Radiomics in Stroke Neuroimaging: Techniques, Applications, and Challenges

Chen, Qian; Xia, Tianyi; Mingyue, Zhang; Xia, Nengzhi; Liu, Jinjin; Yang, Yunjun

doi:10.14336/ad.2020.0421

Cited by 67 publications

(44 citation statements)

References 80 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…It provides numerical acquisition of pixel-level differences in areas that the human eye cannot distinguish on images [22]. In the literature, histogram analysis has been used in the detection and classification of brain tumors in numerous diseases such as multiple sclerosis, acute ischemia, Alzheimer's, and tinnitus [23][24][25][26][27]. In a study on multiple sclerosis patients, histogram analysis showed a difference in the structure of the optic nerve compared to the controls [28].…”

Section: Discussionmentioning

confidence: 99%

MRI histogram analysis of optic nerves in children with type 1 neurofibromatosis

Eroğlu¹,

Baykara²

2022

Journal of Surgery and Medicine

View full text Add to dashboard Cite

Background/Aim: Type 1 neurofibromatosis (NF1) is the most common neurocutaneous disease affecting numerous systems. Optic pathway glioma (OPG) is a common tumor in children with NF1 and often has variable clinical presentations. In this study, histogram analysis parameters of optic nerves were measured in the magnetic resonance images (MRI) of children with NF1 and compared with a control group. Methods: This case-control study consisted of three groups: Ten patients with NF1 without optic pathway glioma (bilateral optic nerve, n: 20), four patients with NF1 with bilateral optic pathway glioma (n: 8), and nineteen healthy controls (n: 38). ROIs were placed on bilateral pre-chiasmatic optic nerves in the images. With histogram analysis, average gray level intensity (mean), the standard deviation, minimum, median, and maximum intensity, uniformity, entropy, kurtosis, variance, skewness, size% M, size% U, size% L, and percentiles were measured. Results: Mean, median, 3%, 5%, 10%, 25%, and 75% values were higher in NF1 patients with optic pathway glioma (NF1-OPG) than in NF1 patients without optic pathway glioma (NF1-woOPG), and the control group (P<0.001). The same values were significantly higher in the NF1-woOPG group compared to the control group (P<0.001). The minimum, maximum, 1%, 90%, 95%, 97%, and 99% values were significantly higher in the NF1-OPG and NF1-woOPG groups than the control group (P<0.001). The entropy value was significantly higher in the NF1-OPG group than the NF1-woOPG and control groups (5.73, 4.93, and 5.25, respectively, P=0.016). Conclusion: MRI histogram analysis revealed significant differences between NF1-OPG, NF1-woOPG, and healthy individuals in terms of optic nerves. Thus, we think that it can be used to monitor the optic nerves of children with NF1.

show abstract

Section: Discussionmentioning

confidence: 99%

MRI histogram analysis of optic nerves in children with type 1 neurofibromatosis

Eroğlu¹,

Baykara²

2022

Journal of Surgery and Medicine

View full text Add to dashboard Cite

show abstract

“…As the images were acquired from different CT scanners with different parameters, we performed data preprocessing before radiomics feature extraction ( Lambin et al, 2017 ; Morin et al, 2018 ; Chen et al, 2021 ). Image resampling and gray-level discretization were used to reduce the variability of radiomics features ( Shafiq-Ul-Hassan et al, 2018 ).…”

Section: Methodsmentioning

confidence: 99%

“…Radiomics is an emerging technology that extracts high-throughput data from medical images ( Zhou et al, 2018 ; Hua et al, 2020 ; Tomaszewski and Gillies, 2021 ). Recently, radiomics is frequently used in cerebrovascular disease researches ( Chen et al, 2021 ; Zhu D. et al, 2021 ; Zhu D. Q. et al, 2021 ). Several researchers have scoped to the whole-brain aneurysms and proved that radiomics features were related to aneurysms rupture status ( Liu et al, 2019 ; Ou et al, 2021 ).…”

