Robustness of PET Radiomics Features: Impact of Co-Registration with MRI

Radiomics-based discriminant analysis of principal components to stratify the treatment response of lung metastases following stereotactic body radiation therapy

PURPOSE

Discriminant analysis of principal components (DAPC) was introduced to describe the clusters of genetically related individuals focusing on the variation between the groups of individuals. Borrowing this approach, we evaluated the potential of DAPC for the evaluation of clusters in terms of treatment response to SBRT of lung lesions using radiomics analysis on pre-treatment CT images.

MATERIALS AND METHODS

80 pulmonary metastases from 56 patients treated with SBRT were analyzed. Treatment response was stratified as complete, incomplete and null responses. For each lesion, 107 radiomics features were extracted using the PyRadiomics software. The concordance correlation coefficients (CCC) between the radiomics features obtained by two segmentations were calculated. DAPC analysis was performed to infer the structure of "radiomically" related lesions for treatment response assessment. The DAPC was performed using the "adegenet" package for the R software.

RESULTS

The overall mean CCC was 0.97 ± 0.14. The analysis yields 14 dimensions in order to explain 95 % of the variance. DAPC was able to group the 80 lesions into the 3 different clusters based on treatment response depending on the radiomics features characteristics. The first Linear Discriminant achieved the best discrimination of individuals into the three pre-defined groups. The greater radiomics loadings who contributed the most to the treatment response differentiation were associated with the "sphericity", "correlation" and "maximal correlation coefficient" features.

CONCLUSION

This study demonstrates that a DAPC analysis based on radiomics features obtained from pretreatment CT is able to provide a reliable stratification of complete, incomplete or null response of lung metastases following SBRT.

A different perspective on 18F-FDG PET radiomics in colorectal cancer patients: The relationship between intra & peritumoral analysis and pathological findings

OBJECTIVE

We aimed to determine the value of 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) based primary tumoral and peritumoral radiomics in the prediction of tumor deposits (TDs), tumor budding (TB) and extramural venous invasion (EMVI) of colorectal cancer (CRC).

METHODS

Our retrospective study included 77 CRC patients who had preoperative 18F-FDG PET/CT between June 2020 and February 2022. A total of 131 radiomic features were extracted from primary tumors and peritumoral areas on PET/CT fusion images. The relationship between TDs, TB, EMVI and T stage in the postoperative pathology of the tumors and radiomic features was investigated. Features with a correlation coefficient (CC) less than 0.8 were analyzed by logistic regression. The area under curve (AUC) obtained from the receiver operating characteristic analysis was used to measure the model performance.

RESULTS

A model was developed from primary tumoral and peritumoral radiomics data to predict T stage (AUC 0.931), and also a predictive model was constructed from primary tumor derived radiomics to predict EMVI (AUC 0.739). Radiomic data derived from the primary tumor was obtained as a predictive prognostic factor in predicting TDs and a peritumoral feature was found to be a prognostic factor in predicting TB.

CONCLUSIONS

Intratumoral and peritumoral radiomics derived from 18F-FDG PET/CT are useful for non-invasive early prediction of pathological features that have important implications in the management of CRC.

Machine learning for differentiation of lipid-poor adrenal adenoma and subclinical pheochromocytoma based on multiphase CT imaging radiomics

BACKGROUND

There is a paucity of research investigating the application of machine learning techniques for distinguishing between lipid-poor adrenal adenoma (LPA) and subclinical pheochromocytoma (sPHEO) based on radiomic features extracted from non-contrast and dynamic contrast-enhanced computed tomography (CT) scans of the abdomen.

METHODS

We conducted a retrospective analysis of multiphase spiral CT scans, including non-contrast, arterial, venous, and delayed phases, as well as thin- and thick-thickness images from 134 patients with surgically and pathologically confirmed. A total of 52 patients with LPA and 44 patients with sPHEO were randomly assigned to training/testing sets in a 7:3 ratio. Additionally, a validation set was comprised of 22 LPA cases and 16 sPHEO cases from two other hospitals. We used 3D Slicer and PyRadiomics to segment tumors and extract radiomic features, respectively. We then applied T-test and least absolute shrinkage and selection operator (LASSO) to select features. Six binary classifiers, including K-nearest neighbor (KNN), logistic regression (LR), decision tree (DT), random forest (RF), support vector machine (SVM), and multi-layer perceptron (MLP), were employed to differentiate LPA from sPHEO. Receiver operating characteristic (ROC) curves and area under the curve (AUC) values were compared using DeLong's method.

