Diagnostic Value of Machine Learning-Based Quantitative Texture Analysis in Differentiating Benign and Malignant Thyroid Nodules

Aim. The aim of this study is to evaluate the diagnostic value of machine learning- (ML-) based quantitative texture analysis in the differentiation of benign and malignant thyroid nodules. Materials and methods. A sum of 306 quantitative textural features of 235 thyroid nodules (102 malignant, 43.4%; 133 benign, 56.4%) of a total of 198 patients were investigated using the random forest ML classifier. Feature selection and dimension reduction were conducted using reproducibility testing and a wrapper method. The diagnostic accuracy, sensitivity, specificity, and area under curve (AUC) of the proposed method were compared with the histopathological or cytopathological findings as reference methods. Results. Of the 306 initial texture features, 284 (92.2%) showed good reproducibility (intraclass correlation ≥0.80). The random forest classifier accurately identified 87 out of 102 malignant thyroid nodules and 117 out of 133 benign thyroid nodules, which is a diagnostic sensitivity of 85.2%, specificity of 87.9%, and accuracy of 86.8%. The AUC of the model was 0.92. Conclusions. Quantitative textural analysis of thyroid nodules using ML classification can accurately discriminate benign and malignant thyroid nodules. Our findings should be validated by multicenter prospective studies using completely independent external data.

Download Full-text

Machine learning-based unenhanced CT texture analysis for predicting BAP1 mutation status of clear cell renal cell carcinomas

Acta Radiologica ◽

10.1177/0284185119881742 ◽

2019 ◽

Vol 61 (6) ◽

pp. 856-864 ◽

Cited By ~ 3

Author(s):

Burak Kocak ◽

Emine Sebnem Durmaz ◽

Ozlem Korkmaz Kaya ◽

Ozgur Kilickesmez

Keyword(s):

Machine Learning ◽

Random Forest ◽

Renal Cell ◽

Clear Cell ◽

Texture Features ◽

Observer Agreement ◽

Mutation Status ◽

Unenhanced Ct ◽

Ct Texture Analysis ◽

Sensitivity Specificity

Background BRCA1-associated protein 1 (BAP1) mutation is an unfavorable factor for overall survival in patients with clear cell renal cell carcinoma (ccRCC). Radiomics literature about BAP1 mutation lacks papers that consider the reliability of texture features in their workflow. Purpose Using texture features with a high inter-observer agreement, we aimed to develop and internally validate a machine learning-based radiomic model for predicting the BAP1 mutation status of ccRCCs. Material and Methods For this retrospective study, 65 ccRCCs were included from a public database. Texture features were extracted from unenhanced computed tomography (CT) images, using two-dimensional manual segmentation. Dimension reduction was done in three steps: (i) inter-observer agreement analysis; (ii) collinearity analysis; and (iii) feature selection. The machine learning classifier was random forest. The model was validated using 10-fold nested cross-validation. The reference standard was the BAP1 mutation status. Results Out of 744 features, 468 had an excellent inter-observer agreement. After the collinearity analysis, the number of features decreased to 17. Finally, the wrapper-based algorithm selected six features. Using selected features, the random forest correctly classified 84.6% of the labelled slices regarding BAP1 mutation status with an area under the receiver operating characteristic curve of 0.897. For predicting ccRCCs with BAP1 mutation, the sensitivity, specificity, and precision were 90.4%, 78.8%, and 81%, respectively. For predicting ccRCCs without BAP1 mutation, the sensitivity, specificity, and precision were 78.8%, 90.4%, and 89.1%, respectively. Conclusion Machine learning-based unenhanced CT texture analysis might be a potential method for predicting the BAP1 mutation status of ccRCCs.

