Skewness and Kurtosis of Apparent Diffusion Coefficient in Human Brain Lesions to Distinguish Benign and Malignant Using MRI

Vijithananda, Sahan M.; Jayatilake, Mohan L.; Weerakoon, Bimali Sanjeevani; Wathsala, P. G. S.; Thevapriya, S.; Thasanky, S.; Kalupahana, Tharindu D.; Wijerathne, Thusitha K.

doi:10.1007/978-981-13-9184-2_17

Cited by 9 publications

(10 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Diffusion Weighted (DW) imaging is a form of magnetic resonance imaging (MRI) technique that is widely used in tumor identification and classification in modern clinical radiology practices [6,7]. This technology is based on measurements of random Brownian motion of water molecules within a voxel of a biological tissue [8][9][10]. The technique allows to visualize the net direction of diffusion of water molecules or collective flow of water molecules in a live tissue.…”

Section: Magnetic Resonance Imagingmentioning

confidence: 99%

“…The higher order statistic provides powerful tools in identifying problems in non linear systems [26]. However, skewness and kurtosis are the examples of third-order and fourth-order statistics, respectively' [8,27]. Here, skewness measures the asymmetry around the mean of probability distribution of a real valued random variable and the values for skewness can be zero (0), positive (+), negative (−) or undefined.…”

Section: Higher Order Momentsmentioning

confidence: 99%

See 1 more Smart Citation

Feature extraction from MRI ADC images for brain tumor classification using machine learning techniques

et al. 2022

Self Cite

View full text Add to dashboard Cite

Background Diffusion-weighted (DW) imaging is a well-recognized magnetic resonance imaging (MRI) technique that is being routinely used in brain examinations in modern clinical radiology practices. This study focuses on extracting demographic and texture features from MRI Apparent Diffusion Coefficient (ADC) images of human brain tumors, identifying the distribution patterns of each feature and applying Machine Learning (ML) techniques to differentiate malignant from benign brain tumors. Methods This prospective study was carried out using 1599 labeled MRI brain ADC image slices, 995 malignant, 604 benign from 195 patients who were radiologically diagnosed and histopathologically confirmed as brain tumor patients. The demographics, mean pixel values, skewness, kurtosis, features of Grey Level Co-occurrence Matrix (GLCM), mean, variance, energy, entropy, contrast, homogeneity, correlation, prominence and shade, were extracted from MRI ADC images of each patient. At the feature selection phase, the validity of the extracted features were measured using ANOVA f-test. Then, these features were used as input to several Machine Learning classification algorithms and the respective models were assessed. Results According to the results of ANOVA f-test feature selection process, two attributes: skewness (3.34) and GLCM homogeneity (3.45) scored the lowest ANOVA f-test scores. Therefore, both features were excluded in continuation of the experiment. From the different tested ML algorithms, the Random Forest classifier was chosen to build the final ML model, since it presented the highest accuracy. The final model was able to predict malignant and benign neoplasms with an 90.41% accuracy after the hyper parameter tuning process. Conclusions This study concludes that the above mentioned features (except skewness and GLCM homogeneity) are informative to identify and differentiate malignant from benign brain tumors. Moreover, they enable the development of a high-performance ML model that has the ability to assist in the decision-making steps of brain tumor diagnosis process, prior to attempting invasive diagnostic procedures, such as brain biopsies.

show abstract

Section: Magnetic Resonance Imagingmentioning

confidence: 99%

Section: Higher Order Momentsmentioning

confidence: 99%

Feature extraction from MRI ADC images for brain tumor classification using machine learning techniques

et al. 2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…Using the hidden characteristics of image textures of higher order moments of ADC (skewness and kurtosis) [8,21], GLCM statistical features [1, 34,35] and patients' demographic data, this work proposes a Machine Learning classification model that can be used to differentiate benign and malignant brain tumors.…”

Section: Objectivesmentioning

confidence: 99%

“…MATLAB 2019 Simulink software was used in all the image processing steps and Python 3.7 in all the feature extraction and analysis processes [8]. The GLCM matrices of each 2D parametric map of ADC brain tumor were derived according to the Equation 4(see Appendix).…”

Section: Feature Extractionmentioning

confidence: 99%

Feature Extraction from MRI ADC Images for Brain Tumor Classification Using Machine Learning Techniques

