DEVELOPMENT OF DIGITAL MAMMOGRAPHY DISPLAYS AND WORKSTATIONS (SBIR/STTR)

Release Date:  April 13, 1999

PA NUMBER:  PA-99-083

P.T.

National Cancer Institute

PURPOSE

The purpose of this Program Announcement (PA) is to alert the investigator
community to the need for and National Cancer Institute (NCI) interest in a
concerted effort to overcome the problems of display for digital mammograms.  The
problem cannot be solved simply with computer programming or hardware
improvements alone.  Integration of advanced hardware, software, and
psychophysics research data is needed to optimize the early diagnosis of breast
cancer using digital mammography.  This Program will utilize the Small Business
Innovation Research (SBIR) and Small Business Technology Transfer (STTR)
mechanisms, but will be run in parallel with a program of identical scientific
scope utilizing the R01 research project grant mechanism (PA-99-082).  This RFA is available at:  
http://www.nih.gov/grants/guide/pa-files/PA-99-082.html).

This PA encourages the small business community to develop displays for digital
mammography.  This PA must be read in conjunction with the OMNIBUS SOLICITATION
OF THE PUBLIC HEALTH SERVICE FOR SMALL BUSINESS INNOVATION RESEARCH GRANT (SBIR)
APPLICATIONS (PHS 99-2) and the OMNIBUS SOLICITATION OF THE NATIONAL INSTITUTES
OF HEALTH FOR SMALL BUSINESS TECHNOLOGY TRANSFER (STTR) GRANT APPLICATIONS (PHS
99-3).  All of the instructions within the omnibus solicitation apply.

HEALTHY PEOPLE 2000

The Public Health Service (PHS) is committed to achieving the health promotion
and disease prevention objectives of "Healthy People 2000," a PHS-led national
activity for setting priority areas.  This PA, Development of Digital Mammography
Displays And Workstations, is related to the priority area of cancer.  Potential
applicants may obtain a copy of "Healthy People 2000" (Full Report:  Stock No.
017-001-00474-0 or Summary Report:  Stock No. 017-001-00473-1) through the
Superintendent of Documents, Government Printing Office, Washington, DC
20402-9325 (telephone 202-512-1800), or at
http://www.crisny.org/health/us/health7.html.

ELIGIBILITY REQUIREMENTS

Eligibility requirements are described in the OMNIBUS SOLICITATIONS.  Any small
business, independently owned by United States citizens and located in the United
States, may apply.  This program will utilize Small Business Innovation Research
(SBIR) and Small Business Technology Transfer (STTR) mechanisms, but will be run
in parallel with a program of identical scientific scope (PA-99-092) for research
project grant (R01) applications by foreign and domestic, for-profit and
non-profit organizations, public and private, such as universities, colleges,
hospitals, laboratories, units of State and local governments, and eligible
agencies of the Federal government.

MECHANISM OF SUPPORT

Support for the PA is through the SBIR and STTR mechanisms, which are set-aside
programs.

Applications may be submitted for support as Phase I STTR (R41) or Phase I SBIR
(R43) grants: Phase II STTR (R42) or Phase II SBIR (R44) grants; or under the
SBIR/STTR  FAST-TRACK option as described in the OMNIBUS SOLICITATIONS.  Phase
II applications in response to this PA will only be accepted as competing
continuations of previously funded NIH Phase I SBIR/STTR awards.  The Phase II
proposal must be a logical extension of the Phase I research.

Information on the FAST-TRACK process and the OMNIBUS SOLICITATIONS are available
at: http://www.nih.gov/grants/funding/sbir.htm

RESEARCH OBJECTIVES

Background

Digital Mammography is one of the most promising research areas for improving
early detection of breast cancer. Experts in the field agree that current
softcopy (i.e., video) display systems remain an impediment to full realization
of the potential of digital mammography.  Softcopy display is essential for
optimizing the use and display of digital mammography data, both in the research
and clinical setting.

On March 9-10, 1998 in Washington, DC, the Public Health Service's Office on
Women's Health and the National Cancer Institute assembled a Joint Working Group
on Digital Mammography: Digital Displays and Workstation Design.  The meeting,
attended by over 100 scientific leaders, was convened to achieve the following
goals: 1) To review the state-of-the-art of display technologies, including
current and future clinical applications and technical challenges; and 2) To
outline research priorities in digital display technology and workstation design.
This Program Announcement is based on the recommendations and research agenda
developed by the Working Group.

