A Neoteric Feature Extraction Technique to Predict the Survival of Gastric Cancer Patients

BACKGROUND

At the time of cancer diagnosis, it is crucial to accurately classify malignant gastric tumors and the possibility that patients will survive.

OBJECTIVE

This study aims to investigate the feasibility of identifying and applying a new feature extraction technique to predict the survival of gastric cancer patients.

METHODS

A retrospective dataset including the computed tomography (CT) images of 135 patients was assembled. Among them, 68 patients survived longer than three years. Several sets of radiomics features were extracted and were incorporated into a machine learning model, and their classification performance was characterized. To improve the classification performance, we further extracted another 27 texture and roughness parameters with 2484 superficial and spatial features to propose a new feature pool. This new feature set was added into the machine learning model and its performance was analyzed. To determine the best model for our experiment, Random Forest (RF) classifier, Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Naïve Bayes (NB) (four of the most popular machine learning models) were utilized. The models were trained and tested using the five-fold cross-validation method.

RESULTS

Using the area under ROC curve (AUC) as an evaluation index, the model that was generated using the new feature pool yields AUC = 0.98 ± 0.01, which was significantly higher than the models created using the traditional radiomics feature set (p < 0.04). RF classifier performed better than the other machine learning models.

CONCLUSIONS

This study demonstrated that although radiomics features produced good classification performance, creating new feature sets significantly improved the model performance.

Applying machine-learning models to differentiate benign and malignant thyroid nodules classified as C-TIRADS 4 based on 2D-ultrasound combined with five contrast-enhanced ultrasound key frames

Objectives

To apply machine learning to extract radiomics features from thyroid two-dimensional ultrasound (2D-US) combined with contrast-enhanced ultrasound (CEUS) images to classify and predict benign and malignant thyroid nodules, classified according to the Chinese version of the thyroid imaging reporting and data system (C-TIRADS) as category 4.

Materials and methods

This retrospective study included 313 pathologically diagnosed thyroid nodules (203 malignant and 110 benign). Two 2D-US images and five CEUS key frames ("2nd second after the arrival time" frame, "time to peak" frame, "2nd second after peak" frame, "first-flash" frame, and "second-flash" frame) were selected to manually label the region of interest using the "Labelme" tool. A total of 7 images of each nodule and their annotates were imported into the Darwin Research Platform for radiomics analysis. The datasets were randomly split into training and test cohorts in a 9:1 ratio. Six classifiers, namely, support vector machine, logistic regression, decision tree, random forest (RF), gradient boosting decision tree and extreme gradient boosting, were used to construct and test the models. Performance was evaluated using a receiver operating characteristic curve analysis. The area under the curve (AUC), sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy (ACC), and F1-score were calculated. One junior radiologist and one senior radiologist reviewed the 2D-US image and CEUS videos of each nodule and made a diagnosis. We then compared their AUC and ACC with those of our best model.

Results

The AUC of the diagnosis of US, CEUS and US combined CEUS by junior radiologist and senior radiologist were 0.755, 0.750, 0.784, 0.800, 0.873, 0.890, respectively. The RF classifier performed better than the other five, with an AUC of 1 for the training cohort and 0.94 (95% confidence interval 0.88-1) for the test cohort. The sensitivity, specificity, accuracy, PPV, NPV, and F1-score of the RF model in the test cohort were 0.82, 0.93, 0.90, 0.85, 0.92, and 0.84, respectively. The RF model with 2D-US combined with CEUS key frames achieved equivalent performance as the senior radiologist (AUC: 0.94 vs. 0.92, P = 0.798; ACC: 0.90 vs. 0.92) and outperformed the junior radiologist (AUC: 0.94 vs. 0.80, P = 0.039, ACC: 0.90 vs. 0.81) in the test cohort.

Conclusions

Our model, based on 2D-US and CEUS key frames radiomics features, had good diagnostic efficacy for thyroid nodules, which are classified as C-TIRADS 4. It shows promising potential in assisting less experienced junior radiologists.

TMM: A comprehensive CAD system for hepatic fibrosis 5-grade METAVIR staging based on liver MRI

BACKGROUND

Precise staging of hepatic fibrosis with MRI is necessary as it can assist precision medicine. Computer aided diagnosis (CAD) system with distinguishing radiomics features and radiologists domain knowledge is expected to obtain 5-grade meta-analysis of histological data in viral hepatitis (METAVIR) staging.

PURPOSE

This study aims to obtain the precise staging of hepatic fibrosis based on MRI as it predicts the risk of future liver-related morbidity and the need for treatment, monitoring and surveillance. Based on METAVIR score, fibrosis can be divided into five stages. Most previous researches focus on binary classification, such as cirrhosis versus non-cirrhosis, early versus advanced fibrosis, and substantial fibrosis or not. In this paper, a comprehensive CAD system TMM is proposed to precisely class hepatic fibrosis into five stages for precision medicine instead of the common binary classification.

