Effect of radiographic disease severity in high-resolution quantitative computed tomography assessment of metacarpophalangeal joint erosion and cysts

Diagnostic accuracy of high-resolution peripheral quantitative computed tomography and X-ray for classifying erosive rheumatoid arthritis

OBJECTIVES

To investigate whether High-Resolution peripheral Quantitative Computed Tomography (HR-pQCT) of two metacarpophalangeal (MCP) joints can more accurately classify patients as having erosive rheumatoid arthritis (RA) compared with conventional radiography (CR) of 44 joints in the hands, wrists, and feet.

METHODS

In this single-centre cross-sectional study, patients with established RA (disease duration ≥ 5 years) were investigated by HR-pQCT and CR. The second and third MCP joints of the dominant hand were assessed for erosions by HR-pQCT. CR of the hands, wrists, and feet were scored according to the Sharp/van der Heijde (SHS) method.

RESULTS

In total, 353 patients were included, 66 (18.7%) patients were classified as having non-erosive RA, and 287 (81.3%) had erosive RA by CR. The sensitivity and specificity (95%CI) of HR-pQCT for classifying patients as having erosive RA when standard CR of hands, wrists, and feet were used as the reference was 89% (84-92)% and 30% (20-43)%, respectively. Using HR-pQCT as the reference, the sensitivity and specificity of CR for classifying patients having erosive RA were 85% (80-89)% and 38% (25-52)%, respectively. McNemar's χ2 test showed no significant difference between the sensitivities of patients classified as having erosive RA by HR-pQCT or by CR (2.14, p= 0.177).

CONCLUSIONS

The diagnostic accuracy of HR-pQCT scanning of only two MCP joints and conventional radiography of 44 joints suggest the two modalities were comparable for classifying patients with established RA as having erosive disease.

Deep learning methods allow fully automated segmentation of metacarpal bones to quantify volumetric bone mineral density

Arthritis patients develop hand bone loss, which leads to destruction and functional impairment of the affected joints. High resolution peripheral quantitative computed tomography (HR-pQCT) allows the quantification of volumetric bone mineral density (vBMD) and bone microstructure in vivo with an isotropic voxel size of 82 micrometres. However, image-processing to obtain bone characteristics is a time-consuming process as it requires semi-automatic segmentation of the bone. In this work, a fully automatic vBMD measurement pipeline for the metacarpal (MC) bone using deep learning methods is introduced. Based on a dataset of HR-pQCT volumes with MC measurements for 541 patients with arthritis, a segmentation network is trained. The best network achieves an intersection over union as high as 0.94 and a Dice similarity coefficient of 0.97 while taking only 33 s to process a whole patient yielding a speedup between 2.5 and 4.0 for the whole workflow. Strong correlation between the vBMD measurements of the expert and of the automatic pipeline are achieved for the average bone density with 0.999 (Pearson) and 0.996 (Spearman's rank) with [Formula: see text] for all correlations. A qualitative assessment of the network predictions and the manual annotations yields a 65.9% probability that the expert favors the network predictions. Further, the steps to integrate the pipeline into the clinical workflow are shown. In order to make these workflow improvements available to others, we openly share the code of this work.