Little research has been done on the hardware, software, and psychophysical
issues involved in workstation design.  Further extensive effort is required for
the successful development, testing and implementation of digital mammography
displays and workstation design for image interpretation.  Manufacturers of
digital mammography systems have generally adapted existing workstations,
developed for CT or MRI, to their digital mammography units.  The dearth of data
on perceptual issues, software applications and hardware comparisons impedes
development of optimal workstations.  Data from observer studies are needed. 
Since mammography demands greater spatial and contrast resolution than any other
clinical area of radiology, it is likely that results obtained from research
targeted toward meeting the needs of mammography will have beneficial effects on
other areas of clinical imaging as well.

Program Initiative

The goal of this program initiative is to advance the state-of-the-art in digital
mammography displays and workstation design in order to obtain the full potential
of digital mammography for improved breast cancer diagnosis.  This PA solicits
research and development in three inter-related areas relevant to digital
mammography: 1) Video Display Hardware; 2) Workstation Software and Design; and
3) Image Perception.  Although these are somewhat distinct areas of research
endeavor, it is important to integrate advances in all three areas in order to
obtain optimal utilization of the image data.

I. Softcopy Display Hardware.

A. Required Features

Digital mammography examinations produce data sets immensely larger than can be
presented or perceived at one time with currently available monitor or flat-panel
display technology.  A digital mammographic examination generates four images of
20 to 200 megabytes each.  A screening study will usually be compared to a prior
exam, and these eight images are the typical data set the radiologist must
interpret.

Direct-view display technologies capable of high brightness and high spatial
resolution are considered to be appropriate candidates for mammography display. 
Current display technologies for 2500 x 3000 pixel images may be suitable for
digital mammography if software functions can help overcome such display
limitations as spatial resolution, limited luminance, and dynamic range. However,
objective data on display performance for clinical tasks are lacking.

B. Examples of Research Priorities for Softcopy Display Hardware

1.  Studies are needed to objectively evaluate display technologies for
mammographic imaging. For example, it is necessary to determine the relation
between the visibility of early signs of breast cancer (such as mass,
microcalcifications, architectural distortion, skin and nipple changes), and
digital display performance parameters (such as luminance, dynamic range, spatial
and contrast resolution and noise characteristics).

High resolution display technologies providing high spatial and contrast
resolution, high luminance, high dynamic range and wide viewing angle at
reasonable cost need to be developed.  For example, research is needed on
materials and device structures for new display technologies.

II. Image Perception

A. Required Features

It has been estimated that at least half of mammography errors are due to faulty
image perception. The efficiency of human perception is influenced by  (1) the
matching of the physical properties of the display to the visual system, (2) the
working environment, (3) the computer interface, (4) the expertise of the
observer, and (5) human factors for both the patient and the observer.  Research
is needed to understand digital display systems and clinical environment
tradeoffs that affect the detection and discrimination of abnormalities in
mammography and to use this information to guide development of hardware,
software and user interfaces.  Predictive models should be developed that
indicate how to improve human performance by changes in images, detectors,
displays, and the environment. Testing every change in the physical parameters
of an imaging system on decision outcome is not feasible. A predictive model
would be much more useful, and scientific effort should be expended on model
development.

B.  Examples of Research Priorities for Image Perception

1.  Perform psychophysical studies of the effect of display parameters on
detection and discrimination of diagnostic features in mammograms.

2. Determine the effect of image navigation and different display protocols on
the detection and discrimination of diagnostic features in mammograms.

3.  Develop an improved understanding of design factors for the image reading
environment.

4. Perform studies to optimize the human user interaction (ergonomics).

5.  Develop computational perception models appropriate for predicting human
detection and discrimination performance using real mammograms, to enable
optimization of display characteristics without the lengthy and prohibitively
costly process of trial-and-error.

6.  Investigate user preference issues with respect to CAD and user directed
image processing and manipulation tools.

7.  Study and quantify the effects of fatigue and vigilance during screening
tasks.

III. Workstation Software and Design

A. Required Features

Softcopy workstations are not currently optimal for mammography.  Workstation
designs must take into account radiologists' needs for both screening and
diagnostic mammography.  Speed and intuitive image review are of paramount
importance. A softcopy workstation must be capable of displaying at least 2
different images at once for comparison; there is typically at total of 8 images
under consideration (2 views of each breast, current and prior). With current
detector technology, this means that workstations must be capable of handling 200
Mbytes or more of raw image data per examination.

For diagnostic studies, film arrangement and manipulation are much more variable
than in screening, and present even greater challenges.  Diagnostic studies
require additional, tailored mammographic views, other modalities (e.g.,
ultrasound, MRI), and techniques (e.g., image analysis, digital tomosynthesis).