METHODS

We propose a novel hepatic fibrosis staging CAD system TMM which includes three modules, Two-level Image Statistical Radiomics Feature (TISRF), Monotonic Error Correcting Output Codes (MECOC) and Monotone Multiclassification with Deep Forest (MMDF). TISRF extracts radiomics features for distinguishing different hepatic fibrosis stages. MECOC is proposed to encode monotonic multiclass by making full use of the progressive severity of hepatic fibrosis and increase the fault tolerance and error correction ability. MMDF combines multiple Deep Forest network to ensure the final five-class classification, which can achieve more precise classification than the common binary classification. The performance of the proposed hepatic fibrosis CAD system is tested on the hepatic data collected from our rabbits models of fibrosis.

RESULTS

A total of 140 regions of interest (ROI) are selected from MRI T1W of liver fibrosis models in 35 rabbits with F0(n = 16), F1(n = 28), F2(n = 29), F3(n = 44) and F4(n = 23). The performance is evaluated by five-fold cross-validation. TMM can achieve the highest total accuracy of 72.14% for five fibrosis stages compared with other popular classifications. To make a comprehensive comparison, a binary classification experiment have been carried out.

CONCLUSIONS

T1WI can obtain precise staging of hepatic fibrosis with the help of comprehensive CAD including radiomics features extraction inspired by radiologists, monotonic multiclass according to the severity of hepatic fibrosis, and deep learning classification.

Diagnosis of cognitive and motor disorders levels in stroke patients through explainable machine learning based on MRI

BACKGROUND

Globally, stroke is the third most significant cause of disability. A stroke may produce motor, sensory, perceptual, or cognitive disorders that result in disability and affect the likelihood of recovery, affecting a person's ability to function. Evaluation post-stroke is critical for optimal stroke care.

PURPOSE

Traditional methods for classifying the clinical disorders of cognitive and motor in stroke patients use assessment and interrogative measures, which are time-consuming, complex, and labor-intensive. In response to the current situation, this study develops an algorithm to automatically classify motor and cognitive disorders in stroke patients by 3D brain MRI to assist physicians in diagnosis.

METHODS

First, radiomics and fusion features are extracted from the OAx T2 Propeller of 3D brain MRI. Then, we use 14 machine learning models and one model ensemble method to predict Fugl-Meyer and MMSE levels of stroke patients. Next, we evaluate the models using accuracy, recall, f1-score, and area under the curve (AUC). Finally, we employ SHAP to explain the output of the model.

RESULTS

The best predictive models come from Random Forest (RF) Classifier with fusion features in cognitive classification and Linear Discriminant Analysis (LDA) with radiomics features in motor classification. The highest accuracies are 92.0 and 82.5% for cognitive and motor disorders.

CONCLUSIONS

MRI brain maps can classify the cognitive and motor disorders of stroke patients. Radiomics features demonstrate its merits. The proposed algorithms with MRI images can efficiently assist physicians in diagnosing the cognitive and motor disorders of stroke patients in clinical practice. Additionally, this lessens labor costs, improves diagnostic effectiveness, and avoids the subjective difference that comes with manual assessment.

Fast prediction of patient-specific organ doses in brain CT scans using support vector regression algorithm

Objectives. This study aims to develop a method for predicting patient-specific head organ doses by training a support vector regression (SVR) model based on radiomics features and graphics processing unit (GPU)-calculated reference doses.Methods. In this study, 237 patients who underwent brain CT scans were selected, and their CT data were transferred to an autosegmentation software to segment head regions of interest (ROIs). Subsequently, radiomics features were extracted from the CT data and ROIs, and the benchmark organ doses were computed using fast GPU-accelerated Monte Carlo (MC) simulations. The SVR organ dose prediction model was then trained using the radiomics features and benchmark doses. For the predicted organ doses, the relative root mean squared error (RRMSE), mean absolute percentage error (MAPE), and coefficient of determination (R2) were evaluated. The robustness of organ dose prediction was verified by changing the patient samples on the training and test sets randomly.Results. For all head organs, the maximal difference between the reference and predicted dose was less than 1 mGy. For the brain, the organ dose was predicted with an absolute error of 1.3%, and theR2reached up to 0.88. For the eyes and lens, the organ doses predicted by SVR achieved an RRMSE of less than 13%, the MAPE ranged from 4.5% to 5.5%, and theR2values were more than 0.7.Conclusions. Patient-specific head organ doses from CT examinations can be predicted within one second with high accuracy, speed, and robustness by training an SVR using radiomics features.

