Prediction of total knee replacement using deep learning analysis of knee MRI
Abstract Current methods for assessing knee osteoarthritis (OA) do not provide comprehensive information to make robust and accurate outcome predictions. Deep learning (DL) risk assessment models were developed to predict the progression of knee OA to total knee replacement (TKR) over a 108-month fo...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2023-04-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-023-33934-1 |
| _version_ | 1797836533451980800 |
|---|---|
| author | Haresh Rengaraj Rajamohan Tianyu Wang Kevin Leung Gregory Chang Kyunghyun Cho Richard Kijowski Cem M. Deniz |
| author_facet | Haresh Rengaraj Rajamohan Tianyu Wang Kevin Leung Gregory Chang Kyunghyun Cho Richard Kijowski Cem M. Deniz |
| author_sort | Haresh Rengaraj Rajamohan |
| collection | DOAJ |
| description | Abstract Current methods for assessing knee osteoarthritis (OA) do not provide comprehensive information to make robust and accurate outcome predictions. Deep learning (DL) risk assessment models were developed to predict the progression of knee OA to total knee replacement (TKR) over a 108-month follow-up period using baseline knee MRI. Participants of our retrospective study consisted of 353 case–control pairs of subjects from the Osteoarthritis Initiative with and without TKR over a 108-month follow-up period matched according to age, sex, ethnicity, and body mass index. A traditional risk assessment model was created to predict TKR using baseline clinical risk factors. DL models were created to predict TKR using baseline knee radiographs and MRI. All DL models had significantly higher (p < 0.001) AUCs than the traditional model. The MRI and radiograph ensemble model and MRI ensemble model (where TKR risk predicted by several contrast-specific DL models were averaged to get the ensemble TKR risk prediction) had the highest AUCs of 0.90 (80% sensitivity and 85% specificity) and 0.89 (79% sensitivity and 86% specificity), respectively, which were significantly higher (p < 0.05) than the AUCs of the radiograph and multiple MRI models (where the DL models were trained to predict TKR risk using single contrast or 2 contrasts together as input). DL models using baseline MRI had a higher diagnostic performance for predicting TKR than a traditional model using baseline clinical risk factors and a DL model using baseline knee radiographs. |
| first_indexed | 2024-04-09T15:10:25Z |
| format | Article |
| id | doaj.art-2be54c251e244660a047431667f0df68 |
| institution | Directory Open Access Journal |
| issn | 2045-2322 |
| language | English |
| last_indexed | 2024-04-09T15:10:25Z |
| publishDate | 2023-04-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj.art-2be54c251e244660a047431667f0df682023-04-30T11:12:36ZengNature PortfolioScientific Reports2045-23222023-04-0113111110.1038/s41598-023-33934-1Prediction of total knee replacement using deep learning analysis of knee MRIHaresh Rengaraj Rajamohan0Tianyu Wang1Kevin Leung2Gregory Chang3Kyunghyun Cho4Richard Kijowski5Cem M. Deniz6Center for Data Science, New York UniversityCenter for Data Science, New York UniversityCourant Institute of Mathematical Sciences, New York UniversityDepartment of Radiology, New York University Langone HealthCenter for Data Science, New York UniversityDepartment of Radiology, New York University Langone HealthDepartment of Radiology, New York University Langone HealthAbstract Current methods for assessing knee osteoarthritis (OA) do not provide comprehensive information to make robust and accurate outcome predictions. Deep learning (DL) risk assessment models were developed to predict the progression of knee OA to total knee replacement (TKR) over a 108-month follow-up period using baseline knee MRI. Participants of our retrospective study consisted of 353 case–control pairs of subjects from the Osteoarthritis Initiative with and without TKR over a 108-month follow-up period matched according to age, sex, ethnicity, and body mass index. A traditional risk assessment model was created to predict TKR using baseline clinical risk factors. DL models were created to predict TKR using baseline knee radiographs and MRI. All DL models had significantly higher (p < 0.001) AUCs than the traditional model. The MRI and radiograph ensemble model and MRI ensemble model (where TKR risk predicted by several contrast-specific DL models were averaged to get the ensemble TKR risk prediction) had the highest AUCs of 0.90 (80% sensitivity and 85% specificity) and 0.89 (79% sensitivity and 86% specificity), respectively, which were significantly higher (p < 0.05) than the AUCs of the radiograph and multiple MRI models (where the DL models were trained to predict TKR risk using single contrast or 2 contrasts together as input). DL models using baseline MRI had a higher diagnostic performance for predicting TKR than a traditional model using baseline clinical risk factors and a DL model using baseline knee radiographs.https://doi.org/10.1038/s41598-023-33934-1 |
| spellingShingle | Haresh Rengaraj Rajamohan Tianyu Wang Kevin Leung Gregory Chang Kyunghyun Cho Richard Kijowski Cem M. Deniz Prediction of total knee replacement using deep learning analysis of knee MRI Scientific Reports |
| title | Prediction of total knee replacement using deep learning analysis of knee MRI |
| title_full | Prediction of total knee replacement using deep learning analysis of knee MRI |
| title_fullStr | Prediction of total knee replacement using deep learning analysis of knee MRI |
| title_full_unstemmed | Prediction of total knee replacement using deep learning analysis of knee MRI |
| title_short | Prediction of total knee replacement using deep learning analysis of knee MRI |
| title_sort | prediction of total knee replacement using deep learning analysis of knee mri |
| url | https://doi.org/10.1038/s41598-023-33934-1 |
| work_keys_str_mv | AT hareshrengarajrajamohan predictionoftotalkneereplacementusingdeeplearninganalysisofkneemri AT tianyuwang predictionoftotalkneereplacementusingdeeplearninganalysisofkneemri AT kevinleung predictionoftotalkneereplacementusingdeeplearninganalysisofkneemri AT gregorychang predictionoftotalkneereplacementusingdeeplearninganalysisofkneemri AT kyunghyuncho predictionoftotalkneereplacementusingdeeplearninganalysisofkneemri AT richardkijowski predictionoftotalkneereplacementusingdeeplearninganalysisofkneemri AT cemmdeniz predictionoftotalkneereplacementusingdeeplearninganalysisofkneemri |