An exemplar pyramid feature extraction based humerus fracture classification method

Machine Learning Solutions for Osteoporosis—A Review

“…( 66,67,71,72,77–86,90,91,93–95 ) Others used features extracted from images or collected from non‐imaging data. ( 68–70,73–77,87–89,92,96,97 ) Studies reported average best AUC of 0.92 (range 0.63 to 1.00) and average best accuracy of 89.8% (range 78.4% to 99.1%]. Surprising findings were that hospital‐related variables, such as the scanner device model, were better predictors of fractures than patients' characteristics or their images (AUC = 0.89 [95% confidence interval (CI) 0.87–0.91] versus 0.79 [95% CI 0.75–0.82] and 0.78 [95% CI 0.74–0.81], respectively), ( 77 ) which the authors acknowledged as a potential bias induced by the triage process.…”

“…( 82 ) Several studies did not report the confidence intervals or standard errors for the performance metrics that they reported. ( 66,68,69,72–74,79,81,83,84,86,88,89,92 ) Four studies obtained both high overall quality scores and near perfect AUCs (AUC ≥ 0.99). ( 78,79,91,94 ) Possible reasons for these unusual performances are the use of images' regions of interest ( 79,91 ) or the presence of potential data set bias, ( 78 ) which can significantly simplify the task.…”

“…Among the 32 studies that investigated fracture detection (Table 3), (66–97 ) 11 were on vertebral fractures, ( 66–76 ) 17 hip fractures, ( 74–90 ) and 10 other fracture sites such as humerus or wrist. ( 75,76,90–97 ) Nineteen studies developed CNN models for image analysis. ( 66,67,71,72,77–86,90,91,93–95 ) Others used features extracted from images or collected from non‐imaging data.…”

“…( 66,67,78,79,84,90,94 ) However, several studies did not use techniques that allowed for adequate understanding of the models' decision‐making (eg, data visualization). ( 68,70–76,80,82,83,85–89,92,93,95,97 ) In some studies, class unbalancing was handled using sampling methods, ( 74,97 ) and overfitting was controlled with data augmentation. ( 80,83,84,95 ) Mutasa and colleagues demonstrated how data augmentation using digitally reconstructed and generated images was able to improve the test AUC from 0.80 (95% CI, not reported) to 0.92 (95% CI, not reported).…”

“…( 33,34,44,52–65,72,73,86,88,92,102 ) In general, the studies where these issues were observed were not adequately acknowledged. ( 29,51,60,72,92 ) Acknowledging potential biases or limitations is fundamental for the accurate interpretation and especially for claims related to clinical implication. Nevertheless, several studies showed a greater awareness of the potential existence of bias and made efforts to address this either directly by controlling for the risks of overfitting, ( 23,28,30,32,34,37–40,43–46,49,53,54,56,57,59,64,77–85,90,91,93–95 ) indirectly by visualizing decision‐making, ( 36,38,48,66,67,78,79,84,90,94,104 ) or by using gold standard strategies for comparison.…”