Deep learning-based automatic scoring models for the disease activity of rheumatoid arthritis based on multimodal ultrasound images
© The Author(s) 2023. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For permissions, please email: journals.permissionsoup.com..
OBJECTIVES: We aimed to investigate the value of deep learning (DL) models based on multimodal ultrasonographic (US) images to quantify RA activity.
METHODS: Static greyscale (SGS), dynamic greyscale (DGS), static power Doppler (SPD) and dynamic power Doppler (DPD) US images were collected and evaluated by two expert radiologists according to the EULAR-OMERACT Synovitis Scoring system. Four DL models were developed based on the ResNet-type structure, evaluated on two separate test cohorts, and finally compared with the performance of 12 radiologists with different levels of experience.
RESULTS: In total, 1244 images were used for the model training, and 152 and 354 for testing (cohort 1 and 2, respectively). The best-performing models for the scores of 0/1/2/3 were the DPD, SGS, DGS and SPD models, respectively (Area Under the receiver operating characteristic Curve [AUC] = 0.87/0.95/0.74/0.95; no significant differences). All the DL models provided results comparable to the experienced radiologists on a per-image basis (intraclass correlation coefficient: 0.239-0.756, P < 0.05). The SPD model performed better than the SGS one on test cohort 1 (score of 0/2/3: AUC = 0.82/0.67/0.95 vs 0.66/0.66/0.75, respectively) and test cohort 2 (score of 0: AUC = 0.89 vs 0.81). The dynamic DL models performed better than the static ones in most of the scoring processes and were more accurate than the most of senior radiologists, especially the DPD model.
CONCLUSION: DL models based on multimodal US images allow a quantitative and objective assessment of RA activity. Dynamic DL models in particular have potential value in assisting radiologists to improve the accuracy of RA US-based grading.
| Medienart: |
E-Artikel |
|---|
| Erscheinungsjahr: |
2024 |
|---|---|
| Erschienen: |
2024 |
| Enthalten in: |
Zur Gesamtaufnahme - volume:63 |
|---|---|
| Enthalten in: |
Rheumatology (Oxford, England) - 63(2024), 3 vom: 01. März, Seite 866-873 |
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
He, Xuelei [VerfasserIn] |
|---|
| Links: |
|---|
| Themen: |
Activity |
|---|
| Anmerkungen: |
Date Completed 04.03.2024 Date Revised 04.03.2024 published: Print Citation Status MEDLINE |
|---|
| doi: |
10.1093/rheumatology/kead366 |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
NLM359720382 |
|---|
| LEADER | 01000caa a22002652 4500 | ||
|---|---|---|---|
| 001 | NLM359720382 | ||
| 003 | DE-627 | ||
| 005 | 20240304231916.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 231226s2024 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1093/rheumatology/kead366 |2 doi | |
| 028 | 5 | 2 | |a pubmed24n1316.xml |
| 035 | |a (DE-627)NLM359720382 | ||
| 035 | |a (NLM)37471602 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a He, Xuelei |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a Deep learning-based automatic scoring models for the disease activity of rheumatoid arthritis based on multimodal ultrasound images |
| 264 | 1 | |c 2024 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a ƒaComputermedien |b c |2 rdamedia | ||
| 338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Date Completed 04.03.2024 | ||
| 500 | |a Date Revised 04.03.2024 | ||
| 500 | |a published: Print | ||
| 500 | |a Citation Status MEDLINE | ||
| 520 | |a © The Author(s) 2023. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For permissions, please email: journals.permissionsoup.com. | ||
| 520 | |a OBJECTIVES: We aimed to investigate the value of deep learning (DL) models based on multimodal ultrasonographic (US) images to quantify RA activity | ||
| 520 | |a METHODS: Static greyscale (SGS), dynamic greyscale (DGS), static power Doppler (SPD) and dynamic power Doppler (DPD) US images were collected and evaluated by two expert radiologists according to the EULAR-OMERACT Synovitis Scoring system. Four DL models were developed based on the ResNet-type structure, evaluated on two separate test cohorts, and finally compared with the performance of 12 radiologists with different levels of experience | ||
| 520 | |a RESULTS: In total, 1244 images were used for the model training, and 152 and 354 for testing (cohort 1 and 2, respectively). The best-performing models for the scores of 0/1/2/3 were the DPD, SGS, DGS and SPD models, respectively (Area Under the receiver operating characteristic Curve [AUC] = 0.87/0.95/0.74/0.95; no significant differences). All the DL models provided results comparable to the experienced radiologists on a per-image basis (intraclass correlation coefficient: 0.239-0.756, P < 0.05). The SPD model performed better than the SGS one on test cohort 1 (score of 0/2/3: AUC = 0.82/0.67/0.95 vs 0.66/0.66/0.75, respectively) and test cohort 2 (score of 0: AUC = 0.89 vs 0.81). The dynamic DL models performed better than the static ones in most of the scoring processes and were more accurate than the most of senior radiologists, especially the DPD model | ||
| 520 | |a CONCLUSION: DL models based on multimodal US images allow a quantitative and objective assessment of RA activity. Dynamic DL models in particular have potential value in assisting radiologists to improve the accuracy of RA US-based grading | ||
| 650 | 4 | |a Journal Article | |
| 650 | 4 | |a RA | |
| 650 | 4 | |a activity | |
| 650 | 4 | |a deep learning | |
| 650 | 4 | |a multimodal ultrasonography | |
| 650 | 4 | |a scoring | |
| 700 | 1 | |a Wang, Ming |e verfasserin |4 aut | |
| 700 | 1 | |a Zhao, Chenyang |e verfasserin |4 aut | |
| 700 | 1 | |a Wang, Qian |e verfasserin |4 aut | |
| 700 | 1 | |a Zhang, Rui |e verfasserin |4 aut | |
| 700 | 1 | |a Liu, Jian |e verfasserin |4 aut | |
| 700 | 1 | |a Zhang, Yixiu |e verfasserin |4 aut | |
| 700 | 1 | |a Qi, Zhenhong |e verfasserin |4 aut | |
| 700 | 1 | |a Su, Na |e verfasserin |4 aut | |
| 700 | 1 | |a Wei, Yao |e verfasserin |4 aut | |
| 700 | 1 | |a Gui, Yang |e verfasserin |4 aut | |
| 700 | 1 | |a Li, Jianchu |e verfasserin |4 aut | |
| 700 | 1 | |a Tian, Xinping |e verfasserin |4 aut | |
| 700 | 1 | |a Zeng, Xiaofeng |e verfasserin |4 aut | |
| 700 | 1 | |a Jiang, Yuxin |e verfasserin |4 aut | |
| 700 | 1 | |a Wang, Kun |e verfasserin |4 aut | |
| 700 | 1 | |a Yang, Meng |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Rheumatology (Oxford, England) |d 1999 |g 63(2024), 3 vom: 01. März, Seite 866-873 |w (DE-627)NLM102581908 |x 1462-0332 |7 nnns |
| 773 | 1 | 8 | |g volume:63 |g year:2024 |g number:3 |g day:01 |g month:03 |g pages:866-873 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.1093/rheumatology/kead366 |3 Volltext |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a GBV_NLM | ||
| 951 | |a AR | ||
| 952 | |d 63 |j 2024 |e 3 |b 01 |c 03 |h 866-873 | ||