Vertebral body insufficiency fractures: detection of vertebrae at risk on standard CT images using texture analysis and machine learning

Muehlematter, Urs J.; Mannil, Manoj; Becker, Anton S.; Vokinger, Kerstin Noëlle; Finkenstaedt, Tim; Osterhoff, Georg; Fischer, Michael A.; Guggenberger, Roman

doi:10.1007/s00330-018-5846-8

Cited by 82 publications

(66 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Predicting the risk of bone loss, osteoporotic fractures, falls, or comorbidities in osteoporotic patients over time was investigated in 14 studies (Table 4). ( 98–111 ) Two of them used unsupervised learning to identify fracture and comorbidity risk groups, respectively. ( 98,99 ) Kruse and colleagues developed a fracture risk clustering model to categorize subgroups of patients at risk.…”

Section: Resultsmentioning

confidence: 99%

“…( 98 ) Wang and colleagues investigated osteoporotic patients' subgroups and their related comorbidity risk. ( 99 ) The 12 remaining studies used supervised learning for the prediction of risk of osteoporosis by bone density loss at 10 years, ( 100 ) incident falls at 6 months ( 102 ) and 1 year, ( 101 ) incident vertebral fracture at ≈ 8 months, ( 103 ) hip fracture prediction at 4, 5, or 10 years, ( 107–111 ) vertebral or hip fractures at ≈ 7.5 years, ( 104 ) major osteoporotic fractures (hip, spine, wrist, or humerus) at ≈ 4.5 years, ( 105 ) and all sort of fracture sites at 1 and 2 years. ( 106 ) Because unsupervised learning is not intended to predict a predetermined outcome, no performance metrics were reported.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Machine Learning Solutions for Osteoporosis—A Review

Smets

Shevroja

Hügle

et al. 2020

Journal of Bone and Mineral Research

View full text Add to dashboard Cite

Osteoporosis and its clinical consequence, bone fracture, is a multifactorial disease that has been the object of extensive research. Recent advances in machine learning (ML) have enabled the field of artificial intelligence (AI) to make impressive breakthroughs in complex data environments where human capacity to identify high‐dimensional relationships is limited. The field of osteoporosis is one such domain, notwithstanding technical and clinical concerns regarding the application of ML methods. This qualitative review is intended to outline some of these concerns and to inform stakeholders interested in applying AI for improved management of osteoporosis. A systemic search in PubMed and Web of Science resulted in 89 studies for inclusion in the review. These covered one or more of four main areas in osteoporosis management: bone properties assessment (n = 13), osteoporosis classification (n = 34), fracture detection (n = 32), and risk prediction (n = 14). Reporting and methodological quality was determined by means of a 12‐point checklist. In general, the studies were of moderate quality with a wide range (mode score 6, range 2 to 11). Major limitations were identified in a significant number of studies. Incomplete reporting, especially over model selection, inadequate splitting of data, and the low proportion of studies with external validation were among the most frequent problems. However, the use of images for opportunistic osteoporosis diagnosis or fracture detection emerged as a promising approach and one of the main contributions that ML could bring to the osteoporosis field. Efforts to develop ML‐based models for identifying novel fracture risk factors and improving fracture prediction are additional promising lines of research. Some studies also offered insights into the potential for model‐based decision‐making. Finally, to avoid some of the common pitfalls, the use of standardized checklists in developing and sharing the results of ML models should be encouraged. © 2021 American Society for Bone and Mineral Research (ASBMR).

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Machine Learning Solutions for Osteoporosis—A Review

Smets

Shevroja

Hügle

et al. 2020

Journal of Bone and Mineral Research

View full text Add to dashboard Cite

show abstract

“…(9)(10)(11) Improvement in image analytics, utilizing machine learning and artificial intelligence algorithms, has been shown to provide accurate diagnoses from radiological images. (12)(13)(14)(15)(16)(17) Computeraided diagnosis (CAD) software employing machine-learning for detection of vertebral fractures have been developed (18)(19)(20)(21) and have the potential to act as a resource-effective screening tool to detect at-risk individuals for secondary prevention of fractures. However, previous studies (12,18,22,23) have tested their respective CAD using relatively small constructed datasets at the training site; ie, where the software was "trained" to detect fracture.…”