Multimodality radiomics prediction of radiotherapy-induced the early proctitis and cystitis in rectal cancer patients: a machine learning study

Purpose.This study aims to predict radiotherapy-induced rectal and bladder toxicity using computed tomography (CT) and magnetic resonance imaging (MRI) radiomics features in combination with clinical and dosimetric features in rectal cancer patients.Methods.A total of sixty-three patients with locally advanced rectal cancer who underwent three-dimensional conformal radiation therapy (3D-CRT) were included in this study. Radiomics features were extracted from the rectum and bladder walls in pretreatment CT and MR-T2W-weighted images. Feature selection was performed using various methods, including Least Absolute Shrinkage and Selection Operator (Lasso), Minimum Redundancy Maximum Relevance (MRMR), Chi-square (Chi2), Analysis of Variance (ANOVA), Recursive Feature Elimination (RFE), and SelectPercentile. Predictive modeling was carried out using machine learning algorithms, such as K-nearest neighbor (KNN), Support Vector Machine (SVM), Logistic Regression (LR), Decision Tree (DT), Random Forest (RF), Naive Bayes (NB), Gradient Boosting (XGB), and Linear Discriminant Analysis (LDA). The impact of the Laplacian of Gaussian (LoG) filter was investigated with sigma values ranging from 0.5 to 2. Model performance was evaluated in terms of the area under the receiver operating characteristic curve (AUC), accuracy, precision, sensitivity, and specificity.Results.A total of 479 radiomics features were extracted, and 59 features were selected. The pre-MRI T2W model exhibited the highest predictive performance with an AUC: 91.0/96.57%, accuracy: 90.38/96.92%, precision: 90.0/97.14%, sensitivity: 93.33/96.50%, and specificity: 88.09/97.14%. These results were achieved with both original image and LoG filter (sigma = 0.5-1.5) based on LDA/DT-RF classifiers for proctitis and cystitis, respectively. Furthermore, for the CT data, AUC: 90.71/96.0%, accuracy: 90.0/96.92%, precision: 88.14/97.14%, sensitivity: 93.0/96.0%, and specificity: 88.09/97.14% were acquired. The highest values were achieved using XGB/DT-XGB classifiers for proctitis and cystitis with LoG filter (sigma = 2)/LoG filter (sigma = 0.5-2), respectively. MRMR/RFE-Chi2 feature selection methods demonstrated the best performance for proctitis and cystitis in the pre-MRI T2W model. MRMR/MRMR-Lasso yielded the highest model performance for CT.Conclusion.Radiomics features extracted from pretreatment CT and MR images can effectively predict radiation-induced proctitis and cystitis. The study found that LDA, DT, RF, and XGB classifiers, combined with MRMR, RFE, Chi2, and Lasso feature selection algorithms, along with the LoG filter, offer strong predictive performance. With the inclusion of a larger training dataset, these models can be valuable tools for personalized radiotherapy decision-making.

A multi-stage fusion framework to classify breast lesions using deep learning and radiomics features computed from four-view mammograms

BACKGROUND

Developing computer aided diagnosis (CAD) schemes of mammograms to classify between malignant and benign breast lesions has attracted a lot of research attention over the last several decades. However, unlike radiologists who make diagnostic decisions based on the fusion of image features extracted from multi-view mammograms, most CAD schemes are single-view-based schemes, which limit CAD performance and clinical utility.

PURPOSE

This study aims to develop and test a novel CAD framework that optimally fuses information extracted from ipsilateral views of bilateral mammograms using both deep transfer learning (DTL) and radiomics feature extraction methods.

METHODS

An image dataset containing 353 benign and 611 malignant cases is assembled. Each case contains four images: the craniocaudal (CC) and mediolateral oblique (MLO) view of the left and right breast. First, we extract four matching regions of interest (ROIs) from images that surround centers of two suspicious lesion regions seen in CC and MLO views, as well as matching ROIs in the contralateral breasts. Next, the handcrafted radiomics (HCRs) features and VGG16 model-generated automated features are extracted from each ROI resulting in eight feature vectors. Then, after reducing feature dimensionality and quantifying the bilateral and ipsilateral asymmetry of four ROIs to yield four new feature vectors, we test four fusion methods to build three support vector machine (SVM) classifiers by an optimal fusion of asymmetrical image features extracted from four view images.

RESULTS

Using a 10-fold cross-validation method, results show that a SVM classifier trained using an optimal fusion of four view images yields the highest classification performance (AUC = 0.876 ± 0.031), which significantly outperforms SVM classifiers trained using one projection view alone, AUC = 0.817 ± 0.026 and 0.792 ± 0.026 for the CC and MLO view of bilateral mammograms, respectively (p < 0.001).

