Aims and objectives
Imaging of the breast exploits a number of different physical properties of breast tissue such as electron density or acoustic impedance.
Dielectric property which determines propagation and absorption of radio waves similarly varies in different tissues of the breast and has been considered for imaging since 1991 [1] and in tumour areas where the cell content is high. The MARIA system is a whole-breast radiofrequency imaging system that exploits the electromagnetic wave scattering contrast between breast tissue types across a 5GHz spectrum to detect lesions...
Methods and materials
System description:
MARIA is a CE-marked radio-frequency (RF) medical imaging system comprising a patient bed,
a Scanning and Data Processing (SDP) unit which is located under the bed (see Figure 1) and a touch-screen console featuring a software user interface that controls both acquisition and review.
The SDP unit contains a hemispherical array of 60 RF antennae which encircle the breast such that uniform transmission and reception is achieved over 1770 bi-static paths and across 101 frequency bins.
The breast lies pendant in the array...
Results
The evaluated studies comprised 17 cases with malignant diagnosis and 31 cases with benign diagnoses (12 Fa and 19 Cy). Figure 4 and Figure 5 show the scatter plots for different combinations of R,
G,
B values.
As can be observed,
the different plots show varyingif clear levels of differentiation betweenmalignant and benign cases.
Specifically,
the best classification corresponds to the R vs B plot (that is,
High against Low frequency magnitude),
which allows to discriminate malignant from benign cases with a PPV of 76%,...
Conclusion
This analysis is encouragingly corroborated byprevious ex-vivo studies which indicated that lesions can be automatically classified based on their frequency dependent dielectric response [4,5,6].
The results presented can be considered a promising proof of concept of how a limited number of parameters extracted from the dielectric radiofrequency response can be used for the automatic discrimination of in-vivo assessed lesions.
Future work includes understanding the behaviour of subtypes within the benign cohort and exploring the use of more elaborated parameters to characterize the frequency response (for...
Personal information
Dr Nick Ridley,
Radiology Department,
Great Western Hospitals NHS Foundation Trust,
Swindon,
SN3 6BB
[email protected]
Dr Ana Iriarte,
Micrima Limited,
One Glass Wharf,
Bristol,
BS2 0EL
[email protected]
Dr Louis Tsui,
Micrima Limited,
One Glass Wharf,
Bristol,
BS2 0EL
[email protected]
Dr Chris Bore,
Micrima Limited,
One Glass Wharf,
Bristol,
BS2 0EL
[email protected]
Dr Mike Shere,
Bristol Breast Care Centre,
Southmead Hospital,
Westbury-on-Trym,
Bristol BS10 5NB
[email protected]
Prof Iain Lyburn,
Thirlestaine Breast Center,
Thirlestaine Road,
Cheltenham,
GL52 7AS
[email protected]
Dr Peter Bannister,
Micrima Limited,
One Glass Wharf,...
References
1. Preece AW,
Robinson MP,
Green JL,
Horrocks M.
Dielectric imaging- an alternative to x-ray mammography? 6th Scientific Meeting of the British Oncological Association (BOA),
Bath,
UK.
1991;64(Supplement XV (P34)):25.
2. Nilavalan R,
Gbedemah A,
Craddock IJ,
Li X,
Hagness SC.
Numerical investigation of breast tumour detection using.
Electronics Letters.
2003;39(25):1787-9.
3. Nilavalan R,
Leendertz J,
Craddock I,
Benjamin R,
Preece A.
Breast tumour detection using a flat 16 element array.
Proceedings of the 16th International Zurich Symposium on Electromagnetic Compatibility-Topical Meeting on Biomedical EMC,...