RESULTS

All six classifiers showed good diagnostic performance for each phase and slice thickness, as well as for the entire CT data, with AUC values ranging from 0.706 to 1. Non-contrast CT densities of LPA were significantly lower than those of sPHEO (P < 0.001). However, using the optimal threshold for non-contrast CT density, sensitivity was only 0.743, specificity 0.744, and AUC 0.828. Delayed phase CT density yielded a sensitivity of 0.971, specificity of 0.641, and AUC of 0.814. In radiomics, AUC values for the testing set using non-contrast CT images were: KNN 0.919, LR 0.979, DT 0.835, RF 0.967, SVM 0.979, and MLP 0.981. In the validation set, AUC values were: KNN 0.891, LR 0.974, DT 0.891, RF 0.964, SVM 0.949, and MLP 0.979.

CONCLUSIONS

The machine learning model based on CT radiomics can accurately differentiate LPA from sPHEO, even using non-contrast CT data alone, making contrast-enhanced CT unnecessary for diagnosing LPA and sPHEO.

A MRI-based radiomics model for predicting the response to anlotinb combined with temozolomide in recurrent malignant glioma patients

OBJECTIVE

Anlotinib is a multitarget anti-angiogenic drug that combined with temozolomide (TMZ) can effectively prolongs the overall survival (OS) of recurrent malignant glioma(rMG),but some patients do not respond to anlotinib combined with TMZ. These patients were associated with a worse prognosis and lack effective identification methods. Therefore, it is necessary to differentiate patients who may have good response to anlotinb in combination with TMZ from those who are not, in order to provide personalized targeted therapies.

METHODS

Fifty three rMG patients (42 in training cohort and 11 in testing cohort) receiving anlotinib combined with TMZ were enrolled. A total of 3668 radiomics features were extracted from the recurrent MRI images. Radiomics features are reduced and filtered by hypothesis testing and Least Absolute Shrinkage And Selection (LASSO) regression. Eight machine learning models construct the radiomics model, and then screen out the optimal model. The performance of the model was assessed by its discrimination, calibration, and clinical usefulness with validation.

RESULTS

Fifty three patients with rMG were enrolled in our study. Thirty four patients displayed effective treatment response, showed a higher survival benefits than non-response group, the median progression-free survival(PFS) was 8.53 months versus 5.33 months (p = 0.06) and the median OS was 19.9 months and 7.33 months (p = 0.029), respectively. Three radiomics features were incorporated into the model construction as final variables after LASSO regression analysis. In testing cohort, Logistic Regression (LR) model has the best performance with an Area Under the Curve (AUC) of 0.93 compared with other models, which can effectively predict the response of rMG patients to anlotinib in combination with TMZ. The calibration curve confirmed the agreement between the observed actual and prediction probability. Within the reasonable threshold probability range (0.38-0.88), the radiomics model shows good clinical utility.

CONCLUSIONS

The above-described radiomics model performed well, which can serve as a clinical tool for individualized prediction of the response to anlotinb combined with TMZ in rMG patients.

Differentiation of retroperitoneal paragangliomas and schwannomas based on computed tomography radiomics