CONCLUSIONS

The study demonstrates that the shift from single-view CAD to four-view CAD and the inclusion of both DTL and radiomics features significantly increases CAD performance in distinguishing between malignant and benign breast lesions.

Presurgical MRI-Based Radiomics Models for Predicting Cerebellar Mutism Syndrome in Children With Posterior Fossa Tumors

BACKGROUND

Current studies have indicated that tumoral morphologic features are associated with cerebellar mutism syndrome (CMS), but the radiomics application in CMS is scarce.

PURPOSE

To develop a model for CMS discrimination based on multiparametric MRI radiomics in patients with posterior fossa tumors.

STUDY TYPE

Retrospective.

POPULATION

A total of 218 patients (males 132, females 86) with posterior fossa tumors, 169 of which were included in the MRI radiomics analysis. The MRI radiomics study cohort (169) was split into training (119) and testing (50) sets with a ratio of 7:3.

FIELD/SEQUENCE

All the MRI were acquired under 1.5/3.0 T scanners. T2-weighted image (T2W), T1-weighted (T1W), fluid attenuated inversion recovery (FLAIR), diffusion-weighted imaging (DWI).

ASSESSMENT

Apparent diffusion coefficient (ADC) maps were generated from DWI. Each MRI dataset generated 1561 radiomics characteristics. Feature selection was performed with univariable logistic analysis, correlation analysis, and least absolute shrinkage and selection operator (LASSO) penalized logistic regression. Significant clinical features were selected with multivariable logistic analysis and used to constructed the clinical model. Radiomics models (based on T1W, T2W, FLAIR, DWI, ADC) were constructed with selected radiomics features. The mix model was based on the multiparametric MRI radiomics features.

STATISTICAL TEST

Multivariable logistic analysis was utilized during clinical features selection. Models' performance was evaluated using the area under the receiver operating characteristic (AUC) curve. Interobserver variability was assessed using Cohen's kappa. Significant threshold was set as P < 0.05.

RESULTS

Sex (aOR = 3.72), tumor location (aOR = 2.81), hydrocephalus (aOR = 2.14), and tumor texture (aOR = 5.08) were significant features in the multivariable analysis and were used to construct the clinical model (AUC = 0.79); totally, 33 radiomics features were selected to construct radiomics models (AUC = 0.63-0.93). Seven of the 33 radiomics features were selected for the mix model (AUC = 0.93).

DATA CONCLUSION

Multiparametric MRI radiomics may be better at predicting CMS than single-parameter MRI models and clinical model.

EVIDENCE LEVEL

4.

TECHNICAL EFFICACY

2.

Value of radiomics in differentiating synchronous double primary lung adenocarcinomas from intrapulmonary metastasis

BACKGROUND

Distinguishing synchronous double primary lung adenocarcinoma (SDPLA) from intrapulmonary metastasis (IPM) of lung cancer has significant therapeutic and prognostic values. This study aimed to develop and validate a CT-based radiomics model to differentiate SDPLA from IPM.

METHODS

A total of 153 patients (93 SDPLA and 60 IPM) with 306 pathologically confirmed lesions were retrospectively studied. CT morphological features were also recorded. Region of interest (ROI) segmentation was performed semiautomatically, and 1,037 radiomics features were extracted from every segmented lesion The differences of radiomics features were defined as the relative net difference in radiomics features between the two lesions on CT. Those low reliable (ICC <0.75) and redundant (r>0.9) features were excluded by intraclass correlation coefficients (ICC) and Pearson's correlation. Multivariate logistic regression (LR) algorithm was used to establish the classification model according to the selected features. The radiomics model was based on the four most contributing differences of radiomics features. Clinical-CT model and MixModel were based on selected clinical and CT features only and the combination of clinical-CT and Rad-score, respectively.

RESULTS

In both the training and testing cohorts, the area under the curves (AUCs) of the radiomics model were larger than those of the clinical-CT model (0.944 vs. 0.793 and 0.886 vs. 0.735 on training and testing cohorts, respectively), and statistically significant differences between the two models in the testing set were found (P<0.001). Meanwhile, three radiologists had sensitivities of 84.2%, 63.9%, and 68.4%, and specificities of 76.9%, 69.2%, and 76.9% in differentiating 19 SDPLA cases from 13 cases of IPM in the testing set. Compared with the performance of the three radiologists, the radiomics model showed better accuracy to the patients in both the training and testing cohorts. Among the three models, the radiomics model showed the best net benefits.

CONCLUSIONS

The differences of radiomics features showed excellent diagnostic performance for preoperative differentiation between synchronous double primary lung adenocarcinoma from interpulmonary metastasis, superior to the clinical model and decisions made by radiologists.

