A deep learning radiomics model may help to improve the prediction performance of preoperative grading in meningioma

Yang, Liping; Xu, Panpan; Zhang, Ying; Cui, Nan; Wang, Menglu; Peng, Mengye; Gao, Chao; Wang, Tianzuo

doi:10.1007/s00234-022-02894-0

Cited by 22 publications

(6 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…6 Some studies have reported prediction of grading, differential diagnosis, and prognosis of meningiomas using radiomic signatures (e.g., T1C, T2, DWI, T1WI, T1C, and ADC mapping). 8,[26][27][28] Hence, we explored the radiomic signatures associated with meningioma sinus invasion. The model based on 20 signatures had the highest AUC for the training and validation cohorts.…”

Section: Discussionmentioning

confidence: 99%

A radiomics model enables prediction venous sinus invasion in meningioma

Wang

Cao

Zhang

et al. 2023

Ann Clin Transl Neurol

View full text Add to dashboard Cite

ObjectivePreoperative prediction of meningioma venous sinus invasion would facilitate the selection of surgical approaches and predicting the prognosis. To predict venous sinus invasion in meningiomas, we used radiomic signatures to construct a model based on preoperative contrast‐enhanced T1‐weighted (T1C) and T2‐weighted (T2) magnetic resonance imaging.MethodsIn total, 599 patients with pathologically confirmed meningioma were retrospectively enrolled. For each patient enrolled in this study, 1595 radiomic signatures were extracted from T1C and T2 image sequences. Pearson correlation analysis and recursive feature elimination were used to select the most relevant signatures extracted from different image sequences, and logistic regression algorithms were used to build a radiomic model for risk prediction of meningioma sinus invasion. Furthermore, a nomogram was built by incorporating clinical characteristics and radiomic signatures, and a decision curve analysis was used to evaluate the clinical utility of the nomogram.ResultsTwenty radiomic signatures that were significantly related to venous sinus invasion were screened from 3190 radiomic signatures. Venous sinus invasion was associated with tumor position, and the clinicoradiomic model that incorporated the above characteristics (20 radiomic signatures and tumor position) had the best discriminating ability. The areas under the curve for the training and validation cohorts were 0.857 (95% confidence interval [CI], 0.824–0.890) and 0.824 (95% CI, 0.752–0.8976), respectively.InterpretationThe clinicoradiomic model had good predictive performance for venous sinus invasion in meningioma, which can aid in devising surgical strategies and predicting prognosis.

show abstract

Section: Discussionmentioning

confidence: 99%

A radiomics model enables prediction venous sinus invasion in meningioma

Wang

Cao

Zhang

et al. 2023

Ann Clin Transl Neurol

View full text Add to dashboard Cite

show abstract

“…Through various ML algorithms applied to RA, we finally extracted two sets of 20 Ki-67- and p53-related features from each patient, which consisted of the first-order features adding more advanced high-order features. The entropy belongs to the GLCM feature pool reflecting the intensity of the spatial distribution, which means that the larger entropy value represents a greater tumor heterogeneity ( 25 ). Regarding the uniformity of the tumor texture of the GLSZM, high-grade meningiomas are featured by a larger proportion of tissue disruption, and thus a higher heterogeneity of the distribution of cells in the tumor lesions compared with low-grade lesions ( 25 ).…”

Section: Discussionmentioning

confidence: 99%

Preoperative prediction of Ki-67 and p53 status in meningioma using a multiparametric MRI-based clinical-radiomic model

et al. 2023

View full text Add to dashboard Cite

PurposeTo investigate the utility of preoperative multiparametric magnetic resonance imaging (mpMRI)-based clinical-radiomic analysis combined with machine learning (ML) algorithms in predicting the expression of the Ki-67 proliferative index and p53 tumor suppressor protein in patients with meningioma.MethodsThis multicenter retrospective study included 483 and 93 patients from two centers. The Ki-67 index was classified into high (Ki-67≥5%) and low (Ki-67<5%)-expressed groups, and the p53 index was classified into positive (p53≥5%) and negative (p53<5%)-expressed groups. Clinical and radiological features were analyzed using univariate and multivariate statistical analyses. Six ML models were performed with different types of classifiers to predict Ki-67 and p53 status.ResultsIn the multivariate analysis, larger tumor volumes (p<0.001), irregular tumor margin (p<0.001), and unclear tumor-brain interface (p<0.001) were independently associated with a high Ki-67 status, whereas the presence of both necrosis (p=0.003) and the dural tail sign (p=0.026) were independently associated with a positive p53 status. A relatively better performance was yielded from the model constructed by combined clinical and radiological features. The area under the curve (AUC) and accuracy of high Ki-67 were 0.820 and 0.867 in the internal test, and 0.666 and 0.773 in the external test, respectively. Regarding p53 positivity, the AUC and accuracy were 0.858 and 0.857 in the internal test, and 0.684 and 0.718 in the external test.ConclusionThe present study developed clinical-radiomic ML models to non-invasively predict Ki-67 and p53 expression in meningioma using mpMRI features, and provides a novel non-invasive strategy for assessing cell proliferation.

show abstract

“…The authors also demonstrated a superior deep learning performance over typical, hand-crafted features. A further enhanced T1-weighted image-based deep learning model was used by Yang et al [ 22 ] in differentiating low- and high-grade meningiomas. In their study, the combined deep learning–radiomics model outperformed both the deep learning and hand-crafted radiomics models working alone (test AUC: 0.935 vs. 0.918 vs. 0.718).…”

Section: Gradingmentioning

confidence: 99%

Meningioma Radiomics: At the Nexus of Imaging, Pathology and Biomolecular Characterization

Ugga

Spadarella

Pinto

et al. 2022

Cancers

View full text Add to dashboard Cite

Meningiomas are the most common extra-axial tumors of the central nervous system (CNS). Even though recurrence is uncommon after surgery and most meningiomas are benign, an aggressive behavior may still be exhibited in some cases. Although the diagnosis can be made by radiologists, typically with magnetic resonance imaging, qualitative analysis has some limitations in regard to outcome prediction and risk stratification. The acquisition of this information could help the referring clinician in the decision-making process and selection of the appropriate treatment. Following the increased attention and potential of radiomics and artificial intelligence in the healthcare domain, including oncological imaging, researchers have investigated their use over the years to overcome the current limitations of imaging. The aim of these new tools is the replacement of subjective and, therefore, potentially variable medical image analysis by more objective quantitative data, using computational algorithms. Although radiomics has not yet fully entered clinical practice, its potential for the detection, diagnostic, and prognostic characterization of tumors is evident. In this review, we present a wide-ranging overview of radiomics and artificial intelligence applications in meningioma imaging.

show abstract

A deep learning radiomics model may help to improve the prediction performance of preoperative grading in meningioma

Cited by 22 publications

References 35 publications

A radiomics model enables prediction venous sinus invasion in meningioma

A radiomics model enables prediction venous sinus invasion in meningioma

Preoperative prediction of Ki-67 and p53 status in meningioma using a multiparametric MRI-based clinical-radiomic model

Meningioma Radiomics: At the Nexus of Imaging, Pathology and Biomolecular Characterization

Contact Info

Product

Resources

About