Home

1
Home
Welcome
1
2
3
4
5
6
6
7
8
Imaging in Cancer Care Cancer care is becoming
more sophisticated, increasing the need for
accurate and detailed information about
individual patients to inform treatment planning
Increasing cancer incidence in an ageing
population combined with constraints on health
care expenditure places limitations on the number
of additional procedures that patients can
undergo TexRAD maximises the information that
can be obtained from the diagnostic images
routinely acquired in current clinical practice
and does not require additional procedures
HOW?
2
What is TexRAD?
What is TexRAD?
1
2
3
4
5
6
6
7
8
TexRAD is a software application that analyses
the textures in existing radiological scans to
assist the clinician in assessing the prognosis
of the cancer patient.
This presentation will guide you through how
TexRAD will improve patient care. Please use
links on left hand side to go directly to
particular sections.
Screen shot of TexRAD software
3
Benefits
Benefits - Patients
Patients Clinicians OEM
4
Benefits Patients
Benefits - Patients
Patients Clinicians OEM

• For Patients - TexRAD
• is non-invasive and minimises discomfort
• may reduce requirement for additional complex
  invasive procedures and exposure to ionising
  radiation
• - may assist with optimisation of cancer care
  pathways

5
Benefits Clinicians
Benefits - Clinicians
Patients Clinicians OEM

• For Clinicians TexRAD is
• - designed to fit into existing workflow
• providing more information than standard
  visualisation of patient
• scans to assist in decision-making
• able to provide a Risk Stratification during
  cancer care
• enabling treatment to be targeted according to
  prognosis
• - a novel cancer imaging biomarker (software)
• - efficient as it uses routine clinical images
• cost effective as can be installed on existing
  hospital imaging
• equipment and PACS

6
Benefits OEMs
Benefits - OEMs
Patients Clinicians OEM

• For Imaging PACS OEMs (Original Equipment
  Manufacturers)
• Novel technology with no current competitors
  (patents pending)
• - Software only - no hardware modifications
  required
• An extra product for all your existing imaging
  clients
• Unique software to enhance your portfolio
  differentiator
• Can be made available to your clients as an
  add-on or bundled
• with your current software offerings