Application of novel hybrid machine learning systems and radiomics features for non-motor outcome prediction in Parkinson's disease

Objectives.Parkinson's disease (PD) is a complex neurodegenerative disorder, affecting 2%-3% of the elderly population. Montreal Cognitive Assessment (MoCA), a rapid nonmotor screening test, assesses different cognitive dysfunctionality aspects. Early MoCA prediction may facilitate better temporal therapy and disease control. Radiomics features (RF), in addition to clinical features (CF), are indicated to increase clinical diagnoses, etc, bridging between medical imaging procedures and personalized medicine. We investigate the effect of RFs, CFs, and conventional imaging features (CIF) to enhance prediction performance using hybrid machine learning systems (HMLS).Methods.We selected 210 patients with 981 features (CFs, CIFs, and RFs) from the Parkinson's Progression-Markers-Initiative database. We generated 4 datasets, namely using (i), (ii) year-0 (D1) or year-1 (D2) features, (iii) longitudinal data (D3, putting datasets in years 0 and 1 longitudinally next to each other), and (iv) timeless data (D4, effectively doubling dataset size by listing both datasets from years 0 and 1 separately). First, we directly applied 23 predictor algorithms (PA) to the datasets to predict year-4 MoCA, which PD patients this year have a higher dementia risk. Subsequently, HMLSs, including 14 attribute extraction and 10 feature selection algorithms followed by PAs were employed to enhance prediction performances. 80% of all datapoints were utilized to select the best model based on minimum mean absolute error (MAE) resulting from 5-fold cross-validation. Subsequently, the remaining 20% was used for hold-out testing of the selected models.Results.When applying PAs without ASAs/FEAs to datasets (MoCA outcome range: [11,30]), Adaboost achieved an MAE of 1.74 ± 0.29 on D4 with a hold-out testing performance of 1.71. When employing HMLSs, D4 + Minimum_Redundancy_Maximum_Relevance (MRMR)+K_Nearest_Neighbor Regressor achieved the highest performance of 1.05 ± 0.25 with a hold-out testing performance of 0.57.Conclusion.Our study shows the importance of using larger datasets (timeless), and utilizing optimized HMLSs, for significantly improved prediction of MoCA in PD patients.

Clinical-radiomic analysis for non-invasive prediction of liver steatosis on non-contrast CT: A pilot study

Purpose: Our aim is to build and validate a clinical-radiomic model for non-invasive liver steatosis prediction based on non-contrast computed tomography (CT). Methods: We retrospectively reviewed 342 patients with suspected NAFLD diagnoses between January 2019 and July 2020 who underwent non-contrast CT and liver biopsy. Radiomics features from hepatic and splenic regions-of-interests (ROIs) were extracted based on abdominal non-contrast CT imaging. The radiomics signature was constructed based on reproducible features by adopting the least absolute shrinkage and selection operator (LASSO) regression. Then, multivariate logistic regression analysis was applied to develop a combined clinical-radiomic nomogram integrating radiomics signature with several independent clinical predictors in a training cohort of 124 patients between January 2019 and December 2019. The performance of models was determined by the area under the receiver operating characteristic curves and calibration curves. We conducted an internal validation during 103 consecutive patients between January 2020 and July 2020. Results: The radiomics signature was composed of four steatosis-related features and positively correlated with pathologic liver steatosis grade (p < 0.01). In both subgroups (Group One, none vs. steatosis; Group Two, none/mild vs. moderate/severe steatosis), the clinical-radiomic model performed best within the validation cohort with an AUC of 0.734 and 0.930, respectively. The calibration curve confirmed the concordance of excellent models. Conclusion: We developed a robust clinical-radiomic model for accurate liver steatosis stage prediction in a non-invasive way, which may improve the clinical decision-making ability.

Novel Survival Features Generated by Clinical Text Information and Radiomics Features May Improve the Prediction of Ischemic Stroke Outcome