Download Full-text

FORMING THE STACK OF TEXTURE FEATURES FOR LIVER ULTRASOUND IMAGES CLASSIFICATION

Herald of Advanced Information Technology ◽

10.15276/hait.04.2020.3 ◽

2020 ◽

Vol 3 (4) ◽

pp. 240-251

Author(s):

Dmitro Yuriiovych Hrishko ◽

Ievgen Arnoldovich Nastenko ◽

Maksym Oleksandrovych Honcharuk ◽

Volodymyr Anatoliyovich Pavlov

Keyword(s):

Random Forest ◽

Texture Analysis ◽

Binary Classification ◽

Texture Features ◽

Normal Liver ◽

Group Method ◽

Ultrasound Images ◽

Liver Ultrasound ◽

Random Forest Models ◽

Occurrence Matrix

This article discusses the use of texture analysis methods to obtain informative features that describe the texture of liver ultrasound images. In total, 317 liver ultrasound images were analyzed, which were provided by the Institute of Nuclear Medicine and Radiation Diagnostics of NAMS of Ukraine. The images were taken by three different sensors (convex, linear, and linear sensor in increased signal level mode). Both images of patients with a normal liver condition and patients with specific liver disease (there were diseases such as: autoimmune hepatitis, Wilson's disease, hepatitis B and C, steatosis, and cirrhosis) were present in the database. Texture analysis was used for “Feature Construction”, which resulted in more than a hundred different informative features that made up a common stack. Among them, there are such features as: three authors’ patented features derived from the grey level co-occurrence matrix; features, obtained with the help of spatial sweep method (working by the principle of group method of data handling), which was applied to ultrasound images; statistical features, calculated on the images, brought to one scale with the help of differential horizontal and vertical matrices, which are proposed by the authors; greyscale pairs ensembles (found using the genetic algorithm), which identify liver pathology on images, transformed with the help of horizontal and vertical differentiations, in the best possible way. The resulting trait stack was used to solve the problem of binary classification (“norma-pathology”) of ultrasound liver images. A Machine Learning method, namely “Random Forest”, was used for this purpose. Before the classification, in order to obtain objective results, the total samples were divided into training (70 %), testing (20 %), and examining (10 %). The result was the best three Random Forest models separately for each sensor, which gave the following recognition rates: 93.4 % for the convex sensor, 92.9 % for the linear sensor, and 92 % for the reinforced linear sensor

Download Full-text

Quantitative Assessment of Thyroid Nodules Using Dual-Energy Computed Tomography: Iodine Concentration Measurement and Multiparametric Texture Analysis for Differentiating between Malignant and Benign Lesions

International Journal of Endocrinology ◽

10.1155/2020/5484671 ◽

2020 ◽

Vol 2020 ◽

pp. 1-8 ◽

Cited By ~ 2

Author(s):

Hayato Tomita ◽

Hirofumi Kuno ◽

Kotaro Sekiya ◽

Katharina Otani ◽

Osamu Sakai ◽

...

Keyword(s):

Computed Tomography ◽

Texture Analysis ◽

Thyroid Nodules ◽

Texture Features ◽

Iodine Content ◽

Iodine Concentration ◽

Dual Energy ◽

Concentration Measurement ◽

Cross Sectional ◽

Dual Energy Computed Tomography

Background and Objectives. Thyroid nodules are increasingly being detected during cross-sectional imaging of the neck and chest. The purpose of this study is to investigate the efficacy of dual-energy computed tomography (DECT) using iodine concentration measurement and multiparametric texture analysis of monochromatic images for differentiating between benign and malignant thyroid nodules. Materials and Methods. This retrospective study included 34 consecutive patients who presented with thyroid nodules and underwent noncontrast DECT between 2015 and 2016. Manual segmentation of each thyroid nodule by monochromatic imaging (40, 60, and 80 keV) was performed, and an in-house developed MATLAB-based texture analysis program was used to extract 41 textures. Iodine material decomposition and CT attenuation slopes were also measured. Histopathologic findings of ultrasound-guided biopsies over a follow-up period of at least one year were used as reference standards. Basic descriptive statistics and areas under receiver operating characteristic curves (AUCs) were evaluated. Results. The 34 nodules comprised 14 benign nodules and 20 malignant nodules. Iodine content and Hounsfield unit curve slopes did not differ significantly between benign and malignant thyroid nodules (P=0.480–0.670). However, significant differences in the texture features of monochromatic images were observed between benign and malignant nodules: histogram mean and median, co-occurrence matrix contrast, gray-level gradient matrix (GLGM) skewness, and mean gradients and variance of gradients for GLGM at 80 keV (P=0.014–0.044). The highest AUC was 0.77, for the histogram mean and median of images acquired at 80 keV. Conclusions. Texture features extracted from monochromatic images using DECT, specifically acquired at high keV, may be a promising diagnostic approach for thyroid nodules. A further large study for incidental thyroid nodules using DECT texture analysis is required to validate our results.