7
Clinical Case Studies
Clinical Application
Colorectal cancer (Liver / Distant Metastatic
Disease) Case Study Clinical Demo
Evidence Lung cancer Case Study Clinical Demo
Evidence Breast cancer Case Study Clinical
Demo Evidence
8
Colorectal Homepage
Colorectal Cancer
Case Study Demo Evidence
PACS workstation
(Liver / Distant Metastatic Disease)
9
Colorectal Case Study
Colorectal Case Study
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
Typical example A patient visits a clinic after
curative colorectal cancer surgery She undergoes
a routine follow-up CT scan The Radiologist
considers that the CT looks normal with no focal
abnormalities However, 18 months later the
patient relapses with focal metastatic disease of
the liver fatal consequence TexRAD could have
assisted the radiologist to improve this scenario
as part of routine clinical procedure
HOW?
10
Colorectal CS 2
Colorectal Case Study
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
How TexRAD supports clinician patient in case
study
From the routine CT scans taken in the clinic,
TexRAD software uniquely extracts and measures
fine, medium and coarse textures - in this
example, from the Liver CT. TexRAD highlights
texture anomalies which are not apparent to
normal visual examination These texture
anomalies can be used to predict the risk of
metastatic disease From this additional
information, radiologist may suggest alternative
treatment pathways to the patient
11
Colorectal Clinic Demonstration
Colorectal - Demo
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
PACS workstation
TexRAD screen shot of Liver
12
Colorectal Demo 2
Colorectal - Demo
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
Screenshot - Liver TexRAD analysis of apparently
normal appearing liver (after curative surgery of
primary tumour) as seen on follow-up CT of a
patient with colorectal cancer could predict the
risk of metastatic disease
13
Colorectal Demo 3
Colorectal - Demo
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
Screenshot - Work flow demonstration sequence for
Liver
14
Colorectal Demo 4
Colorectal - Demo
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
STAGE 1 - Display the target clinical image of
interest A TexRAD analysis is applied to the
appropriate 2D CT image highlighting the liver
(tissue of interest - TOI). The specialist
clinical consultant (e.g. Radiologist) will
select the image containing this TOI.
15
Colorectal Demo 5
Colorectal - Demo
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
STAGE 2 Draw region of interest (ROI) to be
analysed Using TexRADs graphical user interface
tools, image window level/width can be altered to
clearly delineate this TOI, interactive
magnification/panning/centring can be used for
better visualization of this TOI. Clinician can
choose an appropriate ROI tool (e.g. Polygon ROI)
from a list of options based on the application.
This ROI is super-imposed on the TOI within the
original image.
16
Colorectal Demo 6
Colorectal - Demo
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
STAGE 3 Texture Analysis TexRAD employs a
novel algorithm (patent applied for) primarily to
extract subtle but prognostic metrics currently
not available in clinic. The software also
graphically displays clinically relevant fine,
medium and coarse liver textures separately
(below) in addition to their fusion with the
original CT image.
17
Colorectal Demo 7
Colorectal - Demo
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
STAGE 4 Risk Stratification Report A risk
stratification report specific to the colorectal
cancer is generated, which should be used only to
assist the clinician to make an accurate
decision. The report contains patient ID and
scan details, TexRAD analysis result, explanation
and contact information.
18
Colorectal Background 1
Colorectal Evidence
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
Facts about Colorectal Cancer Colorectal cancer
is the second most common malignancy in Western
societies 1. 40 of the patients undergoing
resection of the primary tumour will relapse and
die of their disease, making colorectal cancer
the second leading cause of death related to
cancer 2. The liver is the sole site of
secondary tumour spread (i.e. metastasis) in
20-40 of patients 3 and therefore it is a
common practice to follow-up patients after their
curative resection 4. There is an overall
survival benefit for intensifying the follow-up
of such patients with imaging of the liver being
associated with reduced mortality (Odds ratio
0.66, 95 confidence limits 0.46-0.95) 5.
19
Colorectal Background 2
Colorectal Evidence
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
Facts about Colorectal Cancer The American
Society of Clinical Oncology (ASCO) now
recommends annual CT of the chest and abdomen for
3 years after primary therapy for patients at
higher risk of recurrence 6. However, the risk
of recurrence is not uniform for these patients
and identification of predictive factors that are
linked to outcomes may allow modification of
surveillance strategies for particular sub-groups
and provide effective and optimum patient care.
ASCO has also highlighted the need for research
in this area 6.
20
Colorectal Background 3
Colorectal Evidence
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
Comparison of Risk Stratification
techniques Additional physiological imaging
techniques such as Doppler ultrasound and
quantitative analysis of liver contrast
enhancement or perfusion on CT also have the
potential to identify patients at higher risk of
recurrence 7, 8 These techniques may reflect
alterations of liver hemodynamics associated with
occult metastases 9, 10. However, the
additional image acquisitions required by these
techniques have been a barrier to their adoption
into surveillance programs. (cost/additional
radiation dosage)
21
Colorectal Evidence 1
Colorectal Evidence
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
Why TexRAD is better - evidence TexRAD applied
to routinely acquired CT images highlights subtle
liver changes that may occur in association with
alterations in liver physiology. Furthermore
computer simulations and clinical studies have
suggested that measurements of liver texture on
CT may reflect liver vascularity 11-13. The
ability of Liver TexRAD to detect occult tumour
on CT is further supported by studies
demonstrating textural differences between normal
livers and apparently normal areas of tissue
within livers bearing tumours, areas also known
to exhibit alterations in blood flow 13-15.
22
Colorectal Evidence 2
Colorectal Evidence
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
Why TexRAD is better - evidence Preliminary
results of Liver TexRAD demonstrate the potential
for liver texture on contrast-enhanced CT to
provide a novel parameter that is not only
insensitive to variations in acquisition
parameters but can also act as a marker of
survival for patients following resection of
colorectal cancer 16.
23
Colorectal Evidence 3
Colorectal Evidence
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8