Background: Accurate outcome prediction is of great clinical significance in customizing personalized treatment plans, reducing the situation of poor recovery, and objectively and accurately evaluating the treatment effect. This study intended to evaluate the performance of clinical text information (CTI), radiomics features, and survival features (SurvF) for predicting functional outcomes of patients with ischemic stroke. Methods: SurvF was constructed based on CTI and mRS radiomics features (mRSRF) to improve the prediction of the functional outcome in 3 months (90-day mRS). Ten machine learning models predicted functional outcomes in three situations (2-category, 4-category, and 7-category) using seven feature groups constructed by CTI, mRSRF, and SurvF. Results: For 2-category, ALL (CTI + mRSRF+ SurvF) performed best, with an mAUC of 0.884, mAcc of 0.864, mPre of 0.877, mF1 of 0.86, and mRecall of 0.864. For 4-category, ALL also achieved the best mAuc of 0.787, while CTI + SurvF achieved the best score with mAcc = 0.611, mPre = 0.622, mF1 = 0.595, and mRe-call = 0.611. For 7-category, CTI + SurvF performed best, with an mAuc of 0.788, mPre of 0.519, mAcc of 0.529, mF1 of 0.495, and mRecall of 0.47. Conclusions: The above results indicate that mRSRF + CTI can accurately predict functional outcomes in ischemic stroke patients with proper machine learning models. Moreover, combining SurvF will improve the prediction effect compared with the original features. However, limited by the small sample size, further validation on larger and more varied datasets is necessary.

Elaboration of Multiparametric MRI-Based Radiomics Signature for the Preoperative Quantitative Identification of the Histological Grade in Patients With Non-Small-Cell Lung Cancer

BACKGROUND

The histological grading plays an essential role in the treatment decision of lung cancer. Detected tumors are usually biopsied to confirm histologic grade. How to use MRI extracted radiomics features for accurately grading lung cancer is still challenging.

PURPOSE

To examine the diagnostic utility of multiparametric MRI radiomics and clinical factors for grading non-small-cell lung cancer (NSCLC).

STUDY TYPE

Retrospective.

POPULATION

A total of 148 patients (25.7% female) with postoperative pathologically confirmed NSCLC and divided into the training cohort (N = 110) and the validation cohort (N = 38).

FIELD STRENGTH/SEQUENCE

A 1.5 T; single-shot turbo spin-echo (TSE), T2-weighted imaging (T2WI), and integrated shimming-echo planar imaging (ISHIM-EPI) diffusion-weighted imaging (DWI).

ASSESSMENT

A total of 2775 radiomics features were extracted from carcinomatous regions of interest on T2WI, DWI, and the apparent diffusion coefficient (ADC) maps. The five optimal features were selected by using the Student' s t-test, the least absolute shrinkage and selection operator (LASSO) and stepwise regression. The Radscore combined with clinical factors, which selected by univariate and multivariate analyses, to develop a radiomics-clinical nomogram. Its performance was evaluated in the training cohort and the validation cohort. The potential clinical usefulness was analyzed by the receiver operating characteristic curve (ROC), area under the curve (AUC), and the Hosmer-Lemeshow test.

STATISTICAL TESTS

Student's t-test, univariate analyses, multivariate analyses, LASSO, ROC, AUC, and the Hosmer-Lemeshow test. P < 0.05 was considered statistically significant.

RESULTS

Favorable discrimination performance was obtained for five optimal features (out of the 2775 features), using the training cohorts (AUC 0.761) and validation cohorts (AUC 0.753). In addition, the radiomics-clinical nomogram significantly improved the ability to identify histological grades in the training cohort (AUC 0.814) and the validation cohort (AUC 0.767).

DATA CONCLUSIONS

The radiomics-clinical nomogram based on multiparametric MRI might have the potential to distinguish the histological grade of NSCLC.

EVIDENCE LEVEL

3 TECHNICAL EFFICACY: Stage 2.

Exploring the Value of Radiomics Features Based on B-Mode and Contrast-Enhanced Ultrasound in Discriminating the Nature of Thyroid Nodules

Background

With the improvement of ultrasound imaging resolution and the application of various new technologies, the detection rate of thyroid nodules has increased greatly in recent years. However, there are still challenges in accurately diagnosing the nature of thyroid nodules. This study aimed to evaluate the clinical application value of the radiomics features extracted from B-mode ultrasound (B-US) images combined with contrast-enhanced ultrasound (CEUS) images in the differentiation of benign and malignant thyroid nodules by comparing the diagnostic performance of four logistic models.

Methods

We retrospectively collected and ultimately included B-US images and CEUS images of 123 nodules from 123 patients, and then extracted the corresponding radiomics features from these images respectively. Meanwhile, a senior radiologist combined the thyroid imaging reporting and data system (TI-RADS) and the enhancement pattern of the ultrasonography to make a graded diagnosis of the malignancy of these nodules. Next, based on these radiomics features and grades, logistic regression was used to help build the models (B-US radiomics model, CEUS radiomics model, B-US+CEUS radiomics model, and TI-RADS+CEUS model). Finally, the study assessed the diagnostic performance of these radiomics features with a comparison of the area under the curve (AUC) of the receiver operating characteristic curve of four logistic models for predicting the benignity or malignancy of thyroid nodules.