The purpose of this study was to differentiate the retroperitoneal paragangliomas and schwannomas using computed tomography (CT) radiomics. This study included 112 patients from two centers who pathologically confirmed retroperitoneal pheochromocytomas and schwannomas and underwent preoperative CT examinations. Radiomics features of the entire primary tumor were extracted from non-contrast enhancement (NC), arterial phase (AP) and venous phase (VP) CT images. The least absolute shrinkage and selection operator method was used to screen out key radiomics signatures. Radiomics, clinical and clinical-radiomics combined models were built to differentiate the retroperitoneal paragangliomas and schwannomas. Model performance and clinical usefulness were evaluated by receiver operating characteristic curve, calibration curve and decision curve. In addition, we compared the diagnostic accuracy of radiomics, clinical and clinical-radiomics combined models with radiologists for pheochromocytomas and schwannomas in the same set of data. Three NC, 4 AP, and 3 VP radiomics features were retained as the final radiomics signatures for differentiating the paragangliomas and schwannomas. The CT characteristics CT attenuation value of NC and the enhancement magnitude at AP and VP were found to be significantly different statistically (P < 0.05). The NC, AP, VP, Radiomics and clinical models had encouraging discriminative performance. The clinical-radiomics combined model that combined radiomics signatures and clinical characteristics showed excellent performance, with an area under curve (AUC) values were 0.984 (95% CI 0.952-1.000) in the training cohort, 0.955 (95% CI 0.864-1.000) in the internal validation cohort and 0.871 (95% CI 0.710-1.000) in the external validation cohort. The accuracy, sensitivity and specificity were 0.984, 0.970 and 1.000 in the training cohort, 0.960, 1.000 and 0.917 in the internal validation cohort and 0.917, 0.923 and 0.818 in the external validation cohort, respectively. Additionally, AP, VP, Radiomics, clinical and clinical-radiomics combined models had a higher diagnostic accuracy for pheochromocytomas and schwannomas than the two radiologists. Our study demonstrated the CT-based radiomics models has promising performance in differentiating the paragangliomas and schwannomas.

Phenotyping the Histopathological Subtypes of Non-Small-Cell Lung Carcinoma: How Beneficial Is Radiomics?

The aim of this study was to investigate the usefulness of radiomics in the absence of well-defined standard guidelines. Specifically, we extracted radiomics features from multicenter computed tomography (CT) images to differentiate between the four histopathological subtypes of non-small-cell lung carcinoma (NSCLC). In addition, the results that varied with the radiomics model were compared. We investigated the presence of the batch effects and the impact of feature harmonization on the models' performance. Moreover, the question on how the training dataset composition influenced the selected feature subsets and, consequently, the model's performance was also investigated. Therefore, through combining data from the two publicly available datasets, this study involves a total of 152 squamous cell carcinoma (SCC), 106 large cell carcinoma (LCC), 150 adenocarcinoma (ADC), and 58 no other specified (NOS). Through the matRadiomics tool, which is an example of Image Biomarker Standardization Initiative (IBSI) compliant software, 1781 radiomics features were extracted from each of the malignant lesions that were identified in CT images. After batch analysis and feature harmonization, which were based on the ComBat tool and were integrated in matRadiomics, the datasets (the harmonized and the non-harmonized) were given as an input to a machine learning modeling pipeline. The following steps were articulated: (i) training-set/test-set splitting (80/20); (ii) a Kruskal-Wallis analysis and LASSO linear regression for the feature selection; (iii) model training; (iv) a model validation and hyperparameter optimization; and (v) model testing. Model optimization consisted of a 5-fold cross-validated Bayesian optimization, repeated ten times (inner loop). The whole pipeline was repeated 10 times (outer loop) with six different machine learning classification algorithms. Moreover, the stability of the feature selection was evaluated. Results showed that the batch effects were present even if the voxels were resampled to an isotropic form and whether feature harmonization correctly removed them, even though the models' performances decreased. Moreover, the results showed that a low accuracy (61.41%) was reached when differentiating between the four subtypes, even though a high average area under curve (AUC) was reached (0.831). Further, a NOS subtype was classified as almost completely correct (true positive rate ~90%). The accuracy increased (77.25%) when only the SCC and ADC subtypes were considered, as well as when a high AUC (0.821) was obtained-although harmonization decreased the accuracy to 58%. Moreover, the features that contributed the most to models' performance were those extracted from wavelet decomposed and Laplacian of Gaussian (LoG) filtered images and they belonged to the texture feature class.. In conclusion, we showed that our multicenter data were affected by batch effects, that they could significantly alter the models' performance, and that feature harmonization correctly removed them. Although wavelet features seemed to be the most informative features, an absolute subset could not be identified since it changed depending on the training/testing splitting. Moreover, performance was influenced by the chosen dataset and by the machine learning methods, which could reach a high accuracy in binary classification tasks, but could underperform in multiclass problems. It is, therefore, essential that the scientific community propose a more systematic radiomics approach, focusing on multicenter studies, with clear and solid guidelines to facilitate the translation of radiomics to clinical practice.

