TY - JOUR
T1 - Deep-Learning Generated Synthetic Double Inversion Recovery Images Improve Multiple Sclerosis Lesion Detection
AU - Finck, Tom
AU - Li, Hongwei
AU - Grundl, Lioba
AU - Eichinger, Paul
AU - Bussas, Matthias
AU - Mühlau, Mark
AU - Menze, Bjoern
AU - Wiestler, Benedikt
N1 - Publisher Copyright:
© Wolters Kluwer Health, Inc. All rights reserved.
PY - 2020/5/1
Y1 - 2020/5/1
N2 - The aim of the study was to implement a deep-learning tool to produce synthetic double inversion recovery (synthDIR) images and compare their diagnostic performance to conventional sequences in patients with multiple sclerosis (MS). Materials and Methods For this retrospective analysis, 100 MS patients (65 female, 37 [22-68] years) were randomly selected from a prospective observational cohort between 2014 and 2016. In a subset of 50 patients, an artificial neural network (DiamondGAN) was trained to generate a synthetic DIR (synthDIR) from standard acquisitions (T1, T2, and fluid-attenuated inversion recovery [FLAIR]). With the resulting network, synthDIR was generated for the remaining 50 subjects. These images as well as conventionally acquired DIR (trueDIR) and FLAIR images were assessed for MS lesions by 2 independent readers, blinded to the source of the DIR image. Lesion counts in the different modalities were compared using a Wilcoxon signed-rank test, and interrater analysis was performed. Contrast-to-noise ratios were compared for objective image quality. Results Utilization of synthDIR allowed to detect significantly more lesions compared with the use of FLAIR images (31.4 ± 20.7 vs 22.8 ± 12.7, P < 0.001). This improvement was mainly attributable to an improved depiction of juxtacortical lesions (12.3 ± 10.8 vs 7.2 ± 5.6, P < 0.001). Interrater reliability was excellent in FLAIR 0.92 (95% confidence interval [CI], 0.85-0.95), synthDIR 0.93 (95% CI, 0.87-0.96), and trueDIR 0.95 (95% CI, 0.85-0.98). Contrast-to-noise ratio in synthDIR exceeded that of FLAIR (22.0 ± 6.4 vs 16.7 ± 3.6, P = 0.009); no significant difference was seen in comparison to trueDIR (22.0 ± 6.4 vs 22.4 ± 7.9, P = 0.87). Conclusions Computationally generated DIR images improve lesion depiction compared with the use of standard modalities. This method demonstrates how artificial intelligence can help improving imaging in specific pathologies.
AB - The aim of the study was to implement a deep-learning tool to produce synthetic double inversion recovery (synthDIR) images and compare their diagnostic performance to conventional sequences in patients with multiple sclerosis (MS). Materials and Methods For this retrospective analysis, 100 MS patients (65 female, 37 [22-68] years) were randomly selected from a prospective observational cohort between 2014 and 2016. In a subset of 50 patients, an artificial neural network (DiamondGAN) was trained to generate a synthetic DIR (synthDIR) from standard acquisitions (T1, T2, and fluid-attenuated inversion recovery [FLAIR]). With the resulting network, synthDIR was generated for the remaining 50 subjects. These images as well as conventionally acquired DIR (trueDIR) and FLAIR images were assessed for MS lesions by 2 independent readers, blinded to the source of the DIR image. Lesion counts in the different modalities were compared using a Wilcoxon signed-rank test, and interrater analysis was performed. Contrast-to-noise ratios were compared for objective image quality. Results Utilization of synthDIR allowed to detect significantly more lesions compared with the use of FLAIR images (31.4 ± 20.7 vs 22.8 ± 12.7, P < 0.001). This improvement was mainly attributable to an improved depiction of juxtacortical lesions (12.3 ± 10.8 vs 7.2 ± 5.6, P < 0.001). Interrater reliability was excellent in FLAIR 0.92 (95% confidence interval [CI], 0.85-0.95), synthDIR 0.93 (95% CI, 0.87-0.96), and trueDIR 0.95 (95% CI, 0.85-0.98). Contrast-to-noise ratio in synthDIR exceeded that of FLAIR (22.0 ± 6.4 vs 16.7 ± 3.6, P = 0.009); no significant difference was seen in comparison to trueDIR (22.0 ± 6.4 vs 22.4 ± 7.9, P = 0.87). Conclusions Computationally generated DIR images improve lesion depiction compared with the use of standard modalities. This method demonstrates how artificial intelligence can help improving imaging in specific pathologies.
KW - artificial intelligence
KW - deep learning
KW - double inversion recovery
KW - generative adversarial networks
KW - multiple sclerosis
UR - http://www.scopus.com/inward/record.url?scp=85083003262&partnerID=8YFLogxK
U2 - 10.1097/RLI.0000000000000640
DO - 10.1097/RLI.0000000000000640
M3 - Article
C2 - 31977602
AN - SCOPUS:85083003262
SN - 0020-9996
VL - 55
SP - 318
EP - 323
JO - Investigative Radiology
JF - Investigative Radiology
IS - 5
ER -