Results

The AUC in the differential diagnosis of the nature of thyroid nodules was 0.791 for the B-US radiomics model, 0.766 for the CEUS radiomics model, 0.861 for the B-US+CEUS radiomics model, and 0.785 for the TI-RADS+CEUS model. Compared to the TI-RADS+CEUS model, there was no statistical significance observed in AUC between the B-US radiomics model, CEUS radiomics model, B-US+CEUS radiomics model, and TI-RADS+CEUS model (P>0.05). However, a significant difference was observed between the single B-US radiomics model or CEUS radiomics model and B-US+CEUS radiomics model (P<0.05).

Conclusion

In our study, the B-US radiomics model, CEUS radiomics model, and B-US+CEUS radiomics model demonstrated similar performance with the TI-RADS+CEUS model of senior radiologists in diagnosing the benignity or malignancy of thyroid nodules, while the B-US+CEUS radiomics model showed better diagnostic performance than single B-US radiomics model or CEUS radiomics model. It was proved that B-US radiomics features and CEUS radiomics features are of high clinical value as the combination of the two had better diagnostic performance.

[Predicting postoperative recurrence of stage Ⅰ-Ⅲ renal clear cell carcinoma based on preoperative CT radiomics feature nomogram]

OBJECTIVE

To explore the preoperative radiomics features (RFs) and construct a nomogram for predicting postoperative recurrence of stage Ⅰ-Ⅲ clear cell renal carcinoma (ccRCC).

METHODS

The clinicopathological data and preoperative enhanced CT images collected from 256 patients with ccRCC were used as the training dataset (175 patients) and test dataset (81 patients). The enhanced CT images of the tumor were segmented using ITK-SNAP software, and the RFs were extracted using the PyRadiomics computing platform. In the training dataset, the RFs were screened based on Lasso-CV algorithm, and the Rad_score was calculated. The Clinic factors were screened by univariate and multivariate logistic regression analysis of the clinical and pathological factors and CT characteristics. The Rad_score, Clinic、Rad_score + Clinic nomograms were constructed and verified using the test dataset. The performance, discrimination power and calibration of the nomograms were compared, and their clinical value was evaluated using decision curve analysis.

RESULTS

Six RFs were retained to calculate the Rad_score. The Clinic factors included Rad_score, KPS score, platelet, calcification and TNM clinical stage. In terms of discrimination, the Rad_score + Clinic nomogram showed better performance (AUC=0.84 for training set; AUC=0.85 for test set) than the Rad_score nomogram (AUC=0.78 for training set, P=0.029; AUC=0.77 for Test set, P=0.025) and Clinic nomogram (AUC=0.77 for training set, P=0.014; AUC=0.77 for test set, P=0.011). In terms of calibration, the P value for goodness of fit test of the Rad_score+Clinic nomogram was 0.065 for the training set and 0.628 for the test set. Decision curve analysis showed a greater clinical value of the Rad_score+Clinic nomogram with Rad_score than the Clinic nomogram without Rad_score.

CONCLUSION

The nomogram based on preoperative CT RFs has a high value for predicting postoperative recurrence of stage Ⅰ-Ⅲ ccRCC to facilitate individualized treatment of RCC.

Radiomics Study for Predicting the Expression of PD-L1 and Tumor Mutation Burden in Non-Small Cell Lung Cancer Based on CT Images and Clinicopathological Features

Background

The present study compared the predictive performance of pretreatment computed tomography (CT)-based radiomics signatures and clinicopathological and CT morphological factors for ligand programmed death-ligand 1 (PD-L1) expression level and tumor mutation burden (TMB) status and further explored predictive models in patients with advanced-stage non-small cell lung cancer (NSCLC).

Methods

A total of 120 patients with advanced-stage NSCLC were enrolled in this retrospective study and randomly assigned to a training dataset or validation dataset. Here, 462 radiomics features were extracted from region-of-interest (ROI) segmentation based on pretreatment CT images. The least absolute shrinkage and selection operator (LASSO) and logistic regression were applied to select radiomics features and develop combined models with clinical and morphological factors for PD-L1 expression and TMB status prediction. Ten-fold cross-validation was used to evaluate the accuracy, and the predictive performance of these models was assessed using receiver operating characteristic (ROC) and area under the curve (AUC) analyses.

Results

The PD-L1-positive expression level correlated with differentiation degree (p = 0.005), tumor shape (p = 0.006), and vascular convergence (p = 0.007). Stage (p = 0.023), differentiation degree (p = 0.017), and vacuole sign (p = 0.016) were associated with TMB status. Radiomics signatures showed good performance for predicting PD-L1 and TMB with AUCs of 0.730 and 0.759, respectively. Predictive models that combined radiomics signatures with clinical and morphological factors dramatically improved the predictive efficacy for PD-L1 (AUC = 0.839) and TMB (p = 0.818). The results were verified in the validation datasets.