Dynamic radiomics for predicting the efficacy of antiangiogenic therapy in colorectal liver metastases

Background and objective

For patients with advanced colorectal liver metastases (CRLMs) receiving first-line anti-angiogenic therapy, an accurate, rapid and noninvasive indicator is urgently needed to predict its efficacy. In previous studies, dynamic radiomics predicted more accurately than conventional radiomics. Therefore, it is necessary to establish a dynamic radiomics efficacy prediction model for antiangiogenic therapy to provide more accurate guidance for clinical diagnosis and treatment decisions.

Methods

In this study, we use dynamic radiomics feature extraction method that extracts static features using tomographic images of different sequences of the same patient and then quantifies them into new dynamic features for the prediction of treatmentefficacy. In this retrospective study, we collected 76 patients who were diagnosed with unresectable CRLM between June 2016 and June 2021 in the First Hospital of China Medical University. All patients received standard treatment regimen of bevacizumab combined with chemotherapy in the first-line treatment, and contrast-enhanced abdominal CT (CECT) scans were performed before treatment. Patients with multiple primary lesions as well as missing clinical or imaging information were excluded. Area Under Curve (AUC) and accuracy were used to evaluate model performance. Regions of interest (ROIs) were independently delineated by two radiologists to extract radiomics features. Three machine learning algorithms were used to construct two scores based on the best response and progression-free survival (PFS).

Results

For the task that predict the best response patients will achieve after treatment, by using ROC curve analysis, it can be seen that the relative change rate (RCR) feature performed best among all features and best in linear discriminantanalysis (AUC: 0.945 and accuracy: 0.855). In terms of predicting PFS, the Kaplan-Meier plots suggested that the score constructed using the RCR features could significantly distinguish patients with good response from those with poor response (Two-sided P<0.0001 for survival analysis).

Conclusions

This study demonstrates that the application of dynamic radiomics features can better predict the efficacy of CRLM patients receiving antiangiogenic therapy compared with conventional radiomics features. It allows patients to have a more accurate assessment of the effect of medical treatment before receiving treatment, and this assessment method is noninvasive, rapid, and less expensive. Dynamic radiomics model provides stronger guidance for the selection of treatment options and precision medicine.

Development of MRI-based radiomics predictive model for classifying endometrial lesions

An unbiased and accurate diagnosis of benign and malignant endometrial lesions is essential for the gynecologist, as each type might require distinct treatment. Radiomics is a quantitative method that could facilitate deep mining of information and quantification of the heterogeneity in images, thereby aiding clinicians in proper lesion diagnosis. The aim of this study is to develop an appropriate predictive model for the classification of benign and malignant endometrial lesions, and evaluate potential clinical applicability of the model. 139 patients with pathologically-confirmed endometrial lesions from January 2018 to July 2020 in two independent centers (center A and B) were finally analyzed. Center A was used for training set, while center B was used for test set. The lesions were manually drawn on the largest slice based on the lesion area by two radiologists. After feature extraction and feature selection, the possible associations between radiomics features and clinical parameters were assessed by Uni- and multi- variable logistic regression. The receiver operator characteristic (ROC) curve and DeLong validation were employed to evaluate the possible predictive performance of the models. Decision curve analysis (DCA) was used to evaluate the net benefit of the radiomics nomogram. A radiomics prediction model was established from the 15 selected features, and were found to be relatively high discriminative on the basis of the area under the ROC curve (AUC) for both the training and the test cohorts (AUC = 0.90 and 0.85, respectively). The radiomics nomogram also showed good performance of discrimination for both the training and test cohorts (AUC = 0.91 and 0.86, respectively), and the DeLong test shows that AUCs were significantly different between clinical parameters and nomogram. The result of DCA demonstrated the clinical usefulness of this novel nomogram method. The predictive model constructed based on MRI radiomics and clinical parameters indicated a highly diagnostic efficiency, thereby implying its potential clinical usefulness for the precise identification and prediction of endometrial lesions.

