Support vector machine classification of brain metastasis and radiation necrosis based on texture analysis in MRI

Larroza, Andrés; Moratal, David; Paredes-Sánchez, Alexandra; Soria-Olivas, Emilio; Chust, M.; Arribas, L.; Arana, Estanislao

doi:10.1002/jmri.24913

Cited by 94 publications

(104 citation statements)

References 41 publications

Supporting

Mentioning

100

Contrasting

Order By: Relevance

“…Larozza et al used a support vector machine classification and extracted 7 predictive features based upon texture analysis resulting in a sensitivity of 83% and a specificity of 82% to detect recurrent metastasis following radiosurgery (Larroza et al, 2015). In another study comprising 25 patients with brain metastasis, when using the five best features Tiwari et al observed a detection accuracy of 91% in the training set, resulting, however, in a diagnostic accuracy of only 50% in the validation set (Tiwari et al, 2016).…”

Section: Discussionmentioning

confidence: 99%

“…Textural feature analysis of inconclusive lesions on PET (Lohmann et al, 2017) and MR images (Larroza et al, 2015; Nardone et al, 2016; Pallavi et al, 2014; Tiwari et al, 2016) is another promising approach. It is based on the assumption that the microstructure of a process depends on the underlying pathology and is reflected in subtle differences in the radiological image that cannot be detected by means of human perception but can be made accessible by high-dimensional quantitative image analysis often referred to as “radiomics”.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Combined FET PET/MRI radiomics differentiates radiation injury from recurrent brain metastasis

Lohmann

Köcher

Ceccon

et al. 2018

NeuroImage: Clinical

124

View full text Add to dashboard Cite

BackgroundThe aim of this study was to investigate the potential of combined textural feature analysis of contrast-enhanced MRI (CE-MRI) and static O-(2-[18F]fluoroethyl)-L-tyrosine (FET) PET for the differentiation between local recurrent brain metastasis and radiation injury since CE-MRI often remains inconclusive.MethodsFifty-two patients with new or progressive contrast-enhancing brain lesions on MRI after radiotherapy (predominantly stereotactic radiosurgery) of brain metastases were additionally investigated using FET PET. Based on histology (n = 19) or clinicoradiological follow-up (n = 33), local recurrent brain metastases were diagnosed in 21 patients (40%) and radiation injury in 31 patients (60%). Forty-two textural features were calculated on both unfiltered and filtered CE-MRI and summed FET PET images (20–40 min p.i.), using the software LIFEx. After feature selection, logistic regression models using a maximum of five features to avoid overfitting were calculated for each imaging modality separately and for the combined FET PET/MRI features. The resulting models were validated using cross-validation. Diagnostic accuracies were calculated for each imaging modality separately as well as for the combined model.ResultsFor the differentiation between radiation injury and recurrence of brain metastasis, textural features extracted from CE-MRI had a diagnostic accuracy of 81% (sensitivity, 67%; specificity, 90%). FET PET textural features revealed a slightly higher diagnostic accuracy of 83% (sensitivity, 88%; specificity, 75%). However, the highest diagnostic accuracy was obtained when combining CE-MRI and FET PET features (accuracy, 89%; sensitivity, 85%; specificity, 96%).ConclusionsOur findings suggest that combined FET PET/CE-MRI radiomics using textural feature analysis offers a great potential to contribute significantly to the management of patients with brain metastases.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Combined FET PET/MRI radiomics differentiates radiation injury from recurrent brain metastasis

Lohmann

Köcher

Ceccon

et al. 2018

NeuroImage: Clinical

124

View full text Add to dashboard Cite

show abstract

“…GLCM is a gray-level spatial dependence matrix, consisting of values that show how often specific pairs of pixel value occur in a given spatial relationship in an image [12]. Recently the so-called large space conquest classification method was employed to investigate the textural parameters of brain MRI images, based on support vector machine , 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 or grading machine learning schemes [15]. The extraction of texture information can rely on several techniques, among which the Local Binary Pattern (LBP) method is getting more attention lately, as it provides promising results for a bunch of different applications [16,17].…”

Section: Introductionmentioning

confidence: 99%

“…Textural analysis could differentiate metastases from radiation necrosis after radiotherapy [15] In addition, there are at least two conceptually different texture analysis methods, the 2D and the 3D approaches. Mahmoud-Ghoneim et al showed that 3D method gives a better discrimination between necrosis and solid tumor as well as between edema and solid tumor [23].…”