Conclusions

Quantitative CT-based radiomics features have potential value in the classification of PD-L1 expression levels and TMB status. The combined model further improved the predictive performance and provided sufficient information for the guiding of immunotherapy in clinical practice, and it deserves further analysis.

Morphology-based radiomics signature: a novel determinant to identify multiple intracranial aneurysms rupture

We aimed to develop and validate a morphology-based radiomics signature nomogram for assessing the risk of intracranial aneurysm (IA) rupture. A total of 254 aneurysms in 105 patients with subarachnoid hemorrhage and multiple intracranial aneurysms from three centers were retrospectively reviewed and randomly divided into the derivation and validation cohorts. Radiomics morphological features were automatically extracted from digital subtraction angiography and selected by the least absolute shrinkage and selection operator algorithm to develop a radiomics signature. A radiomics signature-based nomogram was developed by incorporating the signature and traditional morphological features. The performance of calibration, discrimination, and clinical usefulness of the nomogram was assessed. Ten radiomics morphological features were selected to build the radiomics signature model, which showed better discrimination with an area under the curve (AUC) equal to 0.814 and 0.835 in the derivation and validation cohorts compared with 0.747 and 0.666 in the traditional model, which only include traditional morphological features. When radiomics signature and traditional morphological features were combined, the AUC increased to 0.842 and 0.849 in the derivation and validation cohorts, thus showing better performance in assessing aneurysm rupture risk. This novel model could be useful for decision-making and risk stratification for patients with IAs.

Pretreatment CT-Based Radiomics Signature as a Potential Imaging Biomarker for Predicting the Expression of PD-L1 and CD8+TILs in ESCC

Background

The present study constructed and validated models to predict PD-L1 and CD8+TILs expression levels in esophageal squamous cell carcinoma (ESCC) patients using radiomics features and clinical factors.

Patients and Methods

This retrospective study randomly assigned 220 ESCC patients to a discovery dataset (n= 160) and validation dataset (n= 60). A total of 462 radiomics features were extracted from the segmentation of regions of interest (ROIs) based on pretreatment CT images of each patient. The LASSO algorithm was applied to reduce the dimensionality of the data and select features. A multivariable logistic regression analysis was adopted to build radiomics signatures. Receiver operating characteristic (ROC) curve analysis was performed to evaluate the predictive accuracy of these models.

Results

There was no significant difference between the training and validation datasets for any clinical factors in patients with ESCC. The PD-L1 expression level correlated with the differentiation degree (p= 0.011) and tumor stage (p= 0.032). Smoking status (p= 0.043) and differentiation degree (p= 0.025) were associated with CD8+TILs expression levels. The radiomics signatures achieved good performance in predicting PD-L1 and CD8+TILs with AUCs= 0.784 and 0.764, respectively. The combined model showed a favorable predictive ability compared to radiomics signatures or clinical factors alone and improved the AUCs from 0.669 to 0.871 for PD-L1 and from 0.672 to 0.832 for CD8+TILs. These results were verified in the validation dataset with the AUCs of 0.817 and 0.795, respectively.

Conclusion

CT-based radiomics features have a potential value for classifying patients according to PD-L1 and CD8+TILs expression levels. The combination of clinical factors and radiomics signatures significantly improved the predictive performance in ESCC.

Correlation between CT based radiomics features and gene expression data in non-small cell lung cancer

OBJECTIVE

Radiogenomics investigates radiographic imaging phenotypes associated with gene expression patterns. This study aims to explore relationships between CT imaging radiomics features and gene expression data in non-small cell lung cancer (NSCLC).

METHODS

Eighty-nine NSCLC patients are included in the study. Radiomics features are extracted and selected to quantify the phenotype of tumors on CT-scans. Co-expressed genes are also clustered and the first principal component of the cluster is represented, which is defined as a metagene. Then, statistical analysis was performed to assess association of CT radiomics features with metagenes. In addition, predictive models are built and metagene enrichment are conducted to further evaluate performance of NSCLC radiogenomics statistically and biologically.

RESULTS

There are 187 significant pairwise correlations between a CT radiomics feature and a metagene of NSCLC, where eighteen metagenes are annotated with Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) terms. Metagenes are predicted in terms of radiomics features with an accuracy of 41.89% -89.93%.

CONCLUSIONS

This study reveals the associations between CT imaging radiomics features and NSCLC co-expressed gene sets. The findings suggest that CT radiomics features can reflect important biological information of NSCLC patients, which may have a significant clinical impact as CT is routinely used in clinical practice, assisting in improving medical decision-support at low cost.