1
|
Schilcher J, Nilsson A, Andlid O, Eklund A. Fusion of electronic health records and radiographic images for a multimodal deep learning prediction model of atypical femur fractures. Comput Biol Med 2024; 168:107704. [PMID: 37980797 DOI: 10.1016/j.compbiomed.2023.107704] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/18/2023] [Revised: 10/15/2023] [Accepted: 11/07/2023] [Indexed: 11/21/2023]
Abstract
Atypical femur fractures (AFF) represent a very rare type of fracture that can be difficult to discriminate radiologically from normal femur fractures (NFF). AFFs are associated with drugs that are administered to prevent osteoporosis-related fragility fractures, which are highly prevalent in the elderly population. Given that these fractures are rare and the radiologic changes are subtle currently only 7% of AFFs are correctly identified, which hinders adequate treatment for most patients with AFF. Deep learning models could be trained to classify automatically a fracture as AFF or NFF, thereby assisting radiologists in detecting these rare fractures. Historically, for this classification task, only imaging data have been used, using convolutional neural networks (CNN) or vision transformers applied to radiographs. However, to mimic situations in which all available data are used to arrive at a diagnosis, we adopted an approach of deep learning that is based on the integration of image data and tabular data (from electronic health records) for 159 patients with AFF and 914 patients with NFF. We hypothesized that the combinatorial data, compiled from all the radiology departments of 72 hospitals in Sweden and the Swedish National Patient Register, would improve classification accuracy, as compared to using only one modality. At the patient level, the area under the ROC curve (AUC) increased from 0.966 to 0.987 when using the integrated set of imaging data and seven pre-selected variables, as compared to only using imaging data. More importantly, the sensitivity increased from 0.796 to 0.903. We found a greater impact of data fusion when only a randomly selected subset of available images was used to make the image and tabular data more balanced for each patient. The AUC then increased from 0.949 to 0.984, and the sensitivity increased from 0.727 to 0.849. These AUC improvements are not large, mainly because of the already excellent performance of the CNN (AUC of 0.966) when only images are used. However, the improvement is clinically highly relevant considering the importance of accuracy in medical diagnostics. We expect an even greater effect when imaging data from a clinical workflow, comprising a more diverse set of diagnostic images, are used.
Collapse
Affiliation(s)
- Jörg Schilcher
- Department of Orthopedics and Experimental and Clinical Medicine, Faculty of Health Science, Linköping University, Linköping, Sweden; Wallenberg Centre for Molecular Medicine, Linköping University, Linköping, Sweden; Center for Medical Image Science and Visualization (CMIV), Linköping University, Linköping, Sweden
| | - Alva Nilsson
- Department of Biomedical Engineering, Linköping University, Linköping, Sweden
| | - Oliver Andlid
- Department of Biomedical Engineering, Linköping University, Linköping, Sweden
| | - Anders Eklund
- Department of Biomedical Engineering, Linköping University, Linköping, Sweden; Division of Statistics and Machine Learning, Department of Computer and Information Science, Linköping University, Linköping, Sweden; Center for Medical Image Science and Visualization (CMIV), Linköping University, Linköping, Sweden.
| |
Collapse
|
2
|
Lo JC, Yang W, Park-Sigal JJ, Ott SM. Osteoporosis and Fracture Risk among Older US Asian Adults. Curr Osteoporos Rep 2023; 21:592-608. [PMID: 37542683 PMCID: PMC10858302 DOI: 10.1007/s11914-023-00805-7] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Accepted: 06/06/2023] [Indexed: 08/07/2023]
Abstract
PURPOSE OF REVIEW This review summarizes the current knowledge regarding osteoporosis and fracture among older US Asian adults. RECENT FINDINGS Asian adults have lower (areal) bone density than non-Hispanic White adults and thus are more likely to be diagnosed and treated for osteoporosis, despite their lower risk of hip fracture. The latter may relate to favorable characteristics in hip geometry, volumetric bone density, and bone microarchitecture; lower risk of falls; and other clinical factors. The fracture risk calculator FRAX accounts for the lower risk of hip fracture among US Asian adults. However, data on major osteoporotic fracture risk remain limited. Fracture rates also vary by Asian subgroup, which may have implications for fracture risk assessment. Furthermore, among women receiving bisphosphonate drugs, Asian race is a risk factor for atypical femur fracture, an uncommon complication associated with treatment duration. Recent clinical trial efficacy data pertaining to lower bisphosphonate doses and longer dosing intervals may be relevant for Asian adults. More research is needed to inform osteoporosis care of US Asian adults, including risk-benefit considerations and the optimal duration of bisphosphonate treatment. Greater evidence-based guidance for primary fracture prevention among US Asian adults will ensure health equity in the prevention of osteoporotic fractures.