• Why TexRAD is better evidence
• Radiology 2008 In Press

Kaplan-Meier survival curves for colorectal
cancer patients with normal liver appearances on
conventional CT separated by (A) Liver Texture
analysis (HTA) and (B) Liver Perfusion Index
(HPI). Survival curves were significantly
different for HTA (plt0.005) 16.
24
Lung Homepage
Lung Cancer
Case Study Demo Evidence
PACS workstation
TexRAD screen shot of Lung lesion
25
Lung Case Study
Lung Case Study
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
Typical example A patient diagnosed with lung
cancer requires an accurate staging of the tumour
and lymph node disease. Radiologists are unable
to obtain a confident risk stratification from CT
alone, which is critical for early prognosis and
favourable patient outcome. The patient
undergoes FDG PET-CT imaging, which overcomes the
limitations of CT alone. However, this is an
additional and expensive procedure. TexRAD can
improve the performance of CT and potentially be
employed for selection of patients for PET.
HOW?
26
Lung CS 2
Lung Case Study
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
How TexRAD supports clinician patient in case
study
From routine CT scans taken in the clinic TexRAD
software uniquely extracts and measures fine,
medium and coarse textures - in this example,
from the Lung lesion on CT - and classifies
degree of adverse tumour biology These texture
gradation can be used to predict tumour stage and
the risk of metastatic and lymph node
disease Based on this additional information,
radiologist may optimally select patients who
will benefit from FDG PET
27
Lung Clinic Demonstration
Lung TexRAD Demo
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
PACS workstation
TexRAD screen shot of Lung lesion
28
Lung Demo 2
Lung - Demo
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
Lung TexRAD analyses a focal cancerous lesion as
seen on the conventional CT image of a patient
with NSCLC could predict tumour stage, metabolism
and lymph node disease involvement
29
Lung Demo 3
Lung - Demo
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
30
Lung Demo 4
Lung - Demo
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
STAGE 1 - Display the target clinical image of
interest A TexRAD analysis is applied to the
appropriate 2D CT image highlighting the lung
lesion (tissue of interest - TOI). The
specialist clinical consultant (e.g. Radiologist)
will select the image containing this TOI.
31
Lung Demo 5
Lung - Demo
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
STAGE 2 Draw region of interest (ROI) to be
analysed Using TexRADs graphical user interface
tools, image window level/width can be altered to
clearly delineate this TOI, interactive
magnification/panning/centring can be used for
better visualisation of this TOI. Clinician can
choose an appropriate ROI tool (e.g. Elliptical
ROI, which encloses the TOI in an automated
fashion) from a list of options based on the
application. This ROI is super-imposed on the TOI
within the original image.
32
Lung Demo 6
Lung - Demo
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
STAGE 3 Texture Analysis TexRAD employs a
novel algorithm (patent applied for) primarily to
extracts subtle but prognostic metrics currently
not available in clinic. The software also
graphically displays clinically relevant fine,
medium and coarse lung lesion textures separately
(below) in addition to their fusion with the
original CT image.
33
Lung Demo 7
Lung - Demo
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
STAGE 4 Risk Stratification Report A risk
stratification report specific to the lung cancer
is generated, which should be used only to assist
the clinician to make an accurate decision. The
report contains patient ID and scan details,
TexRAD analysis result, explanation and contact
information.
34
Lung Background
Lung - Background
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
Facts about lung Cancer Lung cancer is the most
common form of death related to cancer in men and
second most common in woman 1, 2, accounting
for 1.3 million deaths worldwide annually
3. Non-small cell lung carcinoma (NSCLC) is the
most common form of lung cancer prevalent in 80
of all cases 4. Following initial diagnosis,
patients with NSCLC undergo staging. The most
common imaging staging procedure has been CT.
However, due to low accuracy for CT staging,
clinical guidelines now recommend
Fluoro-deoxy-glucose (FDG) PET-CT unless the
initial CT imaging shows evidence of inoperable
disease. An improvement in the accuracy of CT
could improve the selection of patients for
FDG-PET.
35
Breast Homepage
Breast Cancer
Case Study Demo Evidence
TexRAD screen shot of Breast lesion
36
Breast Case Study
Breast Case Study
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
Typical example A female participant within a
breast screening program undergoes an initial
mammogram examination. Further diagnostic
mammogram confirms presence of cancerous tissue.
Core biopsy identified non-invasive ductal
