Histogram analysis of breast diffusion kurtosis imaging: a comparison between readout-segmented and single-shot...

Area of Science:

• Radiology
• Medical Imaging
• Oncology

Background:

• Histogram analysis of diffusion-weighted imaging (DWI) parameters is established for differentiating breast lesions.
• Comparison of diffusion-kurtosis imaging (DKI) parameters between single-shot echo-planar imaging (ss-EPI) and readout-segmented echo-planar imaging (rs-EPI) sequences in breast cancer has not been previously investigated.
• This study addresses the need to evaluate the diagnostic utility of histogram parameters from different DKI sequences for breast lesion characterization.

Purpose of the Study:

• To investigate the diagnostic accuracy and reliability of histogram parameters derived from rs-EPI and ss-EPI sequences of DKI parameters.
• To compare the effectiveness of rs-EPI and ss-EPI sequences in distinguishing between benign and malignant breast lesions using histogram analysis.
• To determine which DKI sequence offers superior performance for breast lesion characterization.

Main Methods:

• A retrospective cohort study involving 205 patients with breast lesions (65 benign, 140 malignant) who underwent 3T breast MRI.
• Acquisition of DKI data using both rs-EPI and ss-EPI sequences with b values of 0, 50, 1,000, and 2,000 s/mm².
• Manual delineation of regions of interest (ROIs), extraction of histogram parameters from apparent diffusion coefficient (ADC) and apparent diffusional kurtosis (K_app) maps, and statistical analysis including ROC curves.

Main Results:

• Histogram parameters (mean, percentiles, skewness, kurtosis) derived from D_app and K_app maps showed good to excellent intra-observer agreement for both sequences.
• Malignant lesions exhibited significantly lower D_app and higher K_app values compared to benign lesions for both rs-EPI and ss-EPI sequences (P<0.05).
• The rs-EPI sequence demonstrated superior diagnostic performance compared to ss-EPI for most histogram parameters, with the 75th percentile of K_app and 25th percentile of D_app showing the highest AUC values.

Conclusions:

• Histogram parameters derived from the rs-EPI sequence are more reliable and accurate for differentiating malignant from benign breast lesions than those from the ss-EPI sequence.
• The rs-EPI sequence offers enhanced diagnostic value in breast cancer assessment using DKI histogram analysis.
• These findings support the use of rs-EPI sequences for improved characterization of breast lesions.