Risk stratification and prognostic value of multi-modal MRI-based radiomics for extranodal nasal-type NK/T-cell lymphoma

BACKGROUND

Magnetic resonance imaging (MRI) performs well in the locoregional assessment of extranodal nasal-type NK/T-cell lymphoma (ENKTCL). It's important to assess the value of multi-modal MRI-based radiomics for estimating overall survival (OS) in patients with ENKTCL.

METHODS

Patients with ENKTCL in a prospectively cohort were systemically reviewed and all the pretreatment MRI were acquisitioned. An unsupervised spectral clustering method was used to identify risk groups of patients and radiomic features. A nomogram-revised risk index (NRI) plus MRI radiomics signature (NRI-M) was developed, and compared with the NRI.

RESULTS

The 2 distinct type I and II groups of the MRI radiomics signatures were identified. The 5-year OS rates between the type I and type II groups were 87.2% versus 67.3% (P = 0.002) in all patients, and 88.8% versus 69.2% (P = 0.003) in early-stage patients. The discrimination and calibration of the NRI-M for OS prediction demonstrated a better performance than that of either MRI radiomics or NRI, with a mean area under curve (AUC) of 0.748 and 0.717 for predicting the 5-year OS in all-stages and early-stage patients.

CONCLUSIONS

The NRI-M model has good performance for predicting the prognosis of ENKTCL and may help design clinical trials and improve clinical decision making.

CT-based machine learning radiomics predicts CCR5 expression level and survival in ovarian cancer

OBJECTIVE

We aimed to evaluate the prognostic value of C-C motif chemokine receptor type 5 (CCR5) expression level for patients with ovarian cancer and to establish a radiomics model that can predict CCR5 expression level using The Cancer Imaging Archive (TCIA) and The Cancer Genome Atlas (TCGA) database.

METHODS

A total of 343 cases of ovarian cancer from the TCGA were used for the gene-based prognostic analysis. Fifty seven cases had preoperative computed tomography (CT) images stored in TCIA with genomic data in TCGA were used for radiomics feature extraction and model construction. 89 cases with both TCGA and TCIA clinical data were used for radiomics model evaluation. After feature extraction, a radiomics signature was constructed using the least absolute shrinkage and selection operator (LASSO) regression analysis. A prognostic scoring system incorporating radiomics signature based on CCR5 expression level and clinicopathologic risk factors was proposed for survival prediction.

RESULTS

CCR5 was identified as a differentially expressed prognosis-related gene in tumor and normal sample, which were involved in the regulation of immune response and tumor invasion and metastasis. Four optimal radiomics features were selected to predict overall survival. The performance of the radiomics model for predicting the CCR5 expression level with 10-fold cross- validation achieved Area Under Curve (AUCs) of 0.770 and of 0.726, respectively, in the training and validation sets. A predictive nomogram was generated based on the total risk score of each patient, the AUCs of the time-dependent receiver operating characteristic (ROC) curve of the model was 0.8, 0.673 and 0.792 for 1-year, 3-year and 5-year, respectively. Along with clinical features, important imaging biomarkers could improve the overall survival accuracy of the prediction model.

CONCLUSION

The expression levels of CCR5 can affect the prognosis of patients with ovarian cancer. CT-based radiomics could serve as a new tool for prognosis prediction.

A CT based radiomics nomogram for differentiation between focal-type autoimmune pancreatitis and pancreatic ductal adenocarcinoma

Objectives

The purpose of this study was to develop and validate an CT-based radiomics nomogram for the preoperative differentiation of focal-type autoimmune pancreatitis from pancreatic ductal adenocarcinoma.

Methods

96 patients with focal-type autoimmune pancreatitis and pancreatic ductal adenocarcinoma have been enrolled in the study (32 and 64 cases respectively). All cases have been confirmed by imaging, clinical follow-up and/or pathology. The imaging data were considered as: 70% training cohort and 30% test cohort. Pancreatic lesions have been manually delineated by two radiologists and image segmentation was performed to extract radiomic features from the CT images. Independent-sample T tests and LASSO regression were used for feature selection. The training cohort was classified using a variety of machine learning-based classifiers, and 5-fold cross-validation has been performed. The classification performance was evaluated using the test cohort. Multivariate logistic regression analysis was then used to develop a radiomics nomogram model, containing the CT findings and Rad-Score. Calibration curves have been plotted showing the agreement between the predicted and actual probabilities of the radiomics nomogram model. Different patients have been selected to test and evaluate the model prediction process. Finally, receiver operating characteristic curves and decision curves were plotted, and the radiomics nomogram model was compared with a single model to visually assess its diagnostic ability.