carcinoma in situ (DCIS) and the standard breast
conservation excision was performed. Surprisingly
, the final excised specimen provided evidence of
invasive focus, routinely underestimated by core
biopsy, resulting in additional surgical
procedure involving the axilla TexRAD could have
predicted the risk of invasive disease
preoperatively from mammographic lesions
assisting in treatment planning and optimal
selection of biopsy or surgery
HOW?
37
Breast CS 2
Breast Case Study
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
How TexRAD supports clinician patient in case
study
From routine mammographic images taken in the
clinic TexRAD software uniquely extracts and
measures fine, medium and coarse textures - in
this example, from the breast lesion - and
characterises differences in calcification
architecture These texture differences can be
used to preoperatively estimate the likelihood of
invasive focus from among patients with
DCIS This additional information may assist the
clinician in better treatment planning and
optimal selection of sentinel node biopsy or
axillary surgery
38
Breast Clinic Demonstration
Breast - Demo
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
TexRAD screen shot of Breast lesion
39
Breast Demo 2
Breast - Demo
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
Breast TexRAD analysis of an obvious stellate
mass as seen on a mammogram of a patient with
core-biopsy proven breast cancer could estimate
the risk of invasive disease
40
Breast Demo 3
Breast - Demo
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
41
Breast Demo 4
Breast - Demo
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
STAGE 1 - Display the target clinical image of
interest A TexRAD analysis is applied to the
appropriate 2D mammographic image highlighting
the breast lesion (tissue of interest - TOI). The
specialist clinical consultant (e.g. Radiologist)
will select the image containing this TOI.
42
Breast Demo 5
Breast - Demo
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
STAGE 2 Draw region of interest (ROI) to be
analysed Using TexRADs graphical user interface
tools, image window level/width can be altered to
clearly delineate this TOI, interactive
magnification/panning/centring can be used for
better visualisation of this TOI. Clinician can
choose an appropriate ROI tool (e.g. Polygon ROI)
from a list of options based on the application.
This ROI is super-imposed on the TOI within the
original image.
43
Breast Demo 6
Breast - Demo
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
STAGE 3 Texture Analysis TexRAD employs a
novel algorithm (patent applied for) primarily to
extracts subtle but prognostic metrics currently
not available in clinic. The software also
graphically displays clinically relevant fine,
medium and coarse breast lesion textures
separately (below) in addition to their fusion
with the original mammographic image.
44
Breast Demo 7
Breast - Demo
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
STAGE 4 Risk Stratification Report A risk
stratification report specific to the breast
cancer is generated, which should be used only to
assist the clinician to make an accurate
decision. The report contains patient ID and
scan details, TexRAD analysis result, explanation
and contact information.
SAMPLE REPORT ONLY FOR ILLUSTRATION
45
Breast Background 1
Breast - Background
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
Facts about Breast Cancer Invasive cancer of the
breast and DCIS can both present as focal lesions
on mammography. Pure DCIS is not an invasive
process and rarely metastasises to regional lymph
nodes 1. The introduction of mammographic
breast screening has resulted in a dramatic
increase in the diagnosis of DCIS and its
detection rate has reached 15-20 of all
mammographically detected cancers 2-4. Based
on core biopsy specimens, the standard surgical
planning for DCIS is to offer breast conservation
excision.
46
Breast Background 2
Breast - Background
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
Facts about Breast Cancer However, the detection
of DCIS on core biopsy is quite frequently
followed by evidence of invasion within the final
excision specimen. This occurs in 11-44 of
patients and results in the need of second
operative procedure involving the axilla 4, 5.
Therefore the main concern is whether any kind
of axillary staging is indicated in patients
preoperatively diagnosed with DCIS only.
47
Breast Background 3
Breast - Background
Case Study Demo Evidence
1
2
3
4
5
6
6
7
8
Facts about Breast Cancer The most often
employed axillary treatment option for patients
with DCIS is to perform a sentinel node biopsy at
the time of initial excision. Nevertheless it
may lead to reoperation in cases of proven
invasive focus on excision. Therefore an
effective way of estimating the likelihood of an
invasive focus preoperatively in patients
diagnosed with DCIS would assist in better
treatment planning and optimal use of sentinel
node biopsy or axillary surgery.
48
License
License Opportunities