Download Full-text

Diagnostic performance of machine learning applied to texture analysis-derived features for breast lesion characterisation at automated breast ultrasound: a pilot study

European Radiology Experimental ◽

10.1186/s41747-019-0121-6 ◽

2019 ◽

Vol 3 (1) ◽

Cited By ~ 2

Author(s):

Magda Marcon ◽

Alexander Ciritsis ◽

Cristina Rossi ◽

Anton S. Becker ◽

Nicole Berger ◽

...

Keyword(s):

Machine Learning ◽

Pilot Study ◽

Texture Analysis ◽

Area Under The Curve ◽

Texture Features ◽

Breast Ultrasound ◽

Support Vector ◽

Maximum Area ◽

Svm Algorithm ◽

Automated Breast Ultrasound

Abstract Background Our aims were to determine if features derived from texture analysis (TA) can distinguish normal, benign, and malignant tissue on automated breast ultrasound (ABUS); to evaluate whether machine learning (ML) applied to TA can categorise ABUS findings; and to compare ML to the analysis of single texture features for lesion classification. Methods This ethically approved retrospective pilot study included 54 women with benign (n = 38) and malignant (n = 32) solid breast lesions who underwent ABUS. After manual region of interest placement along the lesions’ margin as well as the surrounding fat and glandular breast tissue, 47 texture features (TFs) were calculated for each category. Statistical analysis (ANOVA) and a support vector machine (SVM) algorithm were applied to the texture feature to evaluate the accuracy in distinguishing (i) lesions versus normal tissue and (ii) benign versus malignant lesions. Results Skewness and kurtosis were the only TF significantly different among all the four categories (p < 0.000001). In subsets (i) and (ii), a maximum area under the curve of 0.86 (95% confidence interval [CI] 0.82–0.88) for energy and 0.86 (95% CI 0.82–0.89) for entropy were obtained. Using the SVM algorithm, a maximum area under the curve of 0.98 for both subsets was obtained with a maximum accuracy of 94.4% in subset (i) and 90.7% in subset (ii). Conclusions TA in combination with ML might represent a useful diagnostic tool in the evaluation of breast imaging findings in ABUS. Applying ML techniques to TFs might be superior compared to the analysis of single TF.

Download Full-text

MRI-Based Machine Learning in Differentiation Between Benign and Malignant Breast Lesions

Frontiers in Oncology ◽

10.3389/fonc.2021.552634 ◽

2021 ◽

Vol 11 ◽

Author(s):

Yanjie Zhao ◽

Rong Chen ◽

Ting Zhang ◽

Chaoyue Chen ◽

Muhetaer Muhelisa ◽

...

Keyword(s):

Machine Learning ◽

Random Forest ◽

Decision Tree ◽

Texture Analysis ◽

Training Group ◽

Gradient Boosting ◽

Breast Lesions ◽

Extreme Gradient Boosting ◽

Malignant Breast ◽

Benign Breast Lesions

BackgroundDifferential diagnosis between benign and malignant breast lesions is of crucial importance relating to follow-up treatment. Recent development in texture analysis and machine learning may lead to a new solution to this problem.MethodThis current study enrolled a total number of 265 patients (benign breast lesions:malignant breast lesions = 71:194) diagnosed in our hospital and received magnetic resonance imaging between January 2014 and August 2017. Patients were randomly divided into the training group and validation group (4:1), and two radiologists extracted their texture features from the contrast-enhanced T1-weighted images. We performed five different feature selection methods including Distance correlation, Gradient Boosting Decision Tree (GBDT), least absolute shrinkage and selection operator (LASSO), random forest (RF), eXtreme gradient boosting (Xgboost) and five independent classification models were built based on Linear discriminant analysis (LDA) algorithm.ResultsAll five models showed promising results to discriminate malignant breast lesions from benign breast lesions, and the areas under the curve (AUCs) of receiver operating characteristic (ROC) were all above 0.830 in both training and validation groups. The model with a better discriminating ability was the combination of LDA + gradient boosting decision tree (GBDT). The sensitivity, specificity, AUC, and accuracy in the training group were 0.814, 0.883, 0.922, and 0.868, respectively; LDA + random forest (RF) also suggests promising results with the AUC of 0.906 in the training group.ConclusionThe evidence of this study, while preliminary, suggested that a combination of MRI texture analysis and LDA algorithm could discriminate benign breast lesions from malignant breast lesions. Further multicenter researches in this field would be of great help in the validation of the result.