Collapse
Affiliation(s)
- Joan C Lo
- Division of Research, Kaiser Permanente Northern California, 2000 Broadway, Oakland, CA, 94612, USA.
- The Permanente Medical Group, Oakland, CA, USA.
- Department of Health Systems Science, Kaiser Permanente Bernard J. Tyson School of Medicine, Pasadena, CA, USA.
| | - Wei Yang
- The Permanente Medical Group, Oakland, CA, USA
- Department of Endocrinology, Kaiser Permanente San Jose Medical Center, San Jose, CA, USA
| | - Jennifer J Park-Sigal
- The Permanente Medical Group, Oakland, CA, USA
- Department of Endocrinology, Kaiser Permanente South San Francisco Medical Center, South San Francisco, CA, USA
| | - Susan M Ott
- Department of Medicine, University of Washington School of Medicine, Seattle, WA, USA
| |
Collapse
|
3
|
Hart DA. Are secondary effects of bisphosphonates on the vascular system of bone contributing to increased risk for atypical femoral fractures in osteoporosis? Bioessays 2023; 45:e2200206. [PMID: 36807308 DOI: 10.1002/bies.202200206] [Citation(s) in RCA: 3] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/25/2022] [Revised: 01/27/2023] [Accepted: 01/30/2023] [Indexed: 02/22/2023]
Abstract
Osteoporosis (OP) is a bone disease which affects a number of post-menopausal females and puts many at risk for fractures. A large number of patients are taking bisphosphonates (BPs) to treat their OP and a rare complication is the development of atypical femoral fractures (AFF). No real explanations for the mechanisms underlying the basis for development of where AFF develop while on BPs has emerged. The present hypothesis will discuss the possibility that part of the risk for an AFF is a secondary effect of BPs on a subset of vascular cells in a genetically at-risk population, leading to localized deregulation of the endothelial cell (EC)-bone cell-matrix units in nutrient channels/canals of the femur and increased risk for AFF. This concept of targeting ECs is consistent with location of AFF in the femur, the bilateral risk for occurrence of AFF, and the requirement for long term exposure to the drugs.
Collapse
Affiliation(s)
- David A Hart
- Department of Surgery, Faculty of Kinesiology, McCaig Institute for Bone & Joint Health, University of Calgary, Calgary, Alberta, Canada
| |
Collapse
|
4
|
Haider IT, Loundagin LL, Sawatsky A, Kostenuik PJ, Boyd SK, Edwards WB. Twelve Months of Denosumab and/or Alendronate Is Associated With Improved Bone Fatigue Life, Microarchitecture, and Density in Ovariectomized Cynomolgus Monkeys. J Bone Miner Res 2023; 38:403-413. [PMID: 36533719 DOI: 10.1002/jbmr.4758] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 09/29/2022] [Revised: 11/25/2022] [Accepted: 12/14/2022] [Indexed: 12/23/2022]
Abstract
Prolonged use of antiresorptives such as the bisphosphonate alendronate (ALN) and the RANKL inhibitor denosumab (DMAb) are associated with rare cases of atypical femoral fracture (AFF). The etiology of AFF is unclear, but it has been hypothesized that potent osteoclast inhibitors may reduce bone fatigue resistance. The purpose of this study was to quantify the relationship between antiresorptive treatment and fatigue life (cycles to failure) in bone from ovariectomized cynomolgus monkeys. We analyzed humeral bone from 30 animals across five treatment groups. Animals were treated for 12 months with subcutaneous (sc) vehicle (VEH), sc DMAb (25 mg/kg/month), or intravenous (iv) ALN (50 μg/kg/month). Another group received 6 months VEH followed by 6 months DMAb (VEH-DMAb), and the final group received 6 months ALN followed by 6 months DMAb (ALN-DMAb). A total of 240 cortical beam samples