• License
• The software and interface as shown in this
  presentation exist and are fully working. All
  screenshots are from real case studies. However,
  this is an early stage product and we are open to
  a wide range of options including licensing and
  ongoing collaborative development.
• We are seeking collaboration (licensing
  opportunities) with Imaging equipment, software
  and PACS manufacturers for integration and
  commercialisation of TexRAD technology.
  Interested parties please CONTACT US

49
Project Team
Project Team
Principal Developer Dr. Balaji Ganeshan Clinical
Collaborator Prof. Kenneth Miles Image
Processing Collaborators Dr. Rupert Young Prof.
Chris Chatwin
Business Development Manager Mike
Wylde Intellectual Property Technology
Transfer Office Russell Nicholls
50
Contact
Contact
Dr Balaji Ganeshan or Michael Wylde TexRAD Resear
ch and Enterprise University of Sussex Falmer,
Brighton UK BN1 9SB Tel [phone]
800 Email TexRAD_at_sussex.ac.uk
51
Disclaimer
Disclaimer
Notice TexRAD is meant to be used as an
assistive risk stratification tool in clinical
practice and should not form the basis for
clinical decision making. The results obtained
from TexRAD system should only be interpreted by
a health care professional. IMPORTANT TexRAD is
being further evaluated and not yet available for
routine clinical practice.
52
References
References
Colorectal Lung Breast
Supporting evidence and academic references
53
References Liver 1
References - Liver Cancer
Colorectal Lung Breast
1
2
3
4
5
6
6
7
8

• COLORECTAL (Liver / Distant Metastatic Disease)
• Jemal A, Murray T, Ward E, Samuels A, Tiwari RC,
  Ghafoor A, Feuer EJ, Thun MJ. Cancer statistics,
  2005. CA Cancer J Clin, 2005 5510-30.
• Obrand D, Gordon P. Incidence and patterns of
  recurrence following curative resection of colon
  cancer. Dis Colon Rectum 1997 4015-24.
• Morgan-Parkes JH. Metastases mechanisms,
  pathways, and cascades. Am J Roentgenol. 1995
  164(5)1075-82.
• Johnson FE, Virgo KS, Fossati R. Follow-up for
  patients with colorectal cancer after
  curative-intent primary treatment. Clin Oncol.
  2004 221363-5
• Jeffrey GM, Hickey BE, Hider P. Follow-up
  strategies for patients treated for
  non-metastatic colorectal cancer (Cochrane
  Review). In The Cochrane Library, Issue 2 2003.
  Oxford Update Software.
• Desch CE, Benson AB 3rd, Somerfield MR, Flynn PJ,
  Krause C, Loprinzi CL, Minsky BD, Pfister DG,
  Virgo KS, Petrelli NJ American Society of
  Clinical Oncology. Colorectal cancer
  surveillance 2005 update of an American Society
  of Clinical Oncology practice guideline. J Clin
  Oncol. 2005 238512-9.
• Leen E, Angerson WG, Cooke TG, McArdle CS.
  Prognostic power of Doppler perfusion index in
  colorectal cancer. Correlation with survival. Ann
  Surg. 1996 223199-203.
• Platt JF, Francis IR, Ellis JH, Reige KA. Liver
  metastases Early detection based on abnormal
  contrast material enhancement at dual-phase
  helical CT. Radiology 1997 20549-53.

