Deep Learning Image Processing Enables 40% Faster Spinal MR Scans Which Match or Exceed Quality of Standard of Care

A Shankaranarayanan,W Gibbs,B Johnson,T Zhang,L N Tanenbaum,S Bash

doi:10.1007/s00062-021-01121-2

A Shankaranarayanan, W Gibbs + Show 4 more

Open Access

https://doi.org/10.1007/s00062-021-01121-2

Copy DOI

Abstract

ObjectiveThis prospective multicenter multireader study evaluated the performance of 40% scan-time reduced spinal magnetic resonance imaging (MRI) reconstructed with deep learning (DL).MethodsA total of 61 patients underwent standard of care (SOC) and accelerated (FAST) spine MRI. DL was used to enhance the accelerated set (FAST-DL). Three neuroradiologists were presented with paired side-by-side datasets (666 series). Datasets were blinded and randomized in sequence and left-right display order. Image features were preference rated. Structural similarity index (SSIM) and per pixel L1 was assessed for the image sets pre and post DL-enhancement as a quantitative assessment of image integrity impact.ResultsFAST-DL was qualitatively better than SOC for perceived signal-to-noise ratio (SNR) and artifacts and equivalent for other features. Quantitative SSIM was high, supporting the absence of image corruption by DL processing.ConclusionDL enables 40% spine MRI scan time reduction while maintaining diagnostic integrity and image quality with perceived benefits in SNR and artifact reduction, suggesting potential for clinical practice utility.

Highlights

A clinical practice, standard of care (SOC) study was performed consisting of multiple routine pulse sequences: sagittal T2 (n = 23)/T1 (n = 21)/STIR (n = 18)/PD (n = 12); and axial T2 (n = 20)/T1 (n = 17) for a total of 111 sequences acquired from 61 patients
FAST-deep learning (DL) was statistically better than SOC for perceived signal-to-noise ratio (SNR) (3.4 ± 0.6, p-value
The Wilcox paired test differences were found to be significant (p < 0.001) at 4.363 e–12. This prospective, randomized, multicenter study assessed the ability of DL enhancement to preserve perceived MR spine image quality despite 40% scan time reduction

Summary

Participants

A total of 61 consecutive patients (45.5 ± 17.1 years old) were prospectively recruited and consented for this multicenter, multireader, randomized case-control Institutional Review Board (IRB) approved study. A clinical practice, SOC study was performed consisting of multiple routine pulse sequences: sagittal T2 (n = 23)/T1 (n = 21)/STIR (n = 18)/PD (n = 12); and axial T2 (n = 20)/T1 (n = 17) (average sequence scan time: 171.2 ± 66.4 s) for a total of 111 sequences acquired from 61 patients. The DL model was trained on 1000s of MR DICOM datasets from multiple vendors and clinical sites with a variety of clinical indications and field strengths, experiencing a range of image quality, tissue contrasts, acquisition parameters, and patient anatomies. Wilcoxon rank sum tests were performed to assess the statistical significance of the difference in scores for each feature in comparative datasets (Table 2). While not part of the subjective analysis, SOC images were processed with DL and subjected to SSIM measures (SOC vs. SOC-DL) as an additional method of assessing the impact of DL processing (Table 4)

Results

Discussion

Conclusion