Download Full-text

A machine learning framework for the evaluation of myocardial rotation in patients with noncompaction cardiomyopathy

PLoS ONE ◽

10.1371/journal.pone.0260195 ◽

2021 ◽

Vol 16 (11) ◽

pp. e0260195

Author(s):

Marcelo Dantas Tavares de Melo ◽

Jose de Arimatéia Batista Araujo-Filho ◽

José Raimundo Barbosa ◽

Camila Rocon ◽

Carlos Danilo Miranda Regis ◽

...

Keyword(s):

Machine Learning ◽

Random Forest ◽

Ejection Fraction ◽

Area Under The Curve ◽

Random Forest Algorithm ◽

Noncompaction Cardiomyopathy ◽

2D Echocardiography ◽

Specific Strain ◽

Lv Ejection Fraction ◽

Sensitivity Specificity

Aims Noncompaction cardiomyopathy (NCC) is considered a genetic cardiomyopathy with unknown pathophysiological mechanisms. We propose to evaluate echocardiographic predictors for rigid body rotation (RBR) in NCC using a machine learning (ML) based model. Methods and results Forty-nine outpatients with NCC diagnosis by echocardiography and magnetic resonance imaging (21 men, 42.8±14.8 years) were included. A comprehensive echocardiogram was performed. The layer-specific strain was analyzed from the apical two-, three, four-chamber views, short axis, and focused right ventricle views using 2D echocardiography (2DE) software. RBR was present in 44.9% of patients, and this group presented increased LV mass indexed (118±43.4 vs. 94.1±27.1g/m2, P = 0.034), LV end-diastolic and end-systolic volumes (P< 0.001), E/e’ (12.2±8.68 vs. 7.69±3.13, P = 0.034), and decreased LV ejection fraction (40.7±8.71 vs. 58.9±8.76%, P < 0.001) when compared to patients without RBR. Also, patients with RBR presented a significant decrease of global longitudinal, radial, and circumferential strain. When ML model based on a random forest algorithm and a neural network model was applied, it found that twist, NC/C, torsion, LV ejection fraction, and diastolic dysfunction are the strongest predictors to RBR with accuracy, sensitivity, specificity, area under the curve of 0.93, 0.99, 0.80, and 0.88, respectively. Conclusion In this study, a random forest algorithm was capable of selecting the best echocardiographic predictors to RBR pattern in NCC patients, which was consistent with worse systolic, diastolic, and myocardium deformation indices. Prospective studies are warranted to evaluate the role of this tool for NCC risk stratification.

Download Full-text

Comparative Analysis of Diagnostic Value for Shear Wave Elastography and Real-Time Elastographic Imaging for Thyroid Nodules

Journal of Medical Imaging and Health Informatics ◽

10.1166/jmihi.2019.2594 ◽

2019 ◽

Vol 9 (2) ◽

pp. 334-338

Author(s):

Qing Yang ◽

Wenhong Zhou ◽

Jiyu Li ◽

Guojun Wu ◽

Feng Ding ◽

...

Keyword(s):