Timely display is also important.  Workstations must permit the radiologist to
navigate through hundreds of megabytes of data at speeds approaching those of the
head and eye movements of the radiologist.  It must be possible to load images
rapidly and retrieve other examinations from archives quickly and efficiently.

To make full use of the digital nature of the images, workstations must provide
easy-to-use image manipulation tools, such as contrast, brightness, image re-
ordering, selection of regions of interest, magnification, and other methods to
provide quantitative measurements from the image.

Workstations must be compliant with the DICOM standard, and be capable of
displaying digital mammograms from all DICOM-compliant acquisition systems.
Workstations must accommodate alternative acquisition schema that provide
complementary information (e.g., CAD, stereotaxy, tomosynthesis, digital
subtraction angiography, dual energy subtraction, etc.).  Workstation design also
should facilitate interfacing the digital system with radiology/hospital
information systems (RIS/HIS).

Workstations should also support quality control functions for the digital
mammography acquisition system as well as for displays, and facilitate
quantitative use of digital image data for quality control testing.  This would
allow objective testing of imaging performance and might reduce some of the costs
currently associated with quality control.

B.  Examples of Research Priorities for Workstation Software and Design

1.  Model radiologists' viewing and work patterns in both screening and
diagnostic environments so that critical parameters of work flow can guide
workstation design.

2.  Develop image management and navigation software based on the above modeling.

3.  Define default softcopy image parameters (e.g., contrast resolution, spatial
resolution, maximum luminance, background luminance, and system contrast).

4.  Develop display-specific compensations to permit fidelity of default images
(i.e., compensate for variables such as absolute luminance, luminance non-
uniformity, veiling glare, dynamic range, distortion, noise, modulation transfer,
luminance range, and acquired image size so that default images are the same
regardless of specific display device).

5.  Develop quality control techniques to assure fidelity of standard images
regardless of specific displays.

6.  Develop and evaluate feature specific enhancement algorithms (e.g., for
calcifications, masses or architectural distortion).

7.  Expand work on current CAD algorithms to increase sensitivity and specificity

8. Improve display controllers to provide greater speed of image manipulation,
application of nonlinear lookup tables, image zooming and re-positioning, etc.

9. Evaluate utility and image quality of various methods of image compression for
storage and data transfer.

10. Utilize findings from studies such as those suggested above and state-of-the-
art technologies to assemble softcopy workstations for digital mammography for
both screening and  diagnosis.  Incorporate above work on image perception,
feature-specific image processing, CAD and user control issues into workstation
design. Evaluate  impact on diagnostic accuracy, time efficiency, cost, reader
fatigue, and satisfaction of search in both screening and diagnostic
environments.

Summary

Broadly, it is desired to improve and enhance the detection of breast cancer with
digital mammography by improving the entire system of softcopy reading of
mammograms.  This includes softcopy displays, the workstation design, ergonomics,
and an increased understanding about what properties of the display and
workstation are important for accurate diagnosis.  The examples mentioned above
are not inclusive; investigators are invited to include their own research
priorities.

INCLUSION OF WOMEN AND MINORITIES IN RESEARCH INVOLVING HUMAN SUBJECTS

It is the policy of the NIH that women and members of minority groups and their
subpopulations must be included in all NIH supported biomedical and behavioral
research projects involving human subjects, unless a clear and compelling
rationale and justification is provided that inclusion is inappropriate with
respect to the health of the subjects or the purpose of the research.   This
policy results from the NIH Revitalization Act of 1993 (Section 492B of Public
Law 103-43).

All investigators proposing research involving human subjects should read the
"NIH Guidelines For Inclusion of Women and Minorities as Subjects in Clinical
Research," which have been published in the Federal Register of March 20, 1994
(FR 59 14508-14513) and in the NIH Guide for Grants and Contracts, Volume 23,
Number 11, March 18, 1994 available on the web at the following URL address:
https://grants.nih.gov/grants/guide/notice-files/not94-100.html

Investigators also may obtain copies of the policy from the program staff listed
under INQUIRIES.  Program staff may also provide additional relevant information
concerning the policy.

INCLUSION OF CHILDREN AS PARTICIPANTS IN RESEARCH INVOLVING HUMAN SUBJECTS

It is the policy of NIH that children (i.e., individuals under the age of 21)
must be included in all human subjects research, conducted or supported by the
NIH, unless there are clear and compelling scientific and ethical reasons not to
include them.  This policy applies to all initial (Type 1) applications submitted
for receipt dates after October 1, 1998.

