Fractals improve detection of breast cancer
8 August 2006
Boca Raton, Fla., USA and Bremen, Germany. Researchers from the USA and
Germany have developed new techniques for the the diagnosis and treatment of
breast cancer. They have developed and piloted a unique software platform
for the analysis and display of dynamic magnetic resonance imaging of the
The researchers are from Florida Atlantic University, the Center for
Breast Care at the Women's Center at Boca Raton Community Hospital, and
MeVis, The Center for Diagnostic Systems and Visualization at the University
of Bremen, Germany.
The software, which has been approved by the US FDA, was developed
at MeVis, The Center for Diagnostics Systems and Visualization, under the
direction of Dr. Heinz-Otto Peitgen, who is also a faculty member in the
department of mathematical sciences at FAU's Charles E. Schmidt College of
Science. Peitgen used the mathematical concept of fractals to begin
developing this unique software.
"Fractals are large, irregular geometric patterns made up of infinitely
smaller, but identical, irregular patterns," said Peitgen. "Fractal theory
provided an appropriate platform upon which to build the software program
because the ducts within human breast tissue have fractal properties."
Breast MRI is a relatively new tool used by physicians to diagnose breast
cancer as an adjunct to conventional mammography. Breast MRI displays the
behaviour of a cancerous lesion in three dimensions and approaches a nearly
100% accuracy rate in the detection of invasive cancer. In contrast,
mammography provides a two-dimensional view of the breast and surrounding
tissue and only detects 80 to 85% of tumours. One of the main strengths of
MRI is its precise delineation of soft tissue and its ability to image the
breast in fine sections dynamically by taking multiple MRI images over time.
The percentage of medical centres doing breast MRI is small, but growing.
A recent study spearheaded by Dr. Kathy Schilling, medical director of
Imaging and Intervention at the Center for Breast Care at the Women's Center
at Boca Raton Community Hospital, was published in The American Journal of
Radiology and entitled Assessment of Suspected Breast Cancer by MRI — A
Prospective Clinical Trial Using a Combined Kinetic and Morphologic
Analysis. Findings of this study showed that in over 30% of patients
there were additional tumours in the same breast, and in almost 10% of the
patients there were tumours in the opposite breast.
"These tumours were not found using mammography or ultrasound," said
Schilling. "We also found a resulting change in the course of treatment in
nearly 25% of patients undergoing surgery for newly diagnosed breast
cancer." In addition, findings from this study showed that MRI directed
biopsies using computational clinical imaging led to definitive conclusions,
demonstrating the clinical utility of this unique approach.
According to Dr. Roger Goldwyn, faculty member in FAU's department of
mathematical sciences and director of the proposed Center for the
Development of Computational Clinical Imaging at FAU, "Our approach is to
use innovative image-processing tools to find additional tumours and to help
determine patient management outcomes. These techniques have also led to
biopsies directed by these image-processing tools, surgical planning
modifications, and monitoring effectiveness of chemotherapy."
"Our researchers at FAU will continue to conduct clinical evaluations and
collaborate with the Center for Breast Care and MeVis to further develop
these tools for use by clinicians in more routine settings in order to have
a wider impact on patient care," said Dr. Larry F. Lemanski, vice president
for research at FAU.
American women have a one in nine chance of developing breast cancer
during their lifetime. Early detection is the single most effective tool in
fighting breast cancer. The sooner the cancer is detected, the more options
a woman has for treatment and the better her chance for survival. Methods
for detection include mammography — a safe, low-dose x-ray of the breast —
that can be a vital tool in helping to discover small lumps up to two years
before they can be felt in a physical exam. With mammography, however, only
80 to 85% of tumours are detected, leaving 15-20% of tumours undetected.
These undetected tumours will enlarge and become more lethal before they are
identified in a later screening or when they become clinically apparent.