Novel Imaging Algorithm Shows Promise in DMD Evaluation

Ultrasound attenuation imaging has potential for the evaluation of ambulatory function progression in patients with Duchenne muscular dystrophy (DMD), a new study found. The findings were published in Medical Physics.¹

While patient survival has improved in recent years with the advent of novel care strategies and therapies for DMD, the evaluation tools currently used, like the 6-minute walk test or North Star Ambulatory Assessment, are mostly applicable to patients who are ambulatory and can be limited based on patients’ mood or effort level.

When there are changes in muscle microstructures, acoustic attenuation occurs, which may hold potential as an imaging biomarker for DMD. | Image credit: hafakot - stock.adobe.com

“Ultrasound is increasingly recognized as a preferable alternative for ambulatory assessment tests, primarily due to its efficacy in muscle imaging, cost-effectiveness, noninvasive nature, absence of ionizing radiation, and suitability for pediatric patients, compared with magnetic resonance imaging and computed tomography,” the authors explained.

When there are changes in muscle microstructures, acoustic attenuation occurs, which may hold potential as an imaging biomarker for DMD.² In their study, the authors aimed to expand upon the reference frequency method (RFM)—which estimates local attenuation without a reference phantom—and develop the robust RFM (RRFM) as a new approach of ultrasound attenuation imaging in patients with DMD.¹

“The RRFM integrates an iterative reweighted least squares (IRLS) technique to reduce the influence of outliers, thereby ensuring more reliable attenuation estimation, especially when utilizing a smaller window for attenuation imaging,” the authors wrote. “The subsequent sections will introduce the theoretical background of RFM and the algorithmic details of RRFM.”

Measurements on a standard phantom and a clinical data set of patients with DMD were used to validate the RRFM approach, and its performance was compared with quantitative ultrasound envelope statistics imaging methods, including Nakagami, homodyned K (HK), and entropy analysis.

A total of 161 patients with DMD were included in the study, with 130 in a validation cohort and 31 in a testing cohort. Patients underwent ultrasound scans of the gastrocnemius to capture RRFM-based attenuation imaging, and the diagnostic performance of RRFM for ambulatory function and differentiation between early and late ambulation stages was compared with quantitative ultrasound envelope statistics imaging.

The RRFM method demonstrated more sensitivity vs RFM, with more accuracy matching the attenuation properties of the phantom. RRFM showed a 50% lower measurement bias compared with traditional RFM.

Regarding differentiation between early and late ambulatory stages, RRFM-based attenuation imaging showed an area under the receiver operating characteristic curve (AUROC) of 0.88 in the validation data set, compared with 0.71, 0.73, and 0.74 with Nakagami, HK, and entropy measures, respectively. The AUROC of RRFM-based attenuation imaging in the testing data set was 0.92, compared with 0.77, 0.77, and 0.81 with Nakagami, HK, and entropy measures, respectively. The AUROCs with RRFM-based imaging indicated significantly better performance compared with QUS envelope statistics imaging (P < .05).

The study was limited by a small data set and data imbalance, but these limitations did not undermine the findings related to RRFM-based attenuation imaging vs QUS envelope statistics, according to the authors.

“Our results indicated that RRFM significantly reduces the measurement bias in attenuation estimation compared to conventional RFM,” the authors wrote. “Furthermore, when compared with QUS envelope statistics imaging, RRFM-based attenuation imaging proved beneficial in assessing DMD, including diagnosing patients’ walking function and distinguishing between early and late ambulatory stages. Ultrasound attenuation imaging utilizing RRFM holds considerable potential for enhancing the health management and follow-up care of DMD patients.”

References

1. Yan D, Li Q, Chuang Y, et al. Ultrasound attenuation imaging as a strategy for evaluation of early and late ambulatory functions in Duchenne muscular dystrophy. Med Phys. Published online September 5, 2024. doi:10.1002/mp.17389

2. Berger G, Laugier P, Fink M, Perrin J. Optimal precision in ultrasound attenuation estimation and application to the detection of Duchenne muscular dystrophy carriers. Ultrason Imaging. 1987;9(1):1-17. doi:10.1177/016173468700900101