were cyclically tested in four-point bending at 80, 100, 120, or 140 MPa peak stress. High-resolution imaging and density measurements were performed to evaluate bone microstructure and composition. Samples from the ALN (p = 0.014), ALN-DMAb (p = 0.008), and DMAb (p < 0.001) groups illustrated higher fatigue-life measurements than VEH. For example, at 140 MPa the VEH group demonstrated a median ± interquartile range (IQR) fatigue life of 1987 ± 10593 cycles, while animals in the ALN, ALN-DMAb, and DMAb groups survived 9850 ± 13648 (+395% versus VEH), 10493 ± 16796 (+428%), and 14495 ± 49299 (+629%) cycles, respectively. All antiresorptive treatment groups demonstrated lower porosity, smaller pore size, greater pore spacing, and lower number of canals versus VEH (p < 0.001). Antiresorptive treatment was also associated with greater apparent density, dry density, and ash density (p ≤ 0.03). We did not detect detrimental changes following antiresorptive treatments that would explain their association with AFF. In contrast, 12 months of treatment may have a protective effect against fatigue fractures. © 2022 American Society for Bone and Mineral Research (ASBMR).
Collapse
Affiliation(s)
- Ifaz T Haider
- Human Performance Lab, Faculty of Kinesiology, University of Calgary, Calgary, AB, Canada.,McCaig Institute for Bone and Joint Health, Cumming School of Medicine, University of Calgary, Calgary, AB, Canada
| | - Lindsay L Loundagin
- Human Performance Lab, Faculty of Kinesiology, University of Calgary, Calgary, AB, Canada.,McCaig Institute for Bone and Joint Health, Cumming School of Medicine, University of Calgary, Calgary, AB, Canada.,Department of Anatomy, Physiology and Pharmacology, University of Saskatchewan, Saskatoon, SK, Canada
| | - Andrew Sawatsky
- Human Performance Lab, Faculty of Kinesiology, University of Calgary, Calgary, AB, Canada.,McCaig Institute for Bone and Joint Health, Cumming School of Medicine, University of Calgary, Calgary, AB, Canada
| | - Paul J Kostenuik
- Phylon Pharma Services, Newbury Park, CA, USA.,School of Dentistry, University of Michigan (Adjunct), Ann Arbor, MI, USA
| | - Steven K Boyd
- McCaig Institute for Bone and Joint Health, Cumming School of Medicine, University of Calgary, Calgary, AB, Canada
| | - W Brent Edwards
- Human Performance Lab, Faculty of Kinesiology, University of Calgary, Calgary, AB, Canada.,McCaig Institute for Bone and Joint Health, Cumming School of Medicine, University of Calgary, Calgary, AB, Canada.,Department of Biomedical Engineering, Schulich School of Engineering, University of Calgary, Calgary, AB, Canada
| |
Collapse
|
5
|
Wang M, Wu Y, Girgis CM. Bisphosphonate Drug Holidays: Evidence from Clinical Trials and
Real‐World
Studies. JBMR Plus 2022; 6:e10629. [PMID: 35720669 PMCID: PMC9189912 DOI: 10.1002/jbm4.10629] [Citation(s) in RCA: 12] [Impact Index Per Article: 6.0] [Reference Citation Analysis] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 02/05/2022] [Revised: 03/26/2022] [Accepted: 04/08/2022] [Indexed: 11/09/2022] Open
Affiliation(s)
- Mawson Wang
- Department of Diabetes and Endocrinology Westmead Hospital Westmead NSW Australia
- Faculty of Medicine and Health University of Sydney Sydney NSW Australia
| | - Yu‐Fang Wu
- Department of Diabetes and Endocrinology Westmead Hospital Westmead NSW Australia
| | - Christian M Girgis
- Department of Diabetes and Endocrinology Westmead Hospital Westmead NSW Australia
- Faculty of Medicine and Health University of Sydney Sydney NSW Australia
- Department of Endocrinology Royal North Shore Hospital St Leonards NSW Australia
| |
Collapse
|