Vijithananda

Jayatilake

Hewavithana

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Background: Diffusion-weighted (DW) imaging is a well-recognized magnetic resonance imaging (MRI) technique that is being routinely used in brain examinations in modern clinical radiology practices. This study focuses on extracting demographic and texture features from MRI Apparent Diffusion Coefficient (ADC) images of human brain tumors, identifying the distribution patterns of each feature and applying Machine Learning (ML) techniques to differentiate malignant from benign brain tumors.Methods: This prospective study was carried out using 1599 labeled MRI brain ADC image slices, 995 malignant, 604 benign from 195 patients who were radiologically diagnosed and histopathologically confirmed as brain tumor patients.The demographics, mean pixel values, skewness, kurtosis, features of Grey Level Co-occurrence Matrix (GLCM), mean, variance, energy, entropy, contrast, homogeneity, correlation, prominence and shade, were extracted from MRI ADC images of each patient.At the feature selection phase, the validity of the extracted features were measured using ANOVA f-test. Then, these features were used as input to several Machine Learning classification algorithms and the respective models were assessed.Results: According to the results of ANOVA f-test feature selection process, two attributes: skewness (3.34) and GLCM homogeneity (3.45) scored the lowest ANOVA f-test scores. Therefore both features were excluded in continuation of the experiment. From the different tested ML algorithms, the Random Forest classifier was chosen to build the final ML model since it presented the highest accuracy. The final model was able to predict malignant and benign neoplasms with an 90.41% accuracy after the hyper parameter tuning process.Conclusion: This study concludes that the above mentioned features (except skewness and GLCM homogeneity) are informative to identify and differentiate malignant from benign brain tumors. Moreover, they enable the development of a high-performance ML model that has the ability to assist in the decision-making steps of brain tumor diagnosis process, prior to attempting invasive diagnostic procedures such as brain biopsies.

show abstract

“…However, lower and higher order moments (see Equation 3); mean pixel value (n = 1), skewness (n = 3), kurtosis (n = 4) and texture features of GLCM such as mean, variance, energy, entropy, contrast, homogeneity, correlation, prominence and shade values were studied in this pattern recognition process; The GLCM features were extracted according to Equations 5,6,7,8,9,10,11,12 and Equation 13respectively [24] [25] [39]. Here P i,j be the co-occurrence matrix, N be the number of grey levels in the image, µ be the mean of P i,j , µ i be the mean of row i, µ j be the mean value of column j, σ i be the standard deviation of row i and σ j be the standard deviation of column j.…”

Section: Feature Extractionmentioning

confidence: 99%

Feature Extraction from MRI ADC Images for Brain Tumor Classification Using Machine Learning Techniques

Vijithananda

Jayatilake

Hewavithana

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Background: Diffusion-weighted (DW) imaging is a well-recognized magnetic resonance imaging (MRI) technique that is being routinely used in brain examinations in modern clinical radiology practices. This study focuses on extracting demographic and texture features from MRI Apparent Diffusion Coefficient (ADC) images of human brain tumors, identifying the distribution patterns of each feature and applying Machine Learning (ML) techniques to differentiate malignant from benign brain tumors. Methods: This prospective study was carried out using 1599 labeled MRI brain ADC image slices, 995 malignant, 604 benign from 195 patients who were radiologically diagnosed and histopathologically confirmed as brain tumor patients. The demographics, mean pixel values, skewness, kurtosis, features of Grey Level Co-occurrence Matrix (GLCM), mean, variance, energy, entropy, contrast, homogeneity, correlation, prominence and shade, were extracted from MRI ADC images of each patient. At the feature selection phase, the validity of the extracted features were measured using ANOVA f-test. Then, these features were used as input to several Machine Learning classification algorithms and the respective models were assessed. Results: According to the results of ANOVA f-test feature selection process, two attributes: skewness (3.34) and GLCM homogeneity (3.45) scored the lowest ANOVA f-test scores. Therefore both features were excluded in continuation of the experiment. From the different tested ML algorithms, the Random Forest classifier was chosen to build the final ML model since it presented the highest accuracy. The final model was able to predict malignant and benign neoplasms with an 90.41% accuracy after the hyper parameter tuning process. Conclusion: This study concludes that the above mentioned features (except skewness and GLCM homogeneity) are informative to identify and differentiate malignant from benign brain tumors. Moreover, they enable the development of a high-performance ML model that has the ability to assist in the decision-making steps of brain tumor diagnosis process, prior to attempting invasive diagnostic procedures such as brain biopsies.

show abstract

Skewness and Kurtosis of Apparent Diffusion Coefficient in Human Brain Lesions to Distinguish Benign and Malignant Using MRI

Cited by 9 publications

References 21 publications

Feature extraction from MRI ADC images for brain tumor classification using machine learning techniques

Feature extraction from MRI ADC images for brain tumor classification using machine learning techniques

Feature Extraction from MRI ADC Images for Brain Tumor Classification Using Machine Learning Techniques

Feature Extraction from MRI ADC Images for Brain Tumor Classification Using Machine Learning Techniques

Contact Info

Product

Resources

About