Role of Diffusion MRI in Characterizing Benign and Malignant Breast Lesions

Authors

  • Zaeema Ali, Kynat Rehman, Anam Anwaar, Mahwash Shoaib, Asim Ali, Nadeem Akram Butt

DOI:

https://doi.org/10.53350/pjmhs2023172664

Abstract

Background: Diffusion-weighted MRI imaging (DWI) is frequently employed, particularly to assess organs such as breast, prostate, ovaries, liver, and pancreas.

Aim: The present study aimed to assess the diffusion MRI role in characterization benign breast lesions from malignant breast lesions.

Patients and Methods: This prospective study was conducted on 260 breasts lesions of 200 patients in the Department of Radiology, Jinnah Hospital Lahore and DHQ Hospital, Vihari from January 2021 to September 2022. Females (15 and 75 years) with breast lesion of diameter (>1cm) were enrolled. About 60 patients underwent breast MRI with contrast and DWI. The remaining 140 patients underwent DWI came for sonomammography. ADC values (b-vale 0-300, 0-600, and 0-1000) for all lesions were computed for differentiating the benign and malignant lesions. Benign and malignant breast lesions were categorized based on ADC values (1.3-1.5 mm2/s) and (0.85-1.1 mm2/s) respectively.

Results: Of the total 260 breast lesions, the incidence of benign and malignant breast lesions based on ADC values were 192 (73.8%) and 68 (22.2%) respectively. About 4 lesions were benign based on ADC value but were proven malignant after intervention. This approach of detecting malignant lesions based on ADC values has a specificity 100%, sensitivity 97.22%, positive prediction value 100%, and negative prediction value 99%. The ADC value of benign versus malignant breast lesions at b-value (0-300), (0-600), and (0-1000) were 1.56±0.23 vs. 0.82±0.21, 1.54±0.09 vs. 0.89±0.17, and 1.52±0.16 vs. 0.84±0.18 respectively.

Conclusion: The present study demonstrated that DWI for breast lesions had a good sensitivity and specificity for distinguishing benign from malignant lesions.

Keywords: Breast lesions, DWI, MRI, benign lesions, malignant lesions

Downloads