Differential Diagnosis ◽

Positive Predictive Value ◽

Negative Predictive Value ◽

Shear Wave ◽

Predictive Value ◽

Thyroid Nodules ◽

Shear Wave Elastography ◽

Roc Curves ◽

Diagnostic Value ◽

Sensitivity Specificity

Objective: To compare the diagnostic value of shear wave elastography (SWE) and real-time elastography (RTE) in the diagnosis of benign and malignant thyroid nodules. Methods: A total of 34 patients who ever received thyroidectomy in our hospital from January 2016 to January 2018 were identified. Meanwhile, all the patients received SWE and RTE before surgery, and all the diagnoses were confirmed by pathological examinations. With respect to SWE technique, the Subject Operating Characteristics (ROC) curves were drawn, in order to obtain the optimal threshold and then make differential diagnoses of benign and malignant thyroid nodules. In terms of RTE, the Rago 5 scoring method was utilized to make differential diagnoses of benign and malignant thyroid nodules. Besides, the pathological examinations after surgery could be considered as the golden standard. At last, the sensitivity, specificity, accuracy, positive predictive value, and negative predictive value of SWE and RTE were calculated, respectively. Results: A total of 51 thyroid nodules were identified, and 41 nodules were benign, 10 nodules were malignant. On the basis of ROC curves, with respect to SWE, the best threshold for differential diagnosis of benign and malignant thyroid nodules is 38.3 kPa. The sensitivity, specificity, accuracy, positive predictive value, and negative predictive value of SWE were 72.7% (8/11), 85% (34/40), 82.4% (42/51), 68.4% (13/19), and 87.5% (35/40), respectively. And the diagnostic indicators of RTE were 81.8% (9/11), 87.5% (35/40), 84.3% (43/51), 73.7% (14/19), and 90.0% (36/40). The sensitivity of quasi-static elastography in differential diagnosis of benign and malignant thyroid nodules with diameter ≤1 cm was 87.5% (7/8), and the sensitivity of SWE was 50.0% (5/10). In addition, the accuracy of SWE in differential diagnosis of benign and malignant thyroid nodules with diameter ≥3 cm was 100% (6/6), and the accuracy of RTE for this kind of thyroid nodules was 66.7% (4/6). Conclusion: Both SWE and RTE technology have good application value in differential diagnosis of benign and malignant thyroid nodules. But, SWE is preferable when making diagnosis of benign and malignant thyroid nodules with diameter ≥3 cm, and RTE was superior in detecting benign and malignant thyroid nodules with diameter ≤1 cm.

Download Full-text

An Efficient Ensemble Classifier for Heart Disease Diagnosis and early Prediction

International Journal of Recent Technology and Engineering - 2 ◽

10.35940/ijrte.d4824.119420 ◽

2020 ◽

Vol 9 (4) ◽

pp. 109-114

Keyword(s):

Machine Learning ◽

Cardiovascular Disease ◽

Heart Disease ◽

Random Forest ◽

False Positive Rate ◽

Ensemble Classifier ◽

Disease Pattern ◽

Positive Rate ◽

Sensitivity Specificity ◽

Pattern Prediction

Heart Disease is one of the most significant causes of mortality in the world today. Prediction and Diagnosis of Cardiovascular disease is considered as one of the major challenges in the Medical Field especially for Cardiologists. Artificial Intelligence and Machine learning (ML) was popularly employed for pattern prediction and it was noticed that these Intelligent Mechanisms were used in Medical Feld for better Heart Disease Pattern Prediction. Thus more researchers were focusing Machine Learning based Data Mining Classifiers for Heart Disease Pattern Prediction and Diagnosis in the healthcare Industry especially for Cardiologists. This research work identified the recently proposed Hybrid Random Forest with a Linear Model (HRFLM) Classifier for improving the classification accuracy for the cardiovascular disease patterns prediction well in advance and Diagnosis as well. However, it was noticed that for improving the performances better in terms of Accuracy, Sensitivity, Specificity, Precision, FScore and False Positive Rate FPR, needed an efficient classifier. Thus this work developed and implemented an efficient Classifier ensemble Nu-SVC Classifier and Weighted Random Forest Classifier. From the experimental results, it was noticed that the proposed Ensemble Classifier performs better as compared with that of existing Hybrid Classifier in terms of in terms of Accuracy, Sensitivity, Specificity, Precision, FScore and False Positive Rate FPR

Download Full-text

The value of S-Detect in improving the diagnostic performance of radiologists for the differential diagnosis of thyroid nodules

Medical Ultrasonography ◽

10.11152/mu-2501 ◽

2020 ◽

Vol 22 (4) ◽

pp. 415

Author(s):

Qi Wei ◽

Shu-E Zeng ◽

Li-Ping Wang ◽

Yu-Jing Yan ◽

Ting Wang ◽

...

