Tumor heterogeneity in gastrointestinal stromal tumors of the small bowel: volumetric CT texture analysis as a potential biomarker for risk stratification

Preliminary study on diagnosis of gallbladder neoplastic polyps based on contrast-enhanced ultrasound and grey scale ultrasound radiomics

Objective

Neoplastic gallbladder polyps (GPs), including adenomas and adenocarcinomas, are considered absolute indications for surgery; however, the distinction of neoplastic from non-neoplastic GPs on imaging is often challenging. This study thereby aimed to develop a CEUS radiomics nomogram, and evaluate the role of a combined grey-scale ultrasound and CEUS model for the prediction and diagnosis of neoplastic GPs.

Methods

Patients with GPs of ≥ 1 cm who underwent CEUS between January 2017 and May 2022 were retrospectively enrolled. Grey-scale ultrasound and arterial phase CEUS images of the largest section of the GPs were used for radiomics feature extraction. Features with good reproducibility in terms of intraclass correlation coefficient were selected. Grey-scale ultrasound and CEUS Rad-score models were first constructed using the Mann-Whitney U and LASSO regression test, and were subsequently included in the multivariable logistic regression analysis as independent factors for construction of the combined model.

Results

A total of 229 patients were included in our study. Among them, 118 cholesterol polyps, 68 adenomas, 33 adenocarcinomas, 6 adenomyomatoses, and 4 inflammatory polyps were recorded. A total of 851 features were extracted from each patient. Following screening, 21 and 15 features were retained in the grey-scale and CEUS models, respectively. The combined model demonstrated AUCs of 0.88 (95% CI: 0.83 - 0.93) and 0.84 (95% CI: 0.74 - 0.93) in the training and testing set, respectively. When applied to the whole dataset, the combined model detected 111 of the 128 non-neoplastic GPs, decreasing the resection rate of non-neoplastic GPs to 13.3%.

Conclusion

Our proposed combined model based on grey-scale ultrasound and CEUS radiomics features carries the potential as a non-invasive, radiation-free, and reproducible tool for the prediction and identification of neoplastic GPs. Our model may not only guide the treatment selection for GPs, but may also reduce the surgical burden of such patients.

Comparing endoscopic ultrasonography and double contrast-enhanced ultrasonography in the preoperative diagnosis of gastric stromal tumor

BACKGROUND

This study was designed to perform a comparative analysis between endoscopic ultrasonography (EUS) and double contrast-enhanced ultrasonography (DCEUS) for the preoperative diagnosis of gastric stromal tumors (GSTs).

METHODS

A retrospective study was conducted involving 139 patients with histologically confirmed GSTs. All patients preoperatively underwent DCEUS and EUS. The pathology reports were treated as the baseline and were retrospectively compared with the findings of EUS and DCEUS.

RESULTS

Of the 139 lesions, 120 and 113 were correctly identified by DCEUS and EUS, respectively, with an accuracy of 86.3% and 81.3%. The results revealed an insignificant difference between these two methods (p = 0.189).

CONCLUSIONS

DCEUS can display not only the locations, sizes, shapes, borders, internal echoes, but also show the blood perfusion patterns of GSTs. It is a highly accurate, noninvasive, and convenient method to be used at the pre-treatment stage.

[Update: Small bowel diseases in computed tomography and magnetic resonance imaging]

CLINICAL/METHODICAL ISSUE

Radiological procedures play a crucial role in the diagnosis of small bowel disease. Due to a broad and quite nonspecific spectrum of symptoms, clinical evaluation is often difficult, and endoscopic procedures require significant manpower, are time-consuming and expensive. In contrast, radiologic imaging can provide important information about morphologic and functional variations of the small bowel and help to identify various disease entities, such as inflammation, tumors, vascular problems, and obstruction.

STANDARD RADIOLOGICAL METHODS

The most common radiological modalities in small bowel diagnostics include ultrasound (US), computed tomography (CT), magnetic resonance imaging (MRI), and fluoroscopy. Each of these modalities has its own advantages and limitations, and the choice of imaging modality depends on clinical symptoms and suspected diagnosis in addition to availability.

METHODOLOGICAL INNOVATIONS

In recent years, significant progress has been made, especially in cross-sectional imaging modalities, as a result of new and further technical developments.

PERFORMANCE

These range from increasing detail resolution to functional and molecular imaging techniques that go far beyond simple morphology. In addition, information technology (IT) applications, which include artificial intelligence and radiomics, are assuming an increasing role.

ACHIEVEMENTS

Many of the methods mentioned are still in early stages and need to be further developed for daily practice, but some have already found their way into clinical routine.

PRACTICAL RECOMMENDATIONS

The aim of this work is to provide a review of the most important disease entities of the small intestine, including new and innovative diagnostic approaches.