Section: Introductionmentioning

confidence: 99%

2D and 3D texture analysis to differentiate brain metastases on MR images: proceed with caution

Béresová

Larroza

Arana

et al. 2017

Magn Reson Mater Phy

Self Cite

View full text Add to dashboard Cite

Object: To find structural differences between brain metastases of lung and breast cancer, computing their heterogeneity parameters by means of both 2D and 3D texture analysis (TA). Materials and Methods:Patients with fifty-eight brain metastases from breast (26) and lung cancer (32) were examined by MR imaging. Brain lesions were manually delineated by 2DROIs on the slices of contrast-enhanced T1-weighted (CET1) images, and local binary patterns (LBP) maps were created from each region. Histogram-based (minimum, maximum, mean, standard deviation and variance), and co-occurrence matrix-based (contrast, correlation, energy, entropy and homogeneity) 2D, weighted average of the 2D slices, and true 3D TA were obtained on the CET1 images and LBP maps.Results: For LBP maps and 2D TA contrast, correlation, energy and homogeneity were identified as statistically different heterogeneity parameters (SDHPs) between lung and breast metastasis. The weighted 3D TA identified entropy as an additional SDHP. Only two texture indexes (TI) were significantly different with true 3D TA: entropy and energy. All these TIs discriminated between the two tumor types significantly by ROC analysis. For the CET1 images there was no SDHP at all by 3D TA. Conclusion:Our results indicate that the used textural analysis methods may help with discriminating between brain metastases of different primary tumors.

show abstract

“…brain lesion texture is correlated with presence and type of cancer cells. A recent study23 employed Haralick and wavelet texture features to distinguish radiation necrosis from metastatic brain tumor recurrence with a reported AUC of 94%. However, we believe that the results, reported on a per-slice basis, may have been affected by the classifier being contaminated by slices from the same patient being used both within the training as well as testing sets during classification.…”

Section: Resultsmentioning

confidence: 99%

Co-occurrence of Local Anisotropic Gradient Orientations (CoLlAGe): A new radiomics descriptor

Prasanna

Tiwari

Madabhushi

2016

Sci Rep

115

View full text Add to dashboard Cite

In this paper, we introduce a new radiomic descriptor, Co-occurrence of Local Anisotropic Gradient Orientations (CoLlAGe) for capturing subtle differences between benign and pathologic phenotypes which may be visually indistinguishable on routine anatomic imaging. CoLlAGe seeks to capture and exploit local anisotropic differences in voxel-level gradient orientations to distinguish similar appearing phenotypes. CoLlAGe involves assigning every image voxel an entropy value associated with the co-occurrence matrix of gradient orientations computed around every voxel. The hypothesis behind CoLlAGe is that benign and pathologic phenotypes even though they may appear similar on anatomic imaging, will differ in their local entropy patterns, in turn reflecting subtle local differences in tissue microarchitecture. We demonstrate CoLlAGe’s utility in three clinically challenging classification problems: distinguishing (1) radiation necrosis, a benign yet confounding effect of radiation treatment, from recurrent tumors on T1-w MRI in 42 brain tumor patients, (2) different molecular sub-types of breast cancer on DCE-MRI in 65 studies and (3) non-small cell lung cancer (adenocarcinomas) from benign fungal infection (granulomas) on 120 non-contrast CT studies. For each of these classification problems, CoLlAGE in conjunction with a random forest classifier outperformed state of the art radiomic descriptors (Haralick, Gabor, Histogram of Gradient Orientations).

show abstract

Support vector machine classification of brain metastasis and radiation necrosis based on texture analysis in MRI

Abstract: High classification accuracy (AUC > 0.9) was obtained using texture features and a support vector machine classifier in an approach based on conventional MRI to differentiate between brain metastasis and radiation necrosis.

Cited by 94 publications

References 41 publications

Combined FET PET/MRI radiomics differentiates radiation injury from recurrent brain metastasis

Combined FET PET/MRI radiomics differentiates radiation injury from recurrent brain metastasis

2D and 3D texture analysis to differentiate brain metastases on MR images: proceed with caution

Co-occurrence of Local Anisotropic Gradient Orientations (CoLlAGe): A new radiomics descriptor

Contact Info

Product

Resources

About