Musculoskeletal radiologist-level performance by using deep learning for detection of scaphoid fractures on conventional multi-view radiographs of hand and wrist
© 2022. The Author(s)..
OBJECTIVES: To assess how an artificial intelligence (AI) algorithm performs against five experienced musculoskeletal radiologists in diagnosing scaphoid fractures and whether it aids their diagnosis on conventional multi-view radiographs.
METHODS: Four datasets of conventional hand, wrist, and scaphoid radiographs were retrospectively acquired at two hospitals (hospitals A and B). Dataset 1 (12,990 radiographs from 3353 patients, hospital A) and dataset 2 (1117 radiographs from 394 patients, hospital B) were used for training and testing a scaphoid localization and laterality classification component. Dataset 3 (4316 radiographs from 840 patients, hospital A) and dataset 4 (688 radiographs from 209 patients, hospital B) were used for training and testing the fracture detector. The algorithm was compared with the radiologists in an observer study. Evaluation metrics included sensitivity, specificity, positive predictive value (PPV), area under the characteristic operating curve (AUC), Cohen's kappa coefficient (κ), fracture localization precision, and reading time.
RESULTS: The algorithm detected scaphoid fractures with a sensitivity of 72%, specificity of 93%, PPV of 81%, and AUC of 0.88. The AUC of the algorithm did not differ from each radiologist (0.87 [radiologists' mean], p ≥ .05). AI assistance improved five out of ten pairs of inter-observer Cohen's κ agreements (p < .05) and reduced reading time in four radiologists (p < .001), but did not improve other metrics in the majority of radiologists (p ≥ .05).
CONCLUSIONS: The AI algorithm detects scaphoid fractures on conventional multi-view radiographs at the level of five experienced musculoskeletal radiologists and could significantly shorten their reading time.
KEY POINTS: • An artificial intelligence algorithm automatically detects scaphoid fractures on conventional multi-view radiographs at the same level of five experienced musculoskeletal radiologists. • There is preliminary evidence that automated scaphoid fracture detection can significantly shorten the reading time of musculoskeletal radiologists.
| Medienart: |
E-Artikel |
|---|
| Erscheinungsjahr: |
2023 |
|---|---|
| Erschienen: |
2023 |
| Enthalten in: |
Zur Gesamtaufnahme - volume:33 |
|---|---|
| Enthalten in: |
European radiology - 33(2023), 3 vom: 15. März, Seite 1575-1588 |
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Hendrix, Nils [VerfasserIn] |
|---|
| Links: |
|---|
| Themen: |
Artificial intelligence |
|---|
| Anmerkungen: |
Date Completed 22.02.2023 Date Revised 30.06.2023 published: Print-Electronic Citation Status MEDLINE |
|---|
| doi: |
10.1007/s00330-022-09205-4 |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
NLM348936036 |
|---|
| LEADER | 01000naa a22002652 4500 | ||
|---|---|---|---|
| 001 | NLM348936036 | ||
| 003 | DE-627 | ||
| 005 | 20231226041449.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 231226s2023 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1007/s00330-022-09205-4 |2 doi | |
| 028 | 5 | 2 | |a pubmed24n1163.xml |
| 035 | |a (DE-627)NLM348936036 | ||
| 035 | |a (NLM)36380195 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Hendrix, Nils |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a Musculoskeletal radiologist-level performance by using deep learning for detection of scaphoid fractures on conventional multi-view radiographs of hand and wrist |
| 264 | 1 | |c 2023 | |
| 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 22.02.2023 | ||
| 500 | |a Date Revised 30.06.2023 | ||
| 500 | |a published: Print-Electronic | ||
| 500 | |a Citation Status MEDLINE | ||
| 520 | |a © 2022. The Author(s). | ||
| 520 | |a OBJECTIVES: To assess how an artificial intelligence (AI) algorithm performs against five experienced musculoskeletal radiologists in diagnosing scaphoid fractures and whether it aids their diagnosis on conventional multi-view radiographs | ||
| 520 | |a METHODS: Four datasets of conventional hand, wrist, and scaphoid radiographs were retrospectively acquired at two hospitals (hospitals A and B). Dataset 1 (12,990 radiographs from 3353 patients, hospital A) and dataset 2 (1117 radiographs from 394 patients, hospital B) were used for training and testing a scaphoid localization and laterality classification component. Dataset 3 (4316 radiographs from 840 patients, hospital A) and dataset 4 (688 radiographs from 209 patients, hospital B) were used for training and testing the fracture detector. The algorithm was compared with the radiologists in an observer study. Evaluation metrics included sensitivity, specificity, positive predictive value (PPV), area under the characteristic operating curve (AUC), Cohen's kappa coefficient (κ), fracture localization precision, and reading time | ||
| 520 | |a RESULTS: The algorithm detected scaphoid fractures with a sensitivity of 72%, specificity of 93%, PPV of 81%, and AUC of 0.88. The AUC of the algorithm did not differ from each radiologist (0.87 [radiologists' mean], p ≥ .05). AI assistance improved five out of ten pairs of inter-observer Cohen's κ agreements (p < .05) and reduced reading time in four radiologists (p < .001), but did not improve other metrics in the majority of radiologists (p ≥ .05) | ||
| 520 | |a CONCLUSIONS: The AI algorithm detects scaphoid fractures on conventional multi-view radiographs at the level of five experienced musculoskeletal radiologists and could significantly shorten their reading time | ||
| 520 | |a KEY POINTS: • An artificial intelligence algorithm automatically detects scaphoid fractures on conventional multi-view radiographs at the same level of five experienced musculoskeletal radiologists. • There is preliminary evidence that automated scaphoid fracture detection can significantly shorten the reading time of musculoskeletal radiologists | ||
| 650 | 4 | |a Journal Article | |
| 650 | 4 | |a Artificial intelligence | |
| 650 | 4 | |a Clinical decision support system | |
| 650 | 4 | |a Fractures, bone | |
| 650 | 4 | |a Multicenter study | |
| 650 | 4 | |a Scaphoid bone | |
| 700 | 1 | |a Hendrix, Ward |e verfasserin |4 aut | |
| 700 | 1 | |a van Dijke, Kees |e verfasserin |4 aut | |
| 700 | 1 | |a Maresch, Bas |e verfasserin |4 aut | |
| 700 | 1 | |a Maas, Mario |e verfasserin |4 aut | |
| 700 | 1 | |a Bollen, Stijn |e verfasserin |4 aut | |
| 700 | 1 | |a Scholtens, Alexander |e verfasserin |4 aut | |
| 700 | 1 | |a de Jonge, Milko |e verfasserin |4 aut | |
| 700 | 1 | |a Ong, Lee-Ling Sharon |e verfasserin |4 aut | |
| 700 | 1 | |a van Ginneken, Bram |e verfasserin |4 aut | |
| 700 | 1 | |a Rutten, Matthieu |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t European radiology |d 1991 |g 33(2023), 3 vom: 15. März, Seite 1575-1588 |w (DE-627)NLM087691310 |x 1432-1084 |7 nnns |
| 773 | 1 | 8 | |g volume:33 |g year:2023 |g number:3 |g day:15 |g month:03 |g pages:1575-1588 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.1007/s00330-022-09205-4 |3 Volltext |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a GBV_NLM | ||
| 951 | |a AR | ||
| 952 | |d 33 |j 2023 |e 3 |b 15 |c 03 |h 1575-1588 | ||