scholarly journals Radiomics-based machine learning analysis and characterization of breast lesions with multiparametric diffusion-weighted MR

2021 ◽  
Vol 19 (1) ◽  
Author(s):  
Kun Sun ◽  
Zhicheng Jiao ◽  
Hong Zhu ◽  
Weimin Chai ◽  
Xu Yan ◽  
...  

Abstract Background This study aimed to evaluate the utility of radiomics-based machine learning analysis with multiparametric DWI and to compare the diagnostic performance of radiomics features and mean diffusion metrics in the characterization of breast lesions. Methods This retrospective study included 542 lesions from February 2018 to November 2018. One hundred radiomics features were computed from mono-exponential (ME), biexponential (BE), stretched exponential (SE), and diffusion-kurtosis imaging (DKI). Radiomics-based analysis was performed by comparing four classifiers, including random forest (RF), principal component analysis (PCA), L1 regularization (L1R), and support vector machine (SVM). These four classifiers were trained on a training set with 271 patients via ten-fold cross-validation and tested on an independent testing set with 271 patients. The diagnostic performance of the mean diffusion metrics of ME (mADCall b, mADC0–1000), BE (mD, mD*, mf), SE (mDDC, mα), and DKI (mK, mD) were also calculated for comparison. The area under the receiver operating characteristic curve (AUC) was used to compare the diagnostic performance. Results RF attained higher AUCs than L1R, PCA and SVM. The AUCs of radiomics features for the differential diagnosis of breast lesions ranged from 0.80 (BE_D*) to 0.85 (BE_D). The AUCs of the mean diffusion metrics ranged from 0.54 (BE_mf) to 0.79 (ME_mADC0–1000). There were significant differences in the AUCs between the mean values of all diffusion metrics and radiomics features of AUCs (all P < 0.001) for the differentiation of benign and malignant breast lesions. Of the radiomics features computed, the most important sequence was BE_D (AUC: 0.85), and the most important feature was FO-10 percentile (Feature Importance: 0.04). Conclusions The radiomics-based analysis of multiparametric DWI by RF enables better differentiation of benign and malignant breast lesions than the mean diffusion metrics.

2021 ◽  
Vol 11 ◽  
Author(s):  
Simin Wang ◽  
Ning Mao ◽  
Shaofeng Duan ◽  
Qin Li ◽  
Ruimin Li ◽  
...  

Objective: A limited number of studies have focused on the radiomic analysis of contrast-enhanced mammography (CEM). We aimed to construct several radiomics-based models of CEM for classifying benign and malignant breast lesions.Materials and Methods: The retrospective, double-center study included women who underwent CEM between November 2013 and February 2020. Radiomic analysis was performed using high-energy (HE), low-energy (LE), and dual-energy subtraction (DES) images from CEM. Datasets were randomly divided into the training and testing sets at a ratio of 7:3. The maximum relevance minimum redundancy (mRMR) method and least absolute shrinkage and selection operator (LASSO) logistic regression were used to select the radiomic features and construct the best classification models. The performances of the models were assessed by the area under the receiver operating characteristic curve (AUC) with a 95% confidence interval (CI). Leave-group-out cross-validation (LGOCV) for 100 rounds was performed to obtain the mean AUCs, which were compared by the Wilcoxon rank-sum test and the Kruskal–Wallis rank-sum test.Results: A total of 192 women with 226 breast lesions (101 benign; 125 malignant) were enrolled. The median age was 48 years (range, 22–70 years). For the classification of breast lesions, the AUCs of the best models were 0.931 (95% CI: 0.873–0.989) for HE, 0.897 (95% CI: 0.807–0.981) for LE, 0.882 (95% CI: 0.825–0.987) for DES images and 0.960 (95% CI: 0.910–0.998) for all of the CEM images in the testing set. According to LGOCV, the models constructed with the HE images and all of the CEM images showed the highest mean AUCs for the training (0.931 and 0.938, respectively; P &lt; 0.05 for both) and testing sets (0.892 and 0.889, respectively; P = 0.55 for both), which were significantly higher than those of the two models constructed with the LE and DES images in the training (0.912 and 0.899, respectively; all P &lt; 0.05) and testing sets (0.866 and 0.862, respectively; all P &lt; 0.05).Conclusions: Radiomic analysis of CEM images was valuable for classifying benign and malignant breast lesions. The use of HE images or all three types of CEM images can achieve the best performance.


