Radiomics of locally advanced rectal cancer: machine learning-based prediction of response to neoadjuvant chemoradiotherapy using pre-treatment sagittal T2-weighted MRI.

PURPOSE: Variable response to neoadjuvant chemoradiotherapy (nCRT) is observed among individuals with locally advanced rectal cancer (LARC), having a significant impact on patient management. In this work, we aimed to investigate the potential value of machine learning (ML)-based magnetic resonance imaging (MRI) radiomics in predicting therapeutic response to nCRT in patients with LARC. MATERIALS AND METHODS: Seventy-six patients with LARC were included in this retrospective study. Radiomic features were extracted from pre-treatment sagittal T2-weighted MRI images, with 3D segmentation. Dimension reduction was performed with a reliability analysis, pair-wise correlation analysis, analysis of variance, recursive feature elimination, Kruskal-Wallis, and Relief methods. Models were created using four different algorithms. In addition to radiomic models, clinical only and different combined models were developed and compared. The reference standard was tumor regression grade (TRG) based on the Modified Ryan Scheme (TRG 0 vs TRG 1-3). Models were compared based on net reclassification index (NRI). Clinical utility was assessed with decision curve analysis (DCA). RESULTS: Number of features with excellent reliability is 106. The best result was achieved with radiomic only model using eight features. The area under the curve (AUC), accuracy, sensitivity, and specificity for validation were 0.753 (standard deviation [SD], 0.082), 81.1%, 83.8%, and 75.0%; for testing, 0.705 (SD, 0.145), 73.9%, 81.2%, and 57.1%, respectively. Based on the clinical only model as reference, NRI for radiomic only model was the best. DCA also showed better clinical utility for radiomic only model. CONCLUSIONS: ML-based T2-weighted MRI radiomics might have a potential in predicting response to nCRT in patients with LARC.

Tubular gastric adenocarcinoma: machine learning-based CT texture analysis for predicting lymphovascular and perineural invasion.

PURPOSE: Lymphovascular invasion (LVI) and perineural invasion (PNI) are associated with poor prognosis in gastric cancers. In this work, we aimed to investigate the potential role of computed tomography (CT) texture analysis in predicting LVI and PNI in patients with tubular gastric adenocarcinoma (GAC) using a machine learning (ML) approach. METHODS: Sixty-eight patients who underwent total gastrectomy with curative (R0) resection and D2-lymphadenectomy were included in this retrospective study. Texture features were extracted from the portal venous phase CT images. Dimension reduction was first done with a reproducibility analysis by two radiologists. Then, a feature selection algorithm was used to further reduce the high-dimensionality of the radiomic data. Training and test splits were created with 100 random samplings. ML-based classifications were done using adaptive boosting, k-nearest neighbors, Naive Bayes, neural network, random forest, stochastic gradient descent, support vector machine, and decision tree. Predictive performance of the ML algorithms was mainly evaluated using the mean area under the curve (AUC) metric. RESULTS: Among 271 texture features, 150 features had excellent reproducibility, which were included in the further feature selection process. Dimension reduction steps yielded five texture features for LVI and five for PNI. Considering all eight ML algorithms, mean AUC and accuracy ranges for predicting LVI were 0.777-0.894 and 76%-81.5%, respectively. For predicting PNI, mean AUC and accuracy ranges were 0.482-0.754 and 54%-68.2%, respectively. The best performances for predicting LVI and PNI were achieved with the random forest and Naive Bayes algorithms, respectively. CONCLUSION: ML-based CT texture analysis has a potential for predicting LVI and PNI of the tubular GACs. Overall, the method was more successful in predicting LVI than PNI.

Computed tomography texture analysis in patients with gastric cancer: a quantitative imaging biomarker for preoperative evaluation before neoadjuvant chemotherapy treatment.

PURPOSE: The aim of the study is to explore the role of computed tomography texture analysis (CT-TA) for predicting clinical T and N stages and tumor grade before neoadjuvant chemotherapy treatment in gastric cancer (GC) patients during the preoperative period. MATERIALS AND METHODS: CT images of 114 patients with GC were included in this retrospective study. Following pre-processing steps, textural features were extracted using MaZda software in the portal venous phase. We evaluated and analyzed texture features of six principal categories for differentiating between T stages (T1,2 vs T3,4), N stages (N+ vs N-) and grades (low-intermediate vs. high). Classification was performed based on texture parameters with high model coefficients in linear discriminant analysis (LDA). RESULTS: Dimension-reduction steps yielded five textural features for T stage, three for N stage and two for tumor grade. The discriminatory capacities of T stage, N stage and tumor grade were 90.4%, 81.6% and 64.5%, respectively, when LDA algorithm was employed. CONCLUSION: CT-TA yields potentially useful imaging biomarkers for predicting the T and N stages of patients with GC and can be used for preoperative evaluation before neoadjuvant treatment planning.

Radiogenomics of lower-grade gliomas: machine learning-based MRI texture analysis for predicting 1p/19q codeletion status.

OBJECTIVE: To evaluate the potential value of the machine learning (ML)-based MRI texture analysis for predicting 1p/19q codeletion status of lower-grade gliomas (LGG), using various state-of-the-art ML algorithms. MATERIALS AND METHODS: For this retrospective study, 107 patients with LGG were included from a public database. Texture features were extracted from conventional T2-weighted and contrast-enhanced T1-weighted MRI images, using LIFEx software. Training and unseen validation splits were created using stratified 10-fold cross-validation technique along with minority over-sampling. Dimension reduction was done using collinearity analysis and feature selection (ReliefF). Classifications were done using adaptive boosting, k-nearest neighbours, naive Bayes, neural network, random forest, stochastic gradient descent, and support vector machine. Friedman test and pairwise post hoc analyses were used for comparison of classification performances based on the area under the curve (AUC). RESULTS: Overall, the predictive performance of the ML algorithms were statistically significantly different, χ2(6) = 26.7, p < 0.001. There was no statistically significant difference among the performance of the neural network, naive Bayes, support vector machine, random forest, and stochastic gradient descent, adjusted p > 0.05. The mean AUC and accuracy values of these five algorithms ranged from 0.769 to 0.869 and from 80.1 to 84%, respectively. The neural network had the highest mean rank with mean AUC and accuracy values of 0.869 and 83.8%, respectively. CONCLUSIONS: The ML-based MRI texture analysis might be a promising non-invasive technique for predicting the 1p/19q codeletion status of LGGs. Using this technique along with various ML algorithms, more than four-fifths of the LGGs can be correctly classified. KEY POINTS: • More than four-fifths of the lower-grade gliomas can be correctly classified with machine learning-based MRI texture analysis. Satisfying classification outcomes are not limited to a single algorithm. • A few-slice-based volumetric segmentation technique would be a valid approach, providing satisfactory predictive textural information and avoiding excessive segmentation duration in clinical practice. • Feature selection is sensitive to different patient data set samples so that each sampling leads to the selection of different feature subsets, which needs to be considered in future works.