Section: Introductionmentioning

confidence: 99%

Classifying Ruptured Middle Cerebral Artery Aneurysms With a Machine Learning Based, Radiomics-Morphological Model: A Multicentral Study

et al. 2021

Self Cite

View full text Add to dashboard Cite

ObjectiveRadiomics and morphological features were associated with aneurysms rupture. However, the multicentral study of their predictive power for specific-located aneurysms rupture is rare. We aimed to determine robust radiomics features related to middle cerebral artery (MCA) aneurysms rupture and evaluate the additional value of combining morphological and radiomics features in the classification of ruptured MCA aneurysms.MethodsA total of 632 patients with 668 MCA aneurysms (423 ruptured aneurysms) from five hospitals were included. Radiomics and morphological features of aneurysms were extracted on computed tomography angiography images. The model was developed using a training dataset (407 patients) and validated with the internal (152 patients) and external validation (73 patients) datasets. The support vector machine method was applied for model construction. Optimal radiomics, morphological, and clinical features were used to develop the radiomics model (R-model), morphological model (M-model), radiomics-morphological model (RM-model), clinical-morphological model (CM-model), and clinical-radiomics-morphological model (CRM-model), respectively. A comprehensive nomogram integrating clinical, morphological, and radiomics predictors was generated.ResultsWe found seven radiomics features and four morphological predictors of MCA aneurysms rupture. The R-model obtained an area under the receiver operating curve (AUC) of 0.822 (95% CI, 0.776, 0.867), 0.817 (95% CI, 0.744, 0.890), and 0.691 (95% CI, 0.567, 0.816) in the training, temporal validation, and external validation datasets, respectively. The RM-model showed an AUC of 0.848 (95% CI, 0.810, 0.885), 0.865 (95% CI, 0.807, 0.924), and 0.721 (95% CI, 0.601, 0.841) in the three datasets. The CRM-model obtained an AUC of 0.856 (95% CI, 0.820, 0.892), 0.882 (95% CI, 0.828, 0.936), and 0.738 (95% CI, 0.618, 0.857) in the three datasets. The CRM-model and RM-model outperformed the CM-model and M-model in the internal datasets (p < 0.05), respectively. But these differences were not statistically significant in the external dataset. Decision curve analysis indicated that the CRM-model obtained the highest net benefit for most of the threshold probabilities.ConclusionRobust radiomics features were determined related to MCA aneurysm rupture. The RM-model exhibited good ability in classifying ruptured MCA aneurysms. Integrating radiomics features into conventional models might provide additional value in ruptured MCA aneurysms classification.

show abstract

“…Radiomics is a quantitative mapping approach to structural MRI, aiming at analyzing, extracting and quantifying the high-throughput features of human brain available to clinicians (Lambin et al, 2017 ). MRI-based radiomics, as a rapidly developing field, is an encouraging tool for the identification and quantification of region-specific brain features (Bretzner et al, 2021 ), which has been successfully applied in clinical practice, such as brain tumor (Zhang et al, 2019 ) and stroke (Chen et al, 2021 ). To improve the TUS protocols in age-related brain diseases, except for skull thickness, another transcranial variable, scalp-to-cortex distance (SCD), in the combination of cortical thickness, should be added into the head models for the individuals with different age ranges.…”

Section: The Importance Of Radiomic Features In Tusmentioning

confidence: 99%

Radiomics-Informed Modeling for Transcranial Ultrasound Stimulation: Age Matters

2022

Front. Neurosci.

View full text Add to dashboard Cite

Radiomics in Stroke Neuroimaging: Techniques, Applications, and Challenges

Cited by 67 publications

References 80 publications

MRI histogram analysis of optic nerves in children with type 1 neurofibromatosis

MRI histogram analysis of optic nerves in children with type 1 neurofibromatosis

Classifying Ruptured Middle Cerebral Artery Aneurysms With a Machine Learning Based, Radiomics-Morphological Model: A Multicentral Study

Radiomics-Informed Modeling for Transcranial Ultrasound Stimulation: Age Matters

Contact Info

Product

Resources

About