2020 ◽  
pp. 028418512096142
Author(s):  
Yasemin Altıntas ◽  
Mehmet Bayrak ◽  
Ömer Alabaz ◽  
Medih Celiktas

Background Ultrasound (US) elastography has become a routine instrument in ultrasonographic diagnosis that measures the consistency and stiffness of tissues. Purpose To distinguish benign and malignant breast masses using a single US system by comparing the diagnostic parameters of three kinds of breast elastography simultaneously added to B-mode ultrasonography. Material and Methods A total of 163 breast lesions in 159 consecutive women who underwent US-guided core needle biopsy were included in this prospective study. Before the biopsy, the lesions were examined with B-mode ultrasonography and strain (SE), shear wave (SWE), and point shear wave (STQ) elastography. The strain ratio was computed and the Tsukuba score determined. The mean elasticity values using SWE and STQ were computed and converted to Young’s modulus E (kPa). Results All SE, SWE, and STQ parameters showed similar diagnostic performance. The SE score, SE ratio, SWEmean, SWEmax, STQmean, and STQmax yielded higher specificity than B-mode US alone to differentiate benign and malignant masses. The sensitivity of B-mode US, SWE, and STQ was slightly higher than that of the SE score and SE ratio. The SE score, SE ratio, SWEmean, SWEmax, STQmean, and STQmax had significantly higher positive predictive value and diagnostic accuracy than B-mode US alone. The area under the curve for each of these elastography methods in differentiating benign and malignant breast lesions was 0.93, 0.93, 0.98, 0.97, 0.98, and 0.96, respectively; P<0.001 for all measurements. Conclusion SE (ratio and score), SWE, and STQ had higher diagnostic performance individually than B-mode US alone in distinguishing between malignant and benign breast masses.


Author(s):  
Roaa M. A. Shehata ◽  
Mostafa A. M. El-Sharkawy ◽  
Omar M. Mahmoud ◽  
Hosam M. Kamel

Abstract Background Breast cancer is the most common life-threatening cancer in women worldwide. A high number of women are going through biopsy procedures for characterization of breast masses every day and yet 75% of the pathological results prove these masses to be benign. Ultrasound (US) elastography is a non-invasive technique that measures tissue stiffness. It is convenient for differentiating benign from malignant breast tumors. Our study aims to evaluate the role of qualitative ultrasound elastography scoring (ES), quantitative mass strain ratio (SR), and shear wave elasticity ratio (SWER) in differentiation between benign and malignant breast lesions. Results Among 51 female patients with 77 histopathologically proved breast lesions, 57 breast masses were malignant and 20 were benign. All patients were examined by B-mode ultrasound then strain and shear wave elastographic examinations using ultrasound machine (Logiq E9, GE Medical Systems) with 8.5–12 MHz high-frequency probes. Our study showed that ES best cut-off point > 3 with sensitivity, specificity, PPV, NPP, accuracy was 94.7%, 85%, 94.7%, 85%, 90.9%, respectively, and AUC = 0.926 at P < 0.001, mass SR the best cut-off point > 4.6 with sensitivity, specificity, PPV, NPP, accuracy was 96.5%, 80%, 93.2%, 88.9%, 92.2%, respectively, and AUC = 0.860 at P < 0.001, SWER the best cut-off value > 4.9 with sensitivity, specificity, PPV, NPP and accuracy was 91.2%, 80%, 92.9%, 76.2%, 93.5%, respectively, and AUC = 0.890 at P < 0.001. The mean mass strain ratio for malignant lesions is 10.1 ± 3.7 SD and for solid benign lesions 4.7 ± 4.3 SD (p value 0.001). The mean shear wave elasticity ratio for malignant lesions is 10.6 ± 5.4 SD and for benign (solid and cystic) lesions 3.6 ± 4.2 SD. Using ROC curve and Youden index, the difference in diagnostic performance between ES, SR and SWER was not significant in differentiation between benign and malignant breast lesions and also was non-significant difference when comparing them with conventional US alone. Conclusion ES, SR, and SWER have a high diagnostic performance in differentiating malignant from benign breast lesions with no statistically significant difference between them.


