Diffusion kurtosis imaging with radiomics model could ID lesions, malignancies

Diffusion kurtosis imaging, a new breast MRI technique that doesn't require a contrast agent and utilizes data analysis, could reduce unnecessary breast biopsies and expand treatment options, according to a recent release from the Radiological Society of North America (RSNA).  

The new research, published online Feb. 20 in Radiology, demonstrates how eliminating contrasts agents using diffusion-weighted imaging measurement techniques can provide high-quality images of breast tissue and differentiate between benign lesions and malignancies. 

Researchers from the German Cancer Research Center in Heidelberg analyzed data from women classified under the Breast Imaging Reporting and Data System (BI-RADS) with category four and five breast lesions. A software algorithm and kurtosis-based radiomics were then used to extract lesion characterization and imaging features for data.  

"Diffusion kurtosis imaging has been introduced in diffusion weighted imagingto provide important information on tissue structures at a microscopic level," said lead author Sebastian Bickelhaupt, MD, from the German Cancer Research Center, in a prepared statement. "Since malignant lesions disrupt the tissue structures at this level, diffusion kurtosis might serve as a relevant marker of changes."  

According to study results, the radiomics analysis reduced false-positive findings by 70 percent in a study group of 127 women, while detecting 98 percent of malignant lesions.  

"The model might help to lower the number of BI-RADS 4 lesions suspected of being cancer on the basis of screening mammography while retaining a high sensitivity similar to the sensitivity reported for biopsies themselves," co-author Paul Jaeger, MSc, said in a prepared statement.