All investigators proposing research involving human subjects should read the
"NIH Policy and Guidelines on the Inclusion of Children as Participants in
Research Involving Human Subjects" that was published in the NIH Guide for Grants
and Contracts, March 6, 1998, and is available at the following URL address:
http://www.nih.gov/grants/guide/notice-files/not98-024.html

APPLICATION PROCEDURES

OMNIBUS SOLICITATIONS for both the SBIR and STTR programs are available
electronically through the NIH, Office of Extramural Research "Small Business
Funding Opportunities" web site:  http://www.nih.gov/grants/funding/sbir.htm. 
Hard copies, subject to availability, may be obtained from the PHS SBIR/STTR
Solicitation Office, phone (301)206-9385; FAX (301) 206-9722; email
[email protected].  Helpful information in preparation of the application can be
obtained: https://grants.nih.gov/grants/funding/sbirgrantsmanship.pdf

Applications in response to this PA are to be submitted on the grant application
form PHS 6246-1 (1/99) for SBIR Phase I
[http://www.nih.gov/grants/funding/sbir1/sbir.htm], PHS 6246-3 (1/99) for STTR
Phase I [http://www.nih.gov/grants/funding/sttr1/sttr.htm], PHS 6246-2 (1/99)
for SBIR Phase II [http://www.nih.gov/grants/funding/sbir2/index.htm], and PHS
6246-4 (1/99) for STTR Phase II
[http://www.nih.gov/grants/funding/sttr2/index.html].  The applications are also
located in the back pages of the OMNIBUS SOLICITATIONS.  The title and number of
this PA must be typed in line 2 on the face page of the application.

The prohibition of appendices to SBIR/STTR applications, which is stated in the
OMNIBUS SOLICITATIONS, is waived for this PA.  This waiver is a single case
deviation, but still within the mainstream of NIH grant application procedure. 
Specifically for this PA, applicants may submit glossy photographs or other
supporting documentation of the quality of potential "Development of Digital
Mammography Displays and Workstations."  After submission of the application, and
before submission of appendices, applicants should contact the CSR Scientific
Review Administrator assigned to the review to determine the number of copies of
these appendix materials to be sent.

Potential applicants are encouraged to contact program staff for guidance and to
read advice and information on the web sites.  However, responsibility for
planning, direction, and execution of the proposed research will be solely that
of the applicant.

As stated in the MECHANISM OF SUPPORT section, applications may be submitted for
Phase I alone (R41/43), Phase II (R42/44) alone if there has been previous and
successful Phase I support, or through the FAST TRACK mechanism. Application
instructions specified in the OMNIBUS SOLICITATIONS for each mechanism must be
followed.

The normal level of support and period of time for a Phase I SBIR award is
$100,000 and six months; for Phase II SBIR award, $750,000 and two years.  The
normal level of support and period of time for a Phase I STTR awards is $100,000
and one year; for a Phase II STTR award is $500,000 and two years.  However,
applicants may propose longer periods of time and greater amounts of funds if
necessary for completion of the project.  (See NIH Guide, February 12, 1998;
http://www.nih.gov/grants/guide/notice-files/not98-014.html)

Applicants who plan to submit a Phase II SBIR or STTR application requesting
$500,000 or more per year are advised that it is important that they contact
program staff listed in the OMNIBUS SOLICITATIONS as they begin to develop plans.
Applications received without prior staff contact may be delayed in the review
process or returned to the applicant without review (NIH GUIDE, Volume 22, Number
45, December 17, 1993).

Phase I, FAST-TRACK applications must specify clear, measurable goals that should
be achieved prior to Phase II funding.  Failure to provide measurable goals and
sufficient detail in the Phase II application may be sufficient reason for the
peer review committee to exclude the Phase II application from consideration.

Phase II applications submitted in response to this PA will only be accepted as
continuations of previously funded Phase I grants.  The Phase II proposal must
be a logical extension of the Phase I research but not necessarily a Phase I
supported in response to this PA.