2017 ◽  
Vol 59 (6) ◽  
pp. 657-663 ◽  
Author(s):  
Jin Hee Moon ◽  
Ji-Young Hwang ◽  
Jeong Seon Park ◽  
Sung Hye Koh ◽  
Sun-Young Park

Background Shear wave elastography (SWE) using a region of interest (ROI) can demonstrate the quantitative elasticity of breast lesions. Purpose To prospectively evaluate the impact of two different ROI sizes on the diagnostic performance of SWE for differentiating benign and malignant breast lesions. Material and Methods A total of 154 breast lesions were included. Two types of ROIs were investigated: one involving an approximately 2-mm diameter, small round ROIs placed over the stiffest area of the lesion, as determined by SWE (ROI-S); and another ROI drawn along the margin of the lesion using a touch pen or track ball to encompass the entire lesion (ROI-M). Maximum elasticity (Emax), mean elasticity (Emean), minimum elasticity (Emin), and standard deviation (SD) were measured for the two ROIs. The area under the receiver operating characteristic curve (AUC) as well as the sensitivity and specificity of each elasticity value were determined. Results The AUCs for ROI-S were higher than those for ROI-M when differentiating benign and malignant breast solid lesions. The Emax, Emean, Emin, and SD of the elasticity values for ROI-S were 0.865, 0.857, 0.816, and 0.849, respectively, and for ROI-M were 0.820, 0.780, 0.724, and 0.837, respectively. However, only Emax ( P = 0.0024) and Emean ( P = 0.0015) showed statistically significant differences. For ROI-S, the sensitivity and specificity of Emax were 78.8% and 84.3%, respectively, and those for Emean were 80.8% and 81.4%, respectively. Conclusion Using ROI-S with Emax and Emean has better diagnostic performance than ROI-M for differentiating between benign and malignant breast lesions.


2021 ◽  
Vol 48 (1) ◽  
pp. 53-61
Author(s):  
Wen-tao Kong ◽  
Yin Wang ◽  
Wei-jun Zhou ◽  
Yi-dan Zhang ◽  
Wen-ping Wang ◽  
...  

Author(s):  
Logan Rowe ◽  
Alexander J. Kaczkowski ◽  
Tung-Wei Lin ◽  
Gavin Horn ◽  
Harley Johnson

Abstract A nondestructive photoelastic method is presented for characterizing surface microcracks in monocrystalline silicon wafers, calculating the strength of the wafers, and predicting Weibull parameters under various loading conditions. Defects are first classified from through thickness infrared photoelastic images using a support vector machine learning algorithm. Characteristic wafer strength is shown to vary with the angle of applied uniaxial tensile load, showing greater strength when loaded perpendicular to the direction of wire motion than when loaded along the direction of wire motion. Observed variations in characteristic strength and Weibull shape modulus with applied tensile loading direction stem from the distribution of crack orientations and the bulk stress field acting on the microcracks. Using this method it is possible to improve manufacturing processes for silicon wafers by rapidly, accurately, and nondestructively characterizing large batches in an automated way.


2018 ◽  
Vol 26 (1) ◽  
pp. 141-155 ◽  
Author(s):  
Li Luo ◽  
Fengyi Zhang ◽  
Yao Yao ◽  
RenRong Gong ◽  
Martina Fu ◽  
...  

Surgery cancellations waste scarce operative resources and hinder patients’ access to operative services. In this study, the Wilcoxon and chi-square tests were used for predictor selection, and three machine learning models – random forest, support vector machine, and XGBoost – were used for the identification of surgeries with high risks of cancellation. The optimal performances of the identification models were as follows: sensitivity − 0.615; specificity − 0.957; positive predictive value − 0.454; negative predictive value − 0.904; accuracy − 0.647; and area under the receiver operating characteristic curve − 0.682. Of the three models, the random forest model achieved the best performance. Thus, the effective identification of surgeries with high risks of cancellation is feasible with stable performance. Models and sampling methods significantly affect the performance of identification. This study is a new application of machine learning for the identification of surgeries with high risks of cancellation and facilitation of surgery resource management.


2020 ◽  
Author(s):  
Murad Megjhani ◽  
Kalijah Terilli ◽  
Ayham Alkhachroum ◽  
David J. Roh ◽  
Sachin Agarwal ◽  
...  

AbstractObjectiveTo develop a machine learning based tool, using routine vital signs, to assess delayed cerebral ischemia (DCI) risk over time.MethodsIn this retrospective analysis, physiologic data for 540 consecutive acute subarachnoid hemorrhage patients were collected and annotated as part of a prospective observational cohort study between May 2006 and December 2014. Patients were excluded if (i) no physiologic data was available, (ii) they expired prior to the DCI onset window (< post bleed day 3) or (iii) early angiographic vasospasm was detected on admitting angiogram. DCI was prospectively labeled by consensus of treating physicians. Occurrence of DCI was classified using various machine learning approaches including logistic regression, random forest, support vector machine (linear and kernel), and an ensemble classifier, trained on vitals and subject characteristic features. Hourly risk scores were generated as the posterior probability at time t. We performed five-fold nested cross validation to tune the model parameters and to report the accuracy. All classifiers were evaluated for good discrimination using the area under the receiver operating characteristic curve (AU-ROC) and confusion matrices.ResultsOf 310 patients included in our final analysis, 101 (32.6%) patients developed DCI. We achieved maximal classification of 0.81 [0.75-0.82] AU-ROC. We also predicted 74.7 % of all DCI events 12 hours before typical clinical detection with a ratio of 3 true alerts for every 2 false alerts.ConclusionA data-driven machine learning based detection tool offered hourly assessments of DCI risk and incorporated new physiologic information over time.


