Kevin A Thomas1, Dominik Krzemiński2, Łukasz Kidziński3, Rohan Paul1, Elka B Rubin4, Eni Halilaj5, Marianne S Black4, Akshay Chaudhari1,4, Garry E Gold3,4,6, Scott L Delp3,6,7. 1. Department of Biomedical Data Science, Stanford University, Stanford, CA, USA. 2. Cardiff University Brain Research Imaging Centre, Cardiff University, Cardiff, Wales, UK. 3. Department of Bioengineering, Stanford University, Stanford, CA, USA. 4. Department of Radiology, Stanford University, Stanford, CA, USA. 5. Department of Mechanical Engineering, Carnegie Mellon University, Pittsburgh, PA, USA. 6. Department of Orthopaedic Surgery, Stanford University, Stanford, CA, USA. 7. Department of Mechanical Engineering, Stanford University, Stanford, CA, USA.
Abstract
OBJECTIVE: We evaluated a fully automated femoral cartilage segmentation model for measuring T2 relaxation values and longitudinal changes using multi-echo spin-echo (MESE) magnetic resonance imaging (MRI). We open sourced this model and developed a web app available at https://kl.stanford.edu into which users can drag and drop images to segment them automatically. DESIGN: We trained a neural network to segment femoral cartilage from MESE MRIs. Cartilage was divided into 12 subregions along medial-lateral, superficial-deep, and anterior-central-posterior boundaries. Subregional T2 values and four-year changes were calculated using a radiologist's segmentations (Reader 1) and the model's segmentations. These were compared using 28 held-out images. A subset of 14 images were also evaluated by a second expert (Reader 2) for comparison. RESULTS: Model segmentations agreed with Reader 1 segmentations with a Dice score of 0.85 ± 0.03. The model's estimated T2 values for individual subregions agreed with those of Reader 1 with an average Spearman correlation of 0.89 and average mean absolute error (MAE) of 1.34 ms. The model's estimated four-year change in T2 for individual subregions agreed with Reader 1 with an average correlation of 0.80 and average MAE of 1.72 ms. The model agreed with Reader 1 at least as closely as Reader 2 agreed with Reader 1 in terms of Dice score (0.85 vs. 0.75) and subregional T2 values. CONCLUSIONS: Assessments of cartilage health using our fully automated segmentation model agreed with those of an expert as closely as experts agreed with one another. This has the potential to accelerate osteoarthritis research.
OBJECTIVE: We evaluated a fully automated femoral cartilage segmentation model for measuring T2 relaxation values and longitudinal changes using multi-echo spin-echo (MESE) magnetic resonance imaging (MRI). We open sourced this model and developed a web app available at https://kl.stanford.edu into which users can drag and drop images to segment them automatically. DESIGN: We trained a neural network to segment femoral cartilage from MESE MRIs. Cartilage was divided into 12 subregions along medial-lateral, superficial-deep, and anterior-central-posterior boundaries. Subregional T2 values and four-year changes were calculated using a radiologist's segmentations (Reader 1) and the model's segmentations. These were compared using 28 held-out images. A subset of 14 images were also evaluated by a second expert (Reader 2) for comparison. RESULTS: Model segmentations agreed with Reader 1 segmentations with a Dice score of 0.85 ± 0.03. The model's estimated T2 values for individual subregions agreed with those of Reader 1 with an average Spearman correlation of 0.89 and average mean absolute error (MAE) of 1.34 ms. The model's estimated four-year change in T2 for individual subregions agreed with Reader 1 with an average correlation of 0.80 and average MAE of 1.72 ms. The model agreed with Reader 1 at least as closely as Reader 2 agreed with Reader 1 in terms of Dice score (0.85 vs. 0.75) and subregional T2 values. CONCLUSIONS: Assessments of cartilage health using our fully automated segmentation model agreed with those of an expert as closely as experts agreed with one another. This has the potential to accelerate osteoarthritis research.
Authors: José G Tamez-Peña; Joshua Farber; Patricia C González; Edward Schreyer; Erika Schneider; Saara Totterman Journal: IEEE Trans Biomed Eng Date: 2012-02-03 Impact factor: 4.538
Authors: Gabby B Joseph; Charles E McCulloch; Michael C Nevitt; Jan Neumann; Alexandra S Gersing; Martin Kretzschmar; Benedikt J Schwaiger; John A Lynch; Ursula Heilmeier; Nancy E Lane; Thomas M Link Journal: J Magn Reson Imaging Date: 2017-11-16 Impact factor: 4.813
Authors: Joanne M Jordan; Charles G Helmick; Jordan B Renner; Gheorghe Luta; Anca D Dragomir; Janice Woodard; Fang Fang; Todd A Schwartz; Lauren M Abbate; Leigh F Callahan; William D Kalsbeek; Marc C Hochberg Journal: J Rheumatol Date: 2007-01 Impact factor: 4.666
Authors: Susanne M Eijgenraam; Akshay S Chaudhari; Max Reijman; Sita M A Bierma-Zeinstra; Brian A Hargreaves; Jos Runhaar; Frank W J Heijboer; Garry E Gold; Edwin H G Oei Journal: Eur Radiol Date: 2019-12-16 Impact factor: 5.315