Keyword(s):

Diagnostic Accuracy ◽

Predictive Value ◽

Diagnostic Performance ◽

Thyroid Nodules ◽

Characteristic Curve ◽

Diagnostic Value ◽

Ultrasound Images ◽

Experienced Radiologist ◽

Before And After ◽

Sensitivity Specificity

Aims: To compare the diagnostic value of S-Detect (a computer aided diagnosis system using deep learning) in differentiating thyroid nodules in radiologists with different experience and to assess if S-Detect can improve the diagnostic performance of radiologists.Materials and methods: Between February 2018 and October 2019, 204 thyroid nodules in 181 patients were included. An experienced radiologist performed ultrasound for thyroid nodules and obtained the result of S-Detect. Four radiologists with different experience on thyroid ultrasound (Radiologist 1, 2, 3, 4 with 1, 4, 9, 20 years, respectively) analyzed the conventional ultrasound images of each thyroid nodule and made a diagnosis of “benign” or “malignant” based on the TI-RADS category. After referring to S-Detect results, they re-evaluated the diagnoses. The diagnostic performance of radiologists was analyzed before and after referring to the results of S-Detect.Results: The accuracy, sensitivity, specificity, positive predictive value and negative predictive value of S-Detect were 77.0, 91.3, 65.2, 68.3 and 90.1%, respectively. In comparison with the less experienced radiologists (radiologist 1 and 2), S-Detect had a higher area under receiver operating characteristic curve (AUC), accuracy and specificity (p <0.05). In comparison with the most experienced radiologist, the diagnostic accuracy and AUC were lower (p<0.05). In the less experienced radiologists, the diagnostic accuracy, specificity and AUC were significantly improved when combined with S-Detect (p<0.05), but not for experienced radiologists (radiologist 3 and 4) (p>0.05).Conclusions: S-Detect may become an additional diagnostic method for the diagnosis of thyroid nodules and improve the diagnostic performance of less experienced radiologists.

Download Full-text

Prediction of pituitary adenoma surgical consistency: radiomic data mining and machine learning on T2-weighted MRI

Neuroradiology ◽

10.1007/s00234-020-02502-z ◽

2020 ◽

Vol 62 (12) ◽

pp. 1649-1656 ◽

Cited By ~ 2

Author(s):

Renato Cuocolo ◽

Lorenzo Ugga ◽

Domenico Solari ◽

Sergio Corvino ◽

Alessandra D’Amico ◽

...

Keyword(s):

Machine Learning ◽

Texture Analysis ◽

Preoperative Evaluation ◽

Texture Features ◽

Recursive Feature Elimination ◽

Pituitary Macroadenoma ◽

Mr Images ◽

Endoscopic Endonasal ◽

Tissue Characteristics ◽

Fold Cross Validation

Abstract Purpose Pituitary macroadenoma consistency can influence the ease of lesion removal during surgery, especially when using a transsphenoidal approach. Unfortunately, it is not assessable on standard qualitative MRI. Radiomic texture analysis could help in extracting mineable quantitative tissue characteristics. We aimed to assess the accuracy of texture analysis combined with machine learning in the preoperative evaluation of pituitary macroadenoma consistency in patients undergoing endoscopic endonasal surgery. Methods Data of 89 patients (68 soft and 21 fibrous macroadenomas) who underwent MRI and transsphenoidal surgery at our institution were retrospectively reviewed. After manual segmentation, radiomic texture features were extracted from original and filtered MR images. Feature stability analysis and a multistep feature selection were performed. After oversampling to balance the classes, 80% of the data was used for hyperparameter tuning via stratified 5-fold cross-validation, while a 20% hold-out set was employed for its final testing, using an Extra Trees ensemble meta-algorithm. The reference standard was based on surgical findings. Results A total of 1118 texture features were extracted, of which 741 were stable. After removal of low variance (n = 4) and highly intercorrelated (n = 625) parameters, recursive feature elimination identified a subset of 14 features. After hyperparameter tuning, the Extra Trees classifier obtained an accuracy of 93%, sensitivity of 100%, and specificity of 87%. The area under the receiver operating characteristic and precision-recall curves was 0.99. Conclusion Preoperative T2-weighted MRI texture analysis and machine learning could predict pituitary macroadenoma consistency.

Download Full-text