The small business concern must also submit a concise Product Development Plan
(limited to ten pages) as an Appendix to the Phase II application addressing the
four areas described in the instructions for FAST TRACK applications in the
OMNIBUS SOLICITATIONS.  Applicants are encouraged to seek commitment(s) of funds
and/or resources from an investor or partner organization for commercialization
of the product(s) or services(s) resulting from the SBIR grant.  See the web site
for details http://www.nih.gov/grants/funding/sbir.htm

The completed original application and five legible copies must be sent or
delivered to:

CENTER FOR SCIENTIFIC REVIEW
NATIONAL INSTITUTES OF HEALTH
6701 ROCKLEDGE DRIVE, ROOM 1040, MSC 7710
BETHESDA, MD  20892-7710
BETHESDA, MD  20817 (for express/courier service)

REVIEW CONSIDERATIONS

Review procedure

Upon receipt, applications will be reviewed by the CSR for completeness and for
responsiveness.  Applications not adhering to application instructions described
above and those applications that are incomplete will be returned to the
applicant without review.

Applications will be evaluated for scientific and technical merit convened in
accordance with the standard NIH peer review procedures.  As part of the initial
merit review, all applications will receive a written critique which will be sent
to the Principal Investigator.  Following scientific-technical review, the
applications will receive a second-level review by the appropriate advisory
council.

Review Criteria

Review criteria are as described in the OMNIBUS SOLICITATIONS:

1.  The soundness and technical merit of the proposed approach.

2.  The qualifications of the proposed principal investigator, supporting staff,
and consultants.

3.  The scientific, technical, or technological innovation of the proposed
research.

4.  The potential of the proposed research for commercial application.

5.  The appropriateness of the budget requested.

6.  The adequacy and suitability of the facilities and research environment.

7.  Where appropriate, the adequacy of assurances detailing the proposed means
for (a) safeguarding human or animal subjects and/or (b) protecting against or
minimizing any adverse effect or the environment.

The Phase I application should specify clear, measurable goals (milestones) that
should be achieved prior to initiating Phase II. Failure to provide clear,
measurable goals may be sufficient reason for the study section to judge the
application non-competitive.

The reviewers will also examine: the appropriateness of the proposed project
budget and duration; the adequacy of plans to include both genders and minorities
and their subgroups as appropriate for the scientific goals of the research and
plans for the recruitment and retention of subjects; the provisions for the
protection of human and animal subjects; and the safety of the research
environment; reviewers will be instructed to address the adequacy of plans for
including children as appropriate for the scientific goals of the research, or
justification for exclusion.

AWARD CRITERIA

Applications will compete for available funds with all other approved SBIR and
STTR applications.  Funding decisions for Phase I or Phase II applications will
be based on quality of the proposed project as determined by peer review,
availability of funds, and program priority.

FAST-TRACK, Phase II applications may be funded following submission of the Phase
I progress report and other documents necessary for continuation.  Phase II
applications will be selected for funding based on the initial priority score,
the awarding Institute's assessment of the Phase I progress and determination
that Phase I goals were achieved, the project's potential for commercial success,
and the availability of funds.

INQUIRIES

Inquiries are encouraged.  The opportunity to clarify any issues or questions
from potential applicants is welcome.

Direct inquiries regarding programmatic issues to:

Barbara Y. Croft, Ph.D.
Diagnostic Imaging Program
National Cancer Institute
6130 Executive Boulevard, Room 800
Bethesda, MD  20892
Telephone:  (301) 496-9531
FAX:  (301) 480-5785
Email:  [email protected]

Direct inquiries regarding fiscal matters to:

Ms. Kathleen Shino, MBA
Grants Administration Branch
National Cancer Institute
6120 Executive Boulevard, Room 243
Bethesda, MD  20892-7150
Telephone:  (301) 496-7800 Ext. 248
FAX:  (301) 496-8601
Email:  [email protected]

AUTHORITY AND REGULATIONS

This program is described in the Catalog of Federal Domestic Assistance No.
93.394 (Cancer Detection and Diagnosis Research).  Awards are made under
authorization of the Sections 301 and 405 of the Public Health Service Act, as
amended ( 42 USC 241 and 284) and administered under PHS grants policies and
Federal Regulations 42 CFR 52 and 45 CFR Part 74 and part 92.  This program is
not subject to the intergovernmental review requirements of Executive Order 12372
or Health Systems Agency review.

The PHS strongly encourages all grant and contract recipients to provide a smoke-
free workplace and promote the non-use of all tobacco products.  In addition,
Public Law 103-227, the Pro-Children Act of 1994, prohibits smoking in certain
facilities (or in some cases, any portion of a facility) in which regular or
routine education, library, day care, health care or early childhood development
services are provided to children.  This is consistent with the PHS mission to
protect and advance the physical and mental health of the American people.

Return to Volume Index

Return to NIH Guide Main Index

	Office of Extramural Research (OER)			National Institutes of Health (NIH) 9000 Rockville Pike Bethesda, Maryland 20892			Department of Health and Human Services (HHS)