Results

A total of 158 radiomics features were extracted from each image. 7 features were selected to construct the radiomics model, then a variety of classifiers were used for classification and multinomial logistic regression (MLR) was selected to be the optimal classifier. Combining CT findings with radiomics model, a prediction model based on CT findings and radiomics was finally obtained. The nomogram model showed a good sensitivity and specificity with AUCs of 0.87 and 0.83 in training and test cohorts, respectively. The areas under the curve and decision curve analysis showed that the radiomics nomogram model may provide better diagnostic performance than the single model and achieve greater clinical net benefits than the CT finding model and radiomics signature model individually.

Conclusions

The CT image-based radiomics nomogram model can accurately distinguish between focal-type autoimmune pancreatitis and pancreatic ductal adenocarcinoma patients and provide additional clinical benefits.

Machine learning models predict overall survival and progression free survival of non-surgical esophageal cancer patients with chemoradiotherapy based on CT image radiomics signatures

PURPOSE

To construct machine learning models for predicting progression free survival (PFS) and overall survival (OS) with esophageal squamous cell carcinoma (ESCC) patients.

METHODS

204 ESCC patients were randomly divided into training cohort (n = 143) and test cohort (n = 61) according to the ratio of 7:3. Two radiomics models were constructed by radiomics features, which were selected by LASSO Cox model to predict PFS and OS, respectively. Clinical features were selected by univariate and multivariate Cox proportional hazards model (p < 0.05). Combined radiomics and clinical model was developed by selected clinical and radiomics features. The receiver operating characteristic curve, Kaplan Meier curve and nomogram were used to display the capability of constructed models.

RESULTS

There were 944 radiomics features extracted based on volume of interest in CT images. There were six radiomics features and seven clinical features for PFS prediction and three radiomics features and three clinical features for OS prediction; The radiomics models showed general performance in training cohort and test cohort for prediction for prediction PFS (AUC, 0.664, 0.676. C-index, 0.65, 0.64) and OS (AUC, 0.634, 0.646.C-index, 0.64, 0.65). The combined models displayed high performance in training cohort and test cohort for prediction PFS (AUC, 0.856, 0.833. C-index, 0.81, 0.79) and OS (AUC, 0.742, 0.768. C-index, 0.72, 0.71).

CONCLUSION

We developed combined radiomics and clinical machine learning models with better performance than radiomics or clinical alone, which were used to accurate predict 3 years PFS and OS of non-surgical ESCC patients. The prediction results could provide a reference for clinical decision.

A comparative study between deep learning and radiomics models in grading liver tumors using hepatobiliary phase contrast-enhanced MR images

Purpose

To compare a deep learning model with a radiomics model in differentiating high-grade (LR-3, LR-4, LR-5) liver imaging reporting and data system (LI-RADS) liver tumors from low-grade (LR-1, LR-2) LI-RADS tumors based on the contrast-enhanced magnetic resonance images.

Methods

Magnetic resonance imaging scans of 361 suspected hepatocellular carcinoma patients were retrospectively reviewed. Lesion volume segmentation was manually performed by two radiologists, resulting in 426 lesions from the training set and 83 lesions from the test set. The radiomics model was constructed using a support vector machine (SVM) with pre-defined features, which was first selected using Chi-square test, followed by refining using binary least absolute shrinkage and selection operator (LASSO) regression. The deep learning model was established based on the DenseNet. Performance of the models was quantified by area under the receiver-operating characteristic curve (AUC), accuracy, sensitivity, specificity and F1-score.