Section: Introductionmentioning

confidence: 99%

Clinical Utility of Computer-Aided Diagnosis of Vertebral Fractures From Computed Tomography Images

Kolanu

Silverstone

et al. 2020

Journal of Bone and Mineral Research

View full text Add to dashboard Cite

Osteoporotic vertebral compression fractures (VCFs) are a risk factor for morbidity and mortality, frequently asymptomatic and often present in computed tomography (CT) scans performed for unrelated conditions. Computer‐aided diagnosis (CAD) of VCF from such images can potentially improve identification and treatment of osteoporosis. This single‐blinded, single tertiary center study compared a CAD (Zebra Medical Vision®) to an adjudicated imaging specialist reevaluation using a retrospective consecutive sample of abdominal and thoracic CT scans (n = 2357) performed as part of routine care. Subjects over 50 years between January 1, 2019 and May 12, 2019 were included. Duplicates and unanalyzable scans were excluded resulting in a total of 1696 CT scans. The sensitivity, specificity, and accuracy were calculated for all VCF and for Genant grades 2 or 3 (ie, height loss of >25%) using imaging specialist as the gold standard. Prestudy VCF reporting by hospital‐rostered radiologist was used to calculate the number of scans needed to screen (NNS) to detect one additional VCF using CAD. Prevalence of any VCF was 24% (406/1696) and of Genant 2/3 VCF was 18% (280/1570). The sensitivity and specificity were 54% and 92%, for all fractures, respectively, and 65% and 92% for Genant 2/3 fractures, respectively. Accuracy for any VCF, and for detection of Genant 2/3 VCF, was 83% and 88%, respectively. Of 221 CAD‐detected VCFs, 133 (60.2%) were reported prestudy resulting in 88 additional fractures (72 Genant 2/3) being identified by CAD. NNS to detect one additional VCF was 19 scans for all fractures and 23 for Genant 2/3 fractures. Thus, the CAD tested in this study had a high specificity with moderate sensitivity to detect incidental vertebral fractures in CT scans performed for routine care. A low NNS suggests it is an efficient tool to assist radiologists and clinicians to improve detection and reporting of vertebral fractures. © 2020 American Society for Bone and Mineral Research (ASBMR).

show abstract

“…Different methods for radiological assessment of vertebral fractures exist, including radiographs, CT scans and vertebral fracture software in DXA. The automated detection of prevalent vertebral fracture on CT scans using artificial intelligence technologies will be another avenue for secondary fracture prevention [56,57].…”

Section: What Can We Do To Optimise Osteoporosis Care?mentioning

confidence: 99%

General and Specific Considerations as to why Osteoporosis-Related Care Is Often Suboptimal

Curtis

Woolford

Holmes

et al. 2020

Curr Osteoporos Rep

View full text Add to dashboard Cite

Purpose of Review The assessment of fracture risk and use of antiosteoporosis medications have increased greatly over the last 20-30 years. However, despite this, osteoporosis care remains suboptimal worldwide. Even in patients who have sustained a fragility fracture, fewer than 20% actually receive appropriate antiosteoporosis therapy in the year following the fracture. There is also evidence that treatment rates have declined substantially in the last 5-10 years, in many countries. The goal of this article is to consider the causes for this decline and consider how this situation could be remedied. Recent Findings A number of possible reasons, including the lack of prioritisation of osteoporosis therapy in ageing populations with multimorbidity, disproportionate concerns regarding the rare side effects of anti-resorptives and adverse changes in reimbursement in the USA, have been identified as contributing factors in poor osteoporosis care. Summary Improved secondary prevention strategies; screening measures (primary prevention) and appropriate, cost-effective guideline and treatment threshold development could support the optimisation of osteoporosis care and prevention of future fractures.

show abstract

Vertebral body insufficiency fractures: detection of vertebrae at risk on standard CT images using texture analysis and machine learning

Cited by 82 publications

References 37 publications

Machine Learning Solutions for Osteoporosis—A Review

Machine Learning Solutions for Osteoporosis—A Review

Clinical Utility of Computer-Aided Diagnosis of Vertebral Fractures From Computed Tomography Images

General and Specific Considerations as to why Osteoporosis-Related Care Is Often Suboptimal

Contact Info

Product

Resources

About