Analysis of vertebrae without fracture on spine MRI to assess bone fragility: A Comparison of Traditional Machine Learning and Deep Learning

Ramos, Jonathan S.; Aguiar, Erikson Júlio de; Belizario, Ivar V.; Costa, Márcus V. L.; Maciel, Jamilly Gomes; Cazzolato, Mirela T.; Traina, Caetano; Nogueira-Barbosa, Marcello Henrique; Traina, Agma J. M.

doi:10.1109/cbms55023.2022.00021

Cited by 3 publications

(5 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Of the 57 studies using imaging-related and combined data, 33 analyzed radiograph images [ 6 , 7 , 28 – 31 , 35 – 38 , 40 – 42 , 45 , 47 – 49 , 52 – 57 , 59 , 61 , 62 , 66 – 68 , 72 – 74 , 78 ], 12 analyzed computed tomography (CT) images [ 8 , 9 , 39 , 43 , 50 , 63 , 65 , 69 , 75 , 81 – 83 ], and the remaining studies analyzed other imaging modalities ( S1 Table , and S2 Table ). The most common fracture outcome was vertebral fracture (20 studies) [ 8 , 10 , 11 , 28 , 31 , 34 , 35 , 38 , 44 , 46 , 50 , 51 , 58 , 59 , 65 , 72 , 77 , 80 ,…”

Section: Resultsmentioning

confidence: 99%

“…No universal rule of data separation was found. A different set of split samples was utilized, e.g., 80% training and 20% testing [ 10 , 28 , 47 , 57 , 71 ], 90% training and 10% testing [ 32 , 33 , 56 , 81 ], and 80% training, 10% validation, and 10% testing [ 40 , 41 , 65 , 69 ]. Twenty studies reported the cross-validation with 20-folds [ 66 ], 10-folds [ 8 , 14 , 33 , 34 , 39 , 45 , 50 , 53 , 57 , 64 , 72 , 76 , 80 , 81 ], 5-folds [ 13 , 28 , 32 , 38 , 44 , 46 , 48 , 67 , 74 , 78 , 79 ], and 7-folds [ 83 ].…”

Section: Resultsmentioning

confidence: 99%

“…The selected studies were published between 2007 and 2022, with 73% (48 studies) published in the last three years (Table 1). The studies were conducted in various countries, including Asian countries (26 studies) [6,9,11,, North American countries (19 studies) [14,34,36,[51][52][53][54][55][56][57][58][59][60][61][62][63][64][65][66], European countries (14 studies) [13,59,[67][68][69][70][71][72][73][74][75][76][77][78], Australia (1 study) [79] and Brazil (2 studies) [10,80] (Table 1). Four studies did not provide the country information [81][82][83][84].…”

Section: Study Selection and Characteristicsmentioning

confidence: 99%

“…studies) [8,10,11,28,31,34,35,38,44,46,50,51,58,59,65,72,77,80,83,84], followed by hip [6,13,29,32,33,37,[39][40][41][42][43]48,53,62,64,66,68,79], and other fracture types (Table 1).…”

Section: Plos Digital Healthmentioning

confidence: 99%

“…Artificial Intelligence (AI), encompassing Machine Learning (ML) and Deep Learning (DL), has been extensively employed for fracture outcome prediction due to technological advancements and accessibility. Various imaging modalities, including X-rays [ 6 , 7 ], computed tomography (CT) [ 8 , 9 ], and magnetic resonance imaging (MRI) [ 10 , 11 ], have been used in fracture diagnosis and detection. AI can also predict fractures using tabular data, such as electronic medical records (structured patient-level data).…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Artificial intelligence in fracture detection with different image modalities and data types: A systematic review and meta-analysis

Jung,

Dai,

Liu

et al. 2024

PLOS Digit Health

View full text Add to dashboard Cite

Artificial Intelligence (AI), encompassing Machine Learning and Deep Learning, has increasingly been applied to fracture detection using diverse imaging modalities and data types. This systematic review and meta-analysis aimed to assess the efficacy of AI in detecting fractures through various imaging modalities and data types (image, tabular, or both) and to synthesize the existing evidence related to AI-based fracture detection. Peer-reviewed studies developing and validating AI for fracture detection were identified through searches in multiple electronic databases without time limitations. A hierarchical meta-analysis model was used to calculate pooled sensitivity and specificity. A diagnostic accuracy quality assessment was performed to evaluate bias and applicability. Of the 66 eligible studies, 54 identified fractures using imaging-related data, nine using tabular data, and three using both. Vertebral fractures were the most common outcome (n = 20), followed by hip fractures (n = 18). Hip fractures exhibited the highest pooled sensitivity (92%; 95% CI: 87–96, p< 0.01) and specificity (90%; 95% CI: 85–93, p< 0.01). Pooled sensitivity and specificity using image data (92%; 95% CI: 90–94, p< 0.01; and 91%; 95% CI: 88–93, p < 0.01) were higher than those using tabular data (81%; 95% CI: 77–85, p< 0.01; and 83%; 95% CI: 76–88, p < 0.01), respectively. Radiographs demonstrated the highest pooled sensitivity (94%; 95% CI: 90–96, p < 0.01) and specificity (92%; 95% CI: 89–94, p< 0.01). Patient selection and reference standards were major concerns in assessing diagnostic accuracy for bias and applicability. AI displays high diagnostic accuracy for various fracture outcomes, indicating potential utility in healthcare systems for fracture diagnosis. However, enhanced transparency in reporting and adherence to standardized guidelines are necessary to improve the clinical applicability of AI. Review Registration: PROSPERO (CRD42021240359).

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Section: Study Selection and Characteristicsmentioning

confidence: 99%

“…studies) [8,10,11,28,31,34,35,38,44,46,50,51,58,59,65,72,77,80,83,84], followed by hip [6,13,29,32,33,37,[39][40][41][42][43]48,53,62,64,66,68,79], and other fracture types (Table 1).…”

Section: Plos Digital Healthmentioning

confidence: 99%