Results

A set of 8 most informative features was selected from 1049 features to train the SVM classifier. The AUCs of the radiomics model were 0.857 (95% confidence interval [CI] 0.816–0.888) for the training set and 0.879 (95% CI 0.779–0.935) for the test set. The deep learning method achieved AUCs of 0.838 (95% CI 0.799–0.871) for the training set and 0.717 (95% CI 0.601–0.814) for the test set. The performance difference between these two models was assessed by t-test, which showed the results in both training and test sets were statistically significant.

Conclusion

The deep learning based model can be trained end-to-end with little extra domain knowledge, while the radiomics model requires complex feature selection. However, this process makes the radiomics model achieve better performance in this study with smaller computational cost and more potential on model interpretability.

Prediction of TTF-1 expression in non-small-cell lung cancer using machine learning-based radiomics

PURPOSE

To explore the feasibility and performance of machine learning-based radiomics models in predicting thyroid transcription factor-1 (TTF-1) expression in non-small cell lung cancer (NSCLC).

METHODS

A total of 227 NSCLC patients were included in this retrospective study and divided into the training set and test set with a ratio of 8:2 randomly. Lung tumors on CT images were semi-automatic segmented utilizing 3D Slicer. Radiomic features quantifying tumor intensity, shape, texture, and transformed wavelet were extracted using a Python toolkit. Variance threshold (VT), principal component analysis (PCA), and least absolute shrinkage selection operator (LASSO) were used to reduce features; logistic regression (LR), random forest (RF), and support vector machine (SVM) were used to develop classifier, respectively. The performance of the models was evaluated by areas under the curves (AUC) of receiver operating characteristic (ROC) curves. Different models were compared by the Delong test to determine the optimal algorithms.

RESULTS

Total 1968 radiomic features were extracted from the lung tumors images, and then 13, 15, and 13 stable features were selected by VT, PCA, and LASSO, respectively. Each classifier could discriminate against the TTF-1-positive groups with average AUC ranging from 0.601 to 0.784 in the training set. Among the models, three models constructed by the LASSO method showed satisfactory performance in the test set with AUC ranging from 0.715 to 0.787. The Delong test showed no significant difference between the LASSO models (P > 0.05).

CONCLUSION

Machine learning-based radiomics model could predict the expression of TTF-1 in NSCLC patients.

matRadiomics: A Novel and Complete Radiomics Framework, from Image Visualization to Predictive Model

Radiomics aims to support clinical decisions through its workflow, which is divided into: (i) target identification and segmentation, (ii) feature extraction, (iii) feature selection, and (iv) model fitting. Many radiomics tools were developed to fulfill the steps mentioned above. However, to date, users must switch different software to complete the radiomics workflow. To address this issue, we developed a new free and user-friendly radiomics framework, namely matRadiomics, which allows the user: (i) to import and inspect biomedical images, (ii) to identify and segment the target, (iii) to extract the features, (iv) to reduce and select them, and (v) to build a predictive model using machine learning algorithms. As a result, biomedical images can be visualized and segmented and, through the integration of Pyradiomics into matRadiomics, radiomic features can be extracted. These features can be selected using a hybrid descriptive–inferential method, and, consequently, used to train three different classifiers: linear discriminant analysis, k-nearest neighbors, and support vector machines. Model validation is performed using k-fold cross-Validation and k-fold stratified cross-validation. Finally, the performance metrics of each model are shown in the graphical interface of matRadiomics. In this study, we discuss the workflow, architecture, application, future development of matRadiomics, and demonstrate its working principles in a real case study with the aim of establishing a reference standard for the whole radiomics analysis, starting from the image visualization up to the predictive model implementation.

Image Processing and Analysis for Preclinical and Clinical Applications

Preclinical and clinical imaging aims to characterize and measure biological processes and diseases in animals [...]

Elimination of Defects in Mammograms Caused by a Malfunction of the Device Matrix

Today, the processing and analysis of mammograms is quite an important field of medical image processing. Small defects in images can lead to false conclusions. This is especially true when the distortion occurs due to minor malfunctions in the equipment. In the present work, an algorithm for eliminating a defect is proposed, which includes a change in intensity on a mammogram and deteriorations in the contrast of individual areas. The algorithm consists of three stages. The first is the defect identification stage. The second involves improvement and equalization of the contrasts of different parts of the image outside the defect. The third involves restoration of the defect area via a combination of interpolation and an artificial neural network. The mammogram obtained as a result of applying the algorithm shows significantly better image quality and does not contain distortions caused by changes in brightness of the pixels. The resulting images are evaluated using Blind/Referenceless Image Spatial Quality Evaluator (BRISQUE) and Naturalness Image Quality Evaluator (NIQE) metrics. In total, 98 radiomics features are extracted from the original and obtained images, and conclusions are drawn about the minimum changes in features between the original image and the image obtained by the proposed algorithm.

