Interactive Detection, Evaluation and Classification of Breast
Lesions from 4D MRI Using Direct Volume Visualization
  

W.B. Carruthers*, K.R. Subramanian*, J.P. Brockway**

*Dept. of Computer Science, UNC Charlotte, Charlotte, NC 28223
**Memory Testing Corporation, Davidson, NC 28036; Presbyterian HealthCare, 
 Charlotte, NC 28207.
  

Mammography is currently regarded as the most effective method for early 
detection of breast cancer, but recently its lower sensitivity in 
certain high risk cases has been less than desired. Preliminary 
investigation of these mammographically and ultrasonically occult 
lesions using high resolution MRI dynamic T1 weighted protocols 
suggests that it is possible to detect and characterize such tumors.
Using inexpensive graphics hardware, we have developed 
interactive visualization tools to rapidly process, visualize
and quantify lesions from 4D MRI (MRI volume + time). Our approach
determines a confidence measure for each voxel, representing the probability 
that the voxel is part of the tumor, using a rough goodness-of-fit 
for the shape of the intensity-time curves. Our system
takes advantage of readily available 3D texture mapping hardware to
produce both 2D and 3D visualizations of the segmented MRI volume in 
real-time, allowing enhanced spatial perception of the tumor shape, size, 
distribution, and other characteristics useful in staging and 
treatment courses.  Our tumor detection and visualization system 
permitted (1) volumetric analysis of lesions, (2) identification of 
shape characteristics, (3) tumor extent , (4) tumor size and (5) location 
of the four most common types of breast tumor, in mammographically and 
ultrasonically occult lesions in women who were at high risk.



Figure Caption:

4D MRI Breast Tumor Segmentation revealing ductal cancer in situ 
and ductal cancer invasive. (Left) Coronal, Axial, Sagittal Slice views,
(Right) 3D Volume Rendering.