PeerJ ◽  
2021 ◽  
Vol 9 ◽  
pp. e10884
Author(s):  
Xin Yu ◽  
Qian Yang ◽  
Dong Wang ◽  
Zhaoyang Li ◽  
Nianhang Chen ◽  
...  

Applying the knowledge that methyltransferases and demethylases can modify adjacent cytosine-phosphorothioate-guanine (CpG) sites in the same DNA strand, we found that combining multiple CpGs into a single block may improve cancer diagnosis. However, survival prediction remains a challenge. In this study, we developed a pipeline named “stacked ensemble of machine learning models for methylation-correlated blocks” (EnMCB) that combined Cox regression, support vector regression (SVR), and elastic-net models to construct signatures based on DNA methylation-correlated blocks for lung adenocarcinoma (LUAD) survival prediction. We used methylation profiles from the Cancer Genome Atlas (TCGA) as the training set, and profiles from the Gene Expression Omnibus (GEO) as validation and testing sets. First, we partitioned the genome into blocks of tightly co-methylated CpG sites, which we termed methylation-correlated blocks (MCBs). After partitioning and feature selection, we observed different diagnostic capacities for predicting patient survival across the models. We combined the multiple models into a single stacking ensemble model. The stacking ensemble model based on the top-ranked block had the area under the receiver operating characteristic curve of 0.622 in the TCGA training set, 0.773 in the validation set, and 0.698 in the testing set. When stratified by clinicopathological risk factors, the risk score predicted by the top-ranked MCB was an independent prognostic factor. Our results showed that our pipeline was a reliable tool that may facilitate MCB selection and survival prediction.


mBio ◽  
2020 ◽  
Vol 11 (3) ◽  
Author(s):  
Begüm D. Topçuoğlu ◽  
Nicholas A. Lesniak ◽  
Mack T. Ruffin ◽  
Jenna Wiens ◽  
Patrick D. Schloss

ABSTRACT Machine learning (ML) modeling of the human microbiome has the potential to identify microbial biomarkers and aid in the diagnosis of many diseases such as inflammatory bowel disease, diabetes, and colorectal cancer. Progress has been made toward developing ML models that predict health outcomes using bacterial abundances, but inconsistent adoption of training and evaluation methods call the validity of these models into question. Furthermore, there appears to be a preference by many researchers to favor increased model complexity over interpretability. To overcome these challenges, we trained seven models that used fecal 16S rRNA sequence data to predict the presence of colonic screen relevant neoplasias (SRNs) (n = 490 patients, 261 controls and 229 cases). We developed a reusable open-source pipeline to train, validate, and interpret ML models. To show the effect of model selection, we assessed the predictive performance, interpretability, and training time of L2-regularized logistic regression, L1- and L2-regularized support vector machines (SVM) with linear and radial basis function kernels, a decision tree, random forest, and gradient boosted trees (XGBoost). The random forest model performed best at detecting SRNs with an area under the receiver operating characteristic curve (AUROC) of 0.695 (interquartile range [IQR], 0.651 to 0.739) but was slow to train (83.2 h) and not inherently interpretable. Despite its simplicity, L2-regularized logistic regression followed random forest in predictive performance with an AUROC of 0.680 (IQR, 0.625 to 0.735), trained faster (12 min), and was inherently interpretable. Our analysis highlights the importance of choosing an ML approach based on the goal of the study, as the choice will inform expectations of performance and interpretability. IMPORTANCE Diagnosing diseases using machine learning (ML) is rapidly being adopted in microbiome studies. However, the estimated performance associated with these models is likely overoptimistic. Moreover, there is a trend toward using black box models without a discussion of the difficulty of interpreting such models when trying to identify microbial biomarkers of disease. This work represents a step toward developing more-reproducible ML practices in applying ML to microbiome research. We implement a rigorous pipeline and emphasize the importance of selecting ML models that reflect the goal of the study. These concepts are not particular to the study of human health but can also be applied to environmental microbiology studies.


Sign in / Sign up

Export Citation Format

Share Document