Deep learning discrimination of rheumatoid arthritis from osteoarthritis on hand radiography
© 2023. The Author(s), under exclusive licence to International Skeletal Society (ISS)..
PURPOSE: To develop a deep learning model to distinguish rheumatoid arthritis (RA) from osteoarthritis (OA) using hand radiographs and to evaluate the effects of changing pretraining and training parameters on model performance.
MATERIALS AND METHODS: A convolutional neural network was retrospectively trained on 9714 hand radiograph exams from 8387 patients obtained from 2017 to 2021 at seven hospitals within an integrated healthcare network. Performance was assessed using an independent test set of 250 exams from 146 patients. Binary discriminatory capacity (no arthritis versus arthritis; RA versus not RA) and three-way classification (no arthritis versus OA versus RA) were evaluated. The effects of additional pretraining using musculoskeletal radiographs, using all views as opposed to only the posteroanterior view, and varying image resolution on model performance were also investigated. Area under the receiver operating characteristic curve (AUC) and Cohen's kappa coefficient were used to evaluate diagnostic performance.
RESULTS: For no arthritis versus arthritis, the model achieved an AUC of 0.975 (95% CI: 0.957, 0.989). For RA versus not RA, the model achieved an AUC of 0.955 (95% CI: 0.919, 0.983). For three-way classification, the model achieved a kappa of 0.806 (95% CI: 0.742, 0.866) and accuracy of 87.2% (95% CI: 83.2%, 91.2%) on the test set. Increasing image resolution increased performance up to 1024 × 1024 pixels. Additional pretraining on musculoskeletal radiographs and using all views did not significantly affect performance.
CONCLUSION: A deep learning model can be used to distinguish no arthritis, OA, and RA on hand radiographs with high performance.
| Medienart: |
E-Artikel |
|---|
| Erscheinungsjahr: |
2024 2023 |
|---|---|
| Erschienen: |
2024 |
| Enthalten in: |
Zur Gesamtaufnahme - volume:53 |
|---|---|
| Enthalten in: |
Skeletal radiology - 53(2023), 2 vom: 01. Feb., Seite 377-383 |
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Ma, Yuntong [VerfasserIn] |
|---|
| Links: |
|---|
| Themen: |
Deep learning |
|---|
| Anmerkungen: |
Date Completed 21.12.2023 Date Revised 21.12.2023 published: Print-Electronic Citation Status MEDLINE |
|---|
| doi: |
10.1007/s00256-023-04408-2 |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
NLM360306195 |
|---|
| LEADER | 01000caa a22002652 4500 | ||
|---|---|---|---|
| 001 | NLM360306195 | ||
| 003 | DE-627 | ||
| 005 | 20231227135240.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 231226s2024 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1007/s00256-023-04408-2 |2 doi | |
| 028 | 5 | 2 | |a pubmed24n1234.xml |
| 035 | |a (DE-627)NLM360306195 | ||
| 035 | |a (NLM)37530866 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Ma, Yuntong |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a Deep learning discrimination of rheumatoid arthritis from osteoarthritis on hand radiography |
| 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 21.12.2023 | ||
| 500 | |a Date Revised 21.12.2023 | ||
| 500 | |a published: Print-Electronic | ||
| 500 | |a Citation Status MEDLINE | ||
| 520 | |a © 2023. The Author(s), under exclusive licence to International Skeletal Society (ISS). | ||
| 520 | |a PURPOSE: To develop a deep learning model to distinguish rheumatoid arthritis (RA) from osteoarthritis (OA) using hand radiographs and to evaluate the effects of changing pretraining and training parameters on model performance | ||
| 520 | |a MATERIALS AND METHODS: A convolutional neural network was retrospectively trained on 9714 hand radiograph exams from 8387 patients obtained from 2017 to 2021 at seven hospitals within an integrated healthcare network. Performance was assessed using an independent test set of 250 exams from 146 patients. Binary discriminatory capacity (no arthritis versus arthritis; RA versus not RA) and three-way classification (no arthritis versus OA versus RA) were evaluated. The effects of additional pretraining using musculoskeletal radiographs, using all views as opposed to only the posteroanterior view, and varying image resolution on model performance were also investigated. Area under the receiver operating characteristic curve (AUC) and Cohen's kappa coefficient were used to evaluate diagnostic performance | ||
| 520 | |a RESULTS: For no arthritis versus arthritis, the model achieved an AUC of 0.975 (95% CI: 0.957, 0.989). For RA versus not RA, the model achieved an AUC of 0.955 (95% CI: 0.919, 0.983). For three-way classification, the model achieved a kappa of 0.806 (95% CI: 0.742, 0.866) and accuracy of 87.2% (95% CI: 83.2%, 91.2%) on the test set. Increasing image resolution increased performance up to 1024 × 1024 pixels. Additional pretraining on musculoskeletal radiographs and using all views did not significantly affect performance | ||
| 520 | |a CONCLUSION: A deep learning model can be used to distinguish no arthritis, OA, and RA on hand radiographs with high performance | ||
| 650 | 4 | |a Journal Article | |
| 650 | 4 | |a Deep learning | |
| 650 | 4 | |a Hand radiography | |
| 650 | 4 | |a Model parameters | |
| 650 | 4 | |a Osteoarthritis | |
| 650 | 4 | |a Rheumatoid arthritis | |
| 700 | 1 | |a Pan, Ian |e verfasserin |4 aut | |
| 700 | 1 | |a Kim, Stanley Y |e verfasserin |4 aut | |
| 700 | 1 | |a Wieschhoff, Ged G |e verfasserin |4 aut | |
| 700 | 1 | |a Andriole, Katherine P |e verfasserin |4 aut | |
| 700 | 1 | |a Mandell, Jacob C |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Skeletal radiology |d 1993 |g 53(2023), 2 vom: 01. Feb., Seite 377-383 |w (DE-627)NLM001182935 |x 1432-2161 |7 nnns |
| 773 | 1 | 8 | |g volume:53 |g year:2023 |g number:2 |g day:01 |g month:02 |g pages:377-383 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.1007/s00256-023-04408-2 |3 Volltext |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a GBV_NLM | ||
| 951 | |a AR | ||
| 952 | |d 53 |j 2023 |e 2 |b 01 |c 02 |h 377-383 | ||