Bochen Guan1,2, Fang Liu3, Arya Haj Mizaian4, Shadpour Demehri4, Alexey Samsonov5, Ali Guermazi6, Richard Kijowski7. 1. Department of Radiology, University of Wisconsin, 1111 Highland Avenue, Madison, WI, 53705-2275, USA. bochen.guan@gmail.com. 2. Department of Electrical and Computer Engineering, University of Wisconsin, Madison, WI, USA. bochen.guan@gmail.com. 3. Department of Radiology, Massachusetts General Hospital, Harvard University, Boston, MA, USA. 4. Department of Radiology, Johns Hopkins University, Baltimore, MD, USA. 5. Department of Radiology, University of Wisconsin, 1111 Highland Avenue, Madison, WI, 53705-2275, USA. 6. Department of Radiology, Boston University, Boston, MA, USA. 7. Department of Radiology, New York University, New York, NY, USA.
Abstract
OBJECTIVE: To develop and evaluate deep learning (DL) risk assessment models for predicting pain progression in subjects with or at risk of knee osteoarthritis (OA). MATERIALS AND METHODS: The incidence and progression cohorts of the Osteoarthritis Initiative, a multi-center longitudinal study involving 9348 knees in 4674 subjects with or at risk of knee OA that began in 2004 and is ongoing, were used to conduct this retrospective analysis. A subset of knees without and with pain progression (defined as a 9-point or greater increase in pain score between baseline and two or more follow-up time points over the first 48 months) was randomly stratified into training (4200 knees with a mean age of 61.0 years and 60% female) and hold-out testing (500 knees with a mean age of 60.8 years and 60% female) datasets. A DL model was developed to predict pain progression using baseline knee radiographs. An artificial neural network was used to develop a traditional risk assessment model to predict pain progression using demographic, clinical, and radiographic risk factors. A combined model was developed to combine demographic, clinical, and radiographic risk factors with DL analysis of baseline knee radiographs. Area under the curve (AUC) analysis was performed using the hold-out testing dataset to evaluate model performance. RESULTS: The traditional model had an AUC of 0.692 (66.9% sensitivity and 64.1% specificity). The DL model had an AUC of 0.770 (76.7% sensitivity and 70.5% specificity), which was significantly higher (p < 0.001) than the traditional model. The combined model had an AUC of 0.807 (72.3% sensitivity and 80.9% specificity), which was significantly higher (p < 0.05) than the traditional and DL models. CONCLUSIONS: DL models using baseline knee radiographs had higher diagnostic performance for predicting pain progression than traditional models using demographic, clinical, and radiographic risk factors.
OBJECTIVE: To develop and evaluate deep learning (DL) risk assessment models for predicting pain progression in subjects with or at risk of knee osteoarthritis (OA). MATERIALS AND METHODS: The incidence and progression cohorts of the Osteoarthritis Initiative, a multi-center longitudinal study involving 9348 knees in 4674 subjects with or at risk of knee OA that began in 2004 and is ongoing, were used to conduct this retrospective analysis. A subset of knees without and with pain progression (defined as a 9-point or greater increase in pain score between baseline and two or more follow-up time points over the first 48 months) was randomly stratified into training (4200 knees with a mean age of 61.0 years and 60% female) and hold-out testing (500 knees with a mean age of 60.8 years and 60% female) datasets. A DL model was developed to predict pain progression using baseline knee radiographs. An artificial neural network was used to develop a traditional risk assessment model to predict pain progression using demographic, clinical, and radiographic risk factors. A combined model was developed to combine demographic, clinical, and radiographic risk factors with DL analysis of baseline knee radiographs. Area under the curve (AUC) analysis was performed using the hold-out testing dataset to evaluate model performance. RESULTS: The traditional model had an AUC of 0.692 (66.9% sensitivity and 64.1% specificity). The DL model had an AUC of 0.770 (76.7% sensitivity and 70.5% specificity), which was significantly higher (p < 0.001) than the traditional model. The combined model had an AUC of 0.807 (72.3% sensitivity and 80.9% specificity), which was significantly higher (p < 0.05) than the traditional and DL models. CONCLUSIONS: DL models using baseline knee radiographs had higher diagnostic performance for predicting pain progression than traditional models using demographic, clinical, and radiographic risk factors.
Authors: Tuhina Neogi; Laura Frey-Law; Joachim Scholz; Jingbo Niu; Lars Arendt-Nielsen; Clifford Woolf; Michael Nevitt; Laurence Bradley; David T Felson Journal: Ann Rheum Dis Date: 2013-12-18 Impact factor: 19.103
Authors: Marieke L A Landsmeer; Jos Runhaar; Marienke van Middelkoop; Edwin H G Oei; Dieuwke Schiphof; Patrick J E Bindels; Sita M A Bierma-Zeinstra Journal: J Am Board Fam Med Date: 2019 Jul-Aug Impact factor: 2.657
Authors: Przemyslaw T Paradowski; Martin Englund; L Stefan Lohmander; Ewa M Roos Journal: Health Qual Life Outcomes Date: 2005-09-27 Impact factor: 3.186
Authors: Amanda E Nelson; Thomas H Keefe; Todd A Schwartz; Leigh F Callahan; Richard F Loeser; Yvonne M Golightly; Liubov Arbeeva; J S Marron Journal: PLoS One Date: 2022-05-24 Impact factor: 3.752