocol Management

Full document control including

• Track revisions to protocols• Review reminders• Web-based access• Customizable style sheets and

entry forms

Full document control including

• Track revisions to protocols• Review reminders• Web-based access• Customizable style sheets and

entry forms

Support for multiple scanners in a single protocolSupport for multiple scanners in a single protocol

In Complia nce With HIPAA regulations. Pa tient informa tion listed on the GUI are examples only and do not contain any actua l pa tient information

Protocol Manager ToolBayer HealthCare

EXPOSURE

Order in EMR transfers to RIS – generates Order

Protocol Manager Tool

Bayer Exposure

Radiologists selects a CT ‘Script’

• Prescribing an optimal CT exam

ClinicianOrders


Report completedProtocols

What set of acquisitions create optimal diagnostic content?

Radiologists selects a CT ‘Script’

• Prescribing an optimal CT exam

ClinicianOrders


Report completedProtocols

What set of acquisitions create optimal diagnostic content?

Do radiogists standardize prescription for indications?

CT tech and Nurse: Closest to Prescriptioncan highly affect quality!

ClinicianOrders


Report completed

Protocoldetermined

Correct Patient? Order? Protocol?

Correct Patient? Order? Exam?Prep? Allergies?

Tech Administers X-ray ScriptSelects from CT Scanner

Protocol

• Scanners are NOT a node on the network!

• Scanner settings differ from Protocol Book!

• Differ between scanners

• Established console settings – good!• On the fly changes – variability in exams

• Console settings complicated – need virtual scanner

Scanner

Individuals closest to ‘product’ most able to affect quality

• Toyota Assembly Line worker STOPS the Line!• Procedural Pause – Operating Room

Dose Check!

GE: Establish Dose Check Values by Protocol

Centering the Patient in the GantryProcess Variability

1. Magnification Errors

2. FOV Cut-off AP

3. FOV Cut-off Zaxis

Table Height in gantry

True Perceived

• Lateral Laser Lights – Table Height

• Table Height – affects mA modulation

• Table more LIKELY to be too LOW

Configure x-ray tube at top!

PA Scout AP Scout

5 10 15 20020 15 10 5

5 10 15 20020 15 10 5

1. Mag Error - mA modulation


PA ScoutTable Low

52 cm

AP ScoutTable Center

45 cm

Tube: below above above

AP ScoutTable High

47 cm

2. FOV Cut-off AP

Patient low on lateral scout.

Raise table in gantry.

Acquire AP scout.

Patient low on lateral scout.

Raise table in gantry.

Re-acquire lateral scout.

Acquire AP scout.

Scout misrepresents patient position for FOV.

Scout represents patient position for FOV.


3. FOV Cut-off Zaxis

Exam 1

Properly Centered – No Mag

FOV INSIDE of CT Radiograph!Auto mA is working!

Exam 4

Patient LOW – CT Radiograph isMagnified!

FOV OUTSIDE of CT Radiograph!Auto mA is confused!

LAT = 375 mm AP = 271 mm

mA modulationOff isocenter shift

Protocol name = 5.1 CAP WOPatient position : FFSDLP = 597 mGy.cm

CTDIvol = 28,5 mGyCTDIvol = 28,5 mGy

John DOE PID : 12345 AN : 235647 Study description : CAP CT

Series description : CAPLAT + AP = 646 mm

Diametereff = 31,5 cm

fSSDE = 1,16

IEC BODY DOSIMETRY PHA NTOM ACCEPTACCEPTREDRAWREDRAW

SSDE = 33,06 mGySSDE = 33,06 mGy

IEC BODY PHANTOM

Off isocenter shift

Table height = 163

Delta Y

Delta X

Delta X = 28 mmDelta Y = 33 mm

I 56

S703

Mean = 368 mAMin = 149 mAMax = 475 mA

mA range :

Quality Review of CT Dose

• ACR National Radiation Dose Registry

• ACR CT Accreditation - Medical Physics Review

• Joint Commission SE #47

• California SB 1237

ACR CT Quality Control Manual

©2011 MFMER | slide-75 ©2011 MFMER | slide-76

New Requirements for CT ACR Accreditation

NRDR – ACR Informatics for CT


Routine Head – No Contrast

JC: Institute a process for the review of alldosing protocols…


JC: Investigate Doses outside the range

• 6030_Abdomen_Pelvis scan length 40-60 cm


Toolkit

Integrated Dosimetry


Bayer HealthCare Exposure

Examination Analysis



mSv

Integrated Dosimetry



a

Paden, Robert G. (Gene)PHYSICIST-DIAGNOSTICIM…

Boltz, Thomas F. IIISPEC-DIAG RAD PHYSICS…

Hanson, James A.SPEC-DIAG RAD PHYSICS…

Zhang, MinSCIENTIST/PROGRAMMER…

Wu, Lin-WeiSCIENTIST/PROGRAMMER…

Mango Kaiser, Janice H. …SUPV-CT IMAG-AZ

Loprino, Shirley A., R.T.(R)MGR-RAD IMAG-AZ

Ermer, Ellen R. ANALYST III-BUS SYS…

Sabyan, Stephen J., R.T. …SUPV-IR/CATH-AZ

Wilt, Michelle A., R.T.(R) SUPV-GEN RAD IMAG-AZ

Fike, Rachel D. TECH I-CLINICAL ENG-AZ

Sanchez, Juan J. TL-BIOMED/IMAG-AZ

Ledoux, Elaine M., R.T. ...TECH-CT IMAG II-AZ

Leyk, Linda L., R.T.(R)...TL-CT IMAG-AZ

Osborn, Howard H., R.N. SUPV RN-INPT/365-AZ

Naidu, Sailendra G., M.D. CONS-DIAGNOSTIC RAD

Huettl, Eric A., M.D. CONS-DIAGNOSTIC RAD

Wellnitz, Clinton V., M.D. CONS-DIAGNOSTIC RAD

Kriegshauser, Jeffry S. , M.D. CONS-DIAGNOSTIC RAD

Hara, Amy K., M.D. CONS-DIAGNOSTIC RAD

Informatics - Powerful tools for your toolbox!Thank you!

Documents

Lots of Commercial Dose Radiation Dose Informatics: Using the …amos3.aapm.org/abstracts/pdf/72-20318-285394-90775.pdf · 2013-02-28 · ©2011 MFMER | slide-7 “Inventor” of