54
References Liver 2
References - Liver Cancer
Colorectal Lung Breast
1
2
3
4
5
6
6
7
8

• COLORECTAL (Liver / Distant Metastatic Disease)
• Cuenod CA, Leconte I, Siauve, N et al. Early
  changes in liver perfusion caused by occult
  metastases in rats detection with quantitative
  CT. Radiology 2001 218 556-561.
• Kruskal JB, Thomas P, Kane RA, Goldberg SN.
  Hepatic perfusion changes in mice livers with
  developing colorectal cancer metastases.
  Radiology 2004 231482-490.
• Bezy-Wendling J, Kretowski M, Rolland Y, Le Bidon
  W. Towards a better understanding of texture in
  vascular CT scan simulated images. IEEE Trans
  Biomed Eng. 2001 48120-4.
• Ganeshan B, Miles KA, Young RCD and Chatwin CR.
  In search of biologic correlates for liver
  texture on portal-phase CT. Acad Radiol 2007
  14(9)1058-68.
• Ganeshan B, Miles KA, Young RCD and Chatwin CR.
  Hepatic enhancement in colorectal cancer Texture
  analysis correlates with hepatic hemodynamics and
  patient survival. Acad Radiol 2007
  14(12)1520-30.
• Ganeshan B, Miles KA, Young RCD and Chatwin CR.
  Hepatic entropy and uniformity additional
  parameters that can potentially increase the
  utility of contrast enhancement during abdominal
  CT. Clin Radiol 2007 62(8) 761-768.
• Tsushima Y, Blomley MJK, Yokoyama H, Kusano S,
  Endo K. Does the presence of distant and local
  malignancy alter parenchymal perfusion in
  apparently disease-free areas of the liver? Dig
  Dis Sci 2001 46(10)2113-2119.
• Miles KA, Ganeshan B, Griffiths MR, Young RCD,
  Chatwin CR. Hepatic computed tomography for
  colorectal cancer Texture analysis of portal
  phase images as a potential marker of survival.
  Radiology, 2008 In Press.

55
References - Lung
References - Lung Cancer
Colorectal Lung Breast
1
2
3
4
5
6
6
7
8

• LUNG
• 1. Deaths by cause, sex and mortality stratum
  (PDF). World Health Organization (WHO) 2004.
• 2. Lung Cancer Facts (Women). National Lung
  Cancer Partnership 2006.
• 3. Cancer. World Health Organization (WHO) 2006.
• 4. Travis, WD Travis LB, Devesa SS (January
  1995). Lung cancer. Cancer 1995 75(1) 191-202.

56
References - Breast
References - Breast Cancer
Colorectal Lung Breast
1
2
3
4
5
6
6
7
8

• BREAST
• 1. Yen TW, Hunt KK, et al. Predictors of Invasive
  Breast Cancer in Patients with an initial
  diagnosis of Ductal Carcinoma in Situ A Guide to
  Selective Use of Sentinel Lymph Node Biopsy in
  Management of Ductal Carcinoma in Situ. J Am Coll
  Surg 2005 200 516-526.
• 2. Lagios MD. Ductal carcinoma in situ. Pathology
  and treatment. Surg Clin North America 1990 70,
  853-871.
• 3. Holland R, et al. Extent, distribution and
  mammographic/ histological correlations of breast
  ductal carcinoma in situ. Lancet 1990, 335,
  519-522.
• 4. Leonard GD, Swain SM. Ductal carcinoma in
  situ, complexities and challenges. Journal of the
  National Cancer Institute 2004 96(12)906-20.
• 5. Dillon MF, et al. Predictors of Invasive
  Disease in Breast Cancer When Core Biopsy
  Demonstrates DCIS Only. Journal of Surg Oncology
  2006 93, 559-563.