A New Preclinical Decision Support System Based on PET Radiomics: A Preliminary Study on the Evaluation of an Innovative 64Cu-Labeled Chelator in Mouse Models

The 64Cu-labeled chelator was analyzed in vivo by positron emission tomography (PET) imaging to evaluate its biodistribution in a murine model at different acquisition times. For this purpose, nine 6-week-old female Balb/C nude strain mice underwent micro-PET imaging at three different time points after 64Cu-labeled chelator injection. Specifically, the mice were divided into group 1 (acquisition 1 h after [64Cu] chelator administration, n = 3 mice), group 2 (acquisition 4 h after [64Cu]chelator administration, n = 3 mice), and group 3 (acquisition 24 h after [64Cu] chelator administration, n = 3 mice). Successively, all PET studies were segmented by means of registration with a standard template space (3D whole-body Digimouse atlas), and 108 radiomics features were extracted from seven organs (namely, heart, bladder, stomach, liver, spleen, kidney, and lung) to investigate possible changes over time in [64Cu]chelator biodistribution. The one-way analysis of variance and post hoc Tukey Honestly Significant Difference test revealed that, while heart, stomach, spleen, kidney, and lung districts showed a very low percentage of radiomics features with significant variations (p-value < 0.05) among the three groups of mice, a large number of features (greater than 60% and 50%, respectively) that varied significantly between groups were observed in bladder and liver, indicating a different in vivo uptake of the 64Cu-labeled chelator over time. The proposed methodology may improve the method of calculating the [64Cu]chelator biodistribution and open the way towards a decision support system in the field of new radiopharmaceuticals used in preclinical imaging trials.

Artificial Intelligence Applications on Restaging [18F]FDG PET/CT in Metastatic Colorectal Cancer: A Preliminary Report of Morpho-Functional Radiomics Classification for Prediction of Disease Outcome

The aim of this study was to investigate the application of [18F]FDG PET/CT images-based textural features analysis to propose radiomics models able to early predict disease progression (PD) and survival outcome in metastatic colorectal cancer (MCC) patients after first adjuvant therapy. For this purpose, 52 MCC patients who underwent [18F]FDGPET/CT during the disease restaging process after the first adjuvant therapy were analyzed. Follow-up data were recorded for a minimum of 12 months after PET/CT. Radiomics features from each avid lesion in PET and low-dose CT images were extracted. A hybrid descriptive-inferential method and the discriminant analysis (DA) were used for feature selection and for predictive model implementation, respectively. The performance of the features in predicting PD was performed for per-lesion analysis, per-patient analysis, and liver lesions analysis. All lesions were again considered to assess the diagnostic performance of the features in discriminating liver lesions. In predicting PD in the whole group of patients, on PET features radiomics analysis, among per-lesion analysis, only the GLZLM_GLNU feature was selected, while three features were selected from PET/CT images data set. The same features resulted more accurately by associating CT features with PET features (AUROC 65.22%). In per-patient analysis, three features for stand-alone PET images and one feature (i.e., HUKurtosis) for the PET/CT data set were selected. Focusing on liver metastasis, in per-lesion analysis, the same analysis recognized one PET feature (GLZLM_GLNU) from PET images and three features from PET/CT data set. Similarly, in liver lesions per-patient analysis, we found three PET features and a PET/CT feature (HUKurtosis). In discrimination of liver metastasis from the rest of the other lesions, optimal results of stand-alone PET imaging were found for one feature (SUVbwmin; AUROC 88.91%) and two features for merged PET/CT features analysis (AUROC 95.33%). In conclusion, our machine learning model on restaging [18F]FDGPET/CT was demonstrated to be feasible and potentially useful in the predictive evaluation of disease progression in MCC.