Assessment of idiopathic inflammatory myopathy using a deep learning method for muscle T2 mapping segmentation

Abstract

Objective: To investigate the utility of an automatic deep learning (DL) method for segmentation of T2 maps in patients with idiopathic inflammatory myopathy (IIM) against healthy controls, and also the association of quantitative T2 values in patients with laboratory and pulmonary findings.

Methods: Structural MRI and T2 mapping of bilateral thigh muscles from patients with IIM and healthy volunteers were segmented using dedicated software based on a pre-trained convolutional neural network. Incremental and federated learning were implemented for continuous adaptation and improvement. Muscle T2 values derived from DL segmentation were compared between patients and healthy controls, and T2 values of patients were further analyzed with serum muscle enzymes, and interstitial lung disease (ILD) which was diagnosed and graded based on chest HRCT.

Results: Overall, 64 patients (27 patients with dermatomyositis, 29 with polymyositis, and 8 with antisynthetase syndrome (ASS)) and 10 healthy controls were included. By using DL-based muscle segmentation, T2 values generated from T2 maps accurately differentiated patients from those of controls (p < 0.001) with a cutoff value of 36.4 ms (sensitivity 96.9%, and specificity 100%). In patients with IIM, muscle T2 values positively correlated with all the serum muscle enzymes (all p < 0.05). ILD score of patients with ASS was markedly higher than that of those without ASS (p = 0.011), while dissociation between the severity of muscular involvement and ILD was observed (p = 0.080).

Conclusion: Automatic DL could be used to segment thigh muscles and help quantitatively assess muscular inflammation of IIM through T2 mapping.

Key points: • Muscle T2 mapping automatically segmented by deep learning can differentiate IIM from healthy controls. • T2 value, an indicator of active muscle inflammation, positively correlates with serum muscle enzymes. • T2 mapping can detect muscle disease in patients with normal muscle enzyme levels.

Keywords: Deep learning; Idiopathic inflammatory myopathy; Magnetic resonance imaging; Myositis; Thigh muscles.

Fig. 1

Demonstration of muscle T2 measurement segmented by deep learning (DL). The T2WI in-phase images were used as the structural images for auto-segmentation, with segmentation results marked in purple. Minor manual revision was then performed to erase unnecessary regions, and add missed muscle areas which were marked in red. This revised segmentation was saved as masks which were imported and matched to the corresponding slices of T2 map images by the alignment module. Subsequently, the final region of interest (marked in dark red) was achieved after manual revision if you want and muscle T2 value of the bilateral thigh muscles was provided

Fig. 2

The ROC curve (red line) for the diagnostic performance of T2 mapping. The ROC curve (red line) for the diagnostic performance of T2 mapping in the differentiation of patients with idiopathic inflammatory myopathy from healthy controls. The area under the curve (AUC) was 0.986

Fig. 3

MRI of a 52-year-old female with dermatomyositis. (A) Axial T1-weighted image (T1WI), (B) T2WI, (C) fat-saturated (FS) T2WI, and (D) color-coded T2 map. The serum muscle enzymes of this patient were normal, while slight hyperintensity of quadriceps and adductor magnus (arrows) was noticed on FS T2WI which was further confirmed by an elevation of T2 value (44.9 ms) presenting as light blue in color-coded T2 map