The differentiation between aneurysmal bone cyst and telangiectatic osteosarcoma: a clinical, radiographic and MRI study

Zishan, Umme Sara; Pressney, Ian; Khoo, Michael; Saifuddin, Asif

doi:10.1007/s00256-020-03432-w

Cited by 39 publications

(30 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The expansile growth pattern with well-defined margins exhibited by both of the present tumors is consistent with previously documented observations of rib osteoblastomas; however, these are somewhat aggressive features that are frequently considered to be on the borderline of osteoblastoma and low-grade osteosarcoma [ 38 , 39 ]. Aneurysmal bone cyst ( Figure 3 f) shares several clinical and imaging features with telangiectatic osteosarcoma, such as young age at presentation, large size, and heterogeneous to high T 2 signal intensity corresponding to fluid levels [40] . The “black-box” nature of deep neural networks makes it challenging to explain why our model was able to achieve the correct classification in some of these cases but was similarly misled in others.…”

Section: Discussionmentioning

confidence: 99%

Deep Learning for Classification of Bone Lesions on Routine MRI

Eweje

Bao

et al. 2021

EBioMedicine

View full text Add to dashboard Cite

Background Radiologists have difficulty distinguishing benign from malignant bone lesions because these lesions may have similar imaging appearances. The purpose of this study was to develop a deep learning algorithm that can differentiate benign and malignant bone lesions using routine magnetic resonance imaging (MRI) and patient demographics. Methods 1,060 histologically confirmed bone lesions with T1- and T2-weighted pre-operative MRI were retrospectively identified and included, with lesions from 4 institutions used for model development and internal validation, and data from a fifth institution used for external validation. Image-based models were generated using the EfficientNet-B0 architecture and a logistic regression model was trained using patient age, sex, and lesion location. A voting ensemble was created as the final model. The performance of the model was compared to classification performance by radiology experts. Findings The cohort had a mean age of 30±23 years and was 58.3% male, with 582 benign lesions and 478 malignant. Compared to a contrived expert committee result, the ensemble deep learning model achieved (ensemble vs. experts): similar accuracy (0·76 vs. 0·73, p=0·7), sensitivity (0·79 vs. 0·81, p=1·0) and specificity (0·75 vs. 0·66, p=0·48), with a ROC AUC of 0·82. On external testing, the model achieved ROC AUC of 0·79. Interpretation Deep learning can be used to distinguish benign and malignant bone lesions on par with experts. These findings could aid in the development of computer-aided diagnostic tools to reduce unnecessary referrals to specialized centers from community clinics and limit unnecessary biopsies. Funding This work was funded by a Radiological Society of North America Research Medical Student Grant (#RMS2013) and supported by the Amazon Web Services Diagnostic Development Initiative.

show abstract

Section: Discussionmentioning

confidence: 99%

Deep Learning for Classification of Bone Lesions on Routine MRI

Eweje

Bao

et al. 2021

EBioMedicine

View full text Add to dashboard Cite

show abstract

“…It is important to differentiate between TOS and aneurysmal bone cysts, given that they exhibit similar radiological and pathological phenotypes (27,28). Invasive growth pattern is occasionally accompanied by matrix mineralization, multiple fluid filled cavities of different sizes, surrounded by thick peripheral, septal and nodular tissue, which are helpful for the correct diagnosis of TOS (29).…”

Section: Discussionmentioning

confidence: 99%

Characteristics and prognosis of telangiectatic osteosarcoma: a population-based study using the Surveillance, Epidemiology and End Results (SEER) database

Xu¹,

Dai²,

Yang³

et al. 2021

Ann Transl Med

View full text Add to dashboard Cite

Background: Telangiectatic osteosarcoma (TOS) is a rare type of osteosarcoma for which limited clinical data is available. Furthermore, the clinical characteristics and prognosis of TOS remain unclear.Methods: A large population-based cohort analysis was conducted using the Surveillance, Epidemiology and End Results (SEER) registry. The data of TOS and conventional osteosarcoma (COS) patients from 2000 to 2017 were collected. The categorical variables were assessed by Chi-squared tests. Kaplan-Meier curves and log-rank (Mantel-Cox) tests were used to examine the survival outcomes between the groups.Cox proportional hazard models were used for univariate and multivariate analyses of TOS patient survivalrelated variables.Results: A total of 141 TOS patients and 2961 COS patients were included in this analysis, and the mean age at diagnosis was 23.5 and 29.4 years, respectively. Compared to COS patients, TOS patients were more likely to be under 20 years old (61.7% vs. 51.7%, P=0.022), and without a second peak of incidence after 60 years of age. The median overall survival (mOS) of TOS patients was not reached compared to a median survival of 84 months for COS patients (hazard ratio 0.75, 95% confidence interval 0.59 to 0.95, P=0.0175).After adjusting these data for age at diagnosis, stage, and surgery at the primary site, no significant differences in mOS were observed between the two groups. In univariate analyses, being under 20 years of age, having localized or regional stage disease, and having undergone surgery were associated with a decreased risk of death. Subsequent multivariate analysis indicated that age at diagnosis, stage, and surgery at the primary site were all independent predictors of prognosis in TOS patients.Conclusions: Patients with TOS were younger than patients with COS and did not show a second peak after 60 years of age. Age, summary stage at diagnosis, and surgery at the primary site were independent predictors of survival for TOS patients.

show abstract

“…ABC-like changes most frequently accompany giant cell tumour, chondroblastoma and fibrous dysplasia but are also well-documented in association with osteoblastoma [6]. Telangiectatic osteosarcoma is also a well-recognised cause [7]. The differential diagnosis would include primary ABC, as well as ABC-like change within an osteoid osteoma (although to our knowledge this has not been reported), chondroblastoma or GCT.…”

mentioning

confidence: 88%