Application of radiomics in predicting the preoperative risk stratification of gastric stromal tumors

PURPOSE The stomach is the most common site of gastrointestinal stromal tumors (GISTs). In this study, clinical model, radiomics models, and nomogram were constructed to compare and assess the clinical value of each model in predicting the preoperative risk stratification of gastric stromal tumors (GSTs). METHODS In total, 180 patients with GSTs confirmed postoperatively pathologically were included. 70% was randomly selected from each category as the training group (n = 126), and the remaining 30% was stratified as the testing group (n = 54). The image features and texture characteristics of each patient were analyzed, and predictive model were constructed. The image features and the rad-score of the optimal radiomics model were used to establish the nomogram. The clinical application value of these models was assessed by the receiver operating characteristic (ROC) curve and decision curve analysis (DCA). The calibration of each model was evaluated by the calibration curve. RESULTS The Area Under the Curve (AUC) value of the nomogram was 0.930 (95% confidence interval [CI]: 0.886- 0.973) in the training group and 0.931 (95% CI: 0.869-0.993) in the testing group. The AUC values of the training group and the testing group calculated by the radiomics model were 0.874 (95% CI: 0.814-0.935) and 0.863 (95% CI: 0.76 5-0.960), respectively; the AUC values calculated by the clinical model were 0.871 (95% CI: 0.811-0.931) and 0.854 (95% CI: 0.76 0-0.947). CONCLUSION The proposed nomogram can accurately predict the malignant potential of GSTs and can be used as repeatable imaging markers for decision support to predict the risk stratification of GSTs before surgery noninvasively and effectively.

Building contrast-enhanced CT-based models for preoperatively predicting malignant potential and Ki67 expression of small intestine gastrointestinal stromal tumors (GISTs)

PURPOSE

To assess contrast-enhanced computed tomography (CE-CT) features for predicting malignant potential and Ki67 in small intestinal gastrointestinal stromal tumors (GISTs) and the correlation between them.

METHODS

We retrospectively analyzed the pathological and imaging data for 123 patients (55 male/68 female, mean age: 57.2 years) with a histopathological diagnosis of small intestine GISTs who received CE-CT followed by curative surgery from May 2009 to August 2019. According to postoperatively pathological and immunohistochemical results, patients were categorized by malignant potential and the Ki67 index, respectively. CT features were analyzed to be associated with malignant potential or the Ki67 index using univariate analysis, logistic regression and receiver operating curve analysis. Then, we explored the correlation between the Ki67 index and malignant potential by using the Spearman rank correlation.

RESULTS

Based on univariate and multivariate analysis, a predictive model of malignant potential of small intestine GISTs, consisting of tumor size (p < 0.001) and presence of necrosis (p = 0.033), was developed with the area under the receiver operating curve (AUC) of 0.965 (95% CI, 0.915-0.990; p < 0.001), with 91.53% sensitivity, 96.87% specificity, 96.43% PPV, 92.54% NPV, 94.31% diagnostic accuracy. For high Ki67 expression, a model made up of tumor size (p = 0.051), presence of ulceration (p = 0.054) and metastasis (p = 0.001) may be the best predictive combination with an AUC of 0.785 (95% CI, 0.702-0.854; p < 0.001), 63.33% sensitivity, 76.34% specificity, 46.34% PPV, 86.59% NPV, 73.17% diagnostic accuracy. Ki67 index showed a moderate positive correlation with mitotic count (r = 0.578, p < 0.001), a weak positive correlation with tumor size (r = 0.339, p < 0.001) and with risk stratification (r = 0.364, p < 0.001).

CONCLUSION

Features on CE-CT could preoperatively predict malignant potential and high Ki67 expression of small intestine GISTs, and Ki67 index may be a promising prognostic factor in predicting the prognosis of small intestine GISTs, independent of the risk stratification system.

Ultrasound radiomics model-based nomogram for predicting the risk Stratification of gastrointestinal stromal tumors

This study aimed to develop and evaluate a nomogram based on an ultrasound radiomics model to predict the risk grade of gastrointestinal stromal tumors (GISTs). 216 GIST patients pathologically diagnosed between December 2016 and December 2021 were reviewed and divided into a training cohort (n = 163) and a validation cohort (n = 53) in a ratio of 3:1. The tumor region of interest was depicted on each patient’s ultrasound image using ITK-SNAP, and the radiomics features were extracted. By filtering unstable features and using Spearman’s correlation analysis, and the least absolute shrinkage and selection operator algorithm, a radiomics score was derived to predict the malignant potential of GISTs. a radiomics nomogram that combines the radiomics score and clinical ultrasound predictors was constructed and assessed in terms of calibration, discrimination, and clinical usefulness. The radiomics score from ultrasound images was significantly associated with the malignant potential of GISTs. The radiomics nomogram was superior to the clinical ultrasound nomogram and the radiomics score, and it achieved an AUC of 0.90 in the validation cohort. Based on the decision curve analysis, the radiomics nomogram was found to be more clinically significant and useful. A nomogram consisting of radiomics score and the maximum tumor diameter demonstrated the highest accuracy in the prediction of risk grade in GISTs. The outcomes of our study provide vital insights for important preoperative clinical decisions.

Malignancy risk of gastrointestinal stromal tumors evaluated with noninvasive radiomics: A multi-center study

Purpose

This study was to investigate the diagnostic efficacy of radiomics models based on the enhanced CT images in differentiating the malignant risk of gastrointestinal stromal tumors (GIST) in comparison with the clinical indicators model and traditional CT diagnostic criteria.

Materials and methods

A total of 342 patients with GISTs confirmed histopathologically were enrolled from five medical centers. Data of patients wrom two centers comprised the training group (n=196), and data from the remaining three centers constituted the validation group (n=146). After CT image segmentation and feature extraction and selection, the arterial phase model and venous phase model were established. The maximum diameter of the tumor and internal necrosis were used to establish a clinical indicators model. The traditional CT diagnostic criteria were established for the classification of malignant potential of tumor. The performance of the four models was assessed using the receiver operating characteristics curve.

Reuslts

In the training group, the area under the curves(AUCs) of the arterial phase model, venous phase model, clinical indicators model, and traditional CT diagnostic criteria were 0.930 [95% confidence interval (CI): 0.895-0.965), 0.933 (95%CI 0.898-0.967), 0.917 (95%CI 0.872-0.961) and 0.782 (95%CI 0.717-0.848), respectively. In the validation group, the AUCs of the models were 0.960 (95%CI 0.930-0.990), 0.961 (95% CI 0.930-0.992), 0.922 (95%CI 0.884-0.960) and 0.768 (95%CI 0.692-0.844), respectively. No significant difference was detected in the AUC between the arterial phase model, venous phase model, and clinical indicators model by the DeLong test, whereas a significant difference was observed between the traditional CT diagnostic criteria and the other three models.

Conclusion

The radiomics model using the morphological features of GISTs play a significant role in tumor risk stratification and can provide a reference for clinical diagnosis and treatment plan.

Gastrointestinal stromal tumors: a comprehensive radiological review

Gastrointestinal stromal tumors (GISTs) originating from the interstitial cells of Cajal in the muscularis propria are the most common mesenchymal tumor of the gastrointestinal tract. Multiple modalities, including computed tomography (CT), magnetic resonance imaging (MRI), fluorodeoxyglucose positron emission tomography, ultrasonography, digital subtraction angiography, and endoscopy, have been performed to evaluate GISTs. CT is most frequently used for diagnosis, staging, surveillance, and response monitoring during molecularly targeted therapy in clinical practice. The diagnosis of GISTs is sometimes challenging because of the diverse imaging findings, such as anatomical location (esophagus, stomach, duodenum, small bowel, colorectum, appendix, and peritoneum), growth pattern, and enhancement pattern as well as the presence of necrosis, calcification, ulceration, early venous return, and metastasis. Imaging findings of GISTs treated with antineoplastic agents are quite different from those of other neoplasms (e.g. adenocarcinomas) because only subtle changes in size are seen even in responsive lesions. Furthermore, the recurrence pattern of GISTs is different from that of other neoplasms. This review discusses the advantages and disadvantages of each imaging modality, describes imaging findings obtained before and after treatment, presents a few cases of complicated GISTs, and discusses recent investigations performed using CT and MRI to predict histological risk grade, gene mutations, and patient outcomes.

Malignant Small Bowel Neoplasms:a review of post-contrast multiphasic multidetector computed tomography

Small bowel neoplasms are rare and account for 3-6% of all gastrointestinal neoplasms. For the diagnosis of small bowel neoplasms, differentiating normal bowel tissue from tumor is critical and depends on imaging modality and scanning techniques. The detection and characterization of small bowel neoplasms have recently improved with the advance of computed tomography (CT) technology. Post-contrast multiphasic CT is an aid to detection and recognition of the vascular nature of small bowel neoplasms. Understanding the typical post-contrast multiphasic CT features of small bowel neoplasms is important because of overlapping features and the necessity of evaluating associated complications and metastases to lymph node and other organs. However, accurate classification of pathologies is still challenging in clinical practice. Texture analysis can quantify complex mathematical patterns within the gray-level distribution of the pixels and voxels of digital images, and texture analysis of the post-contrast multidetector CT data of various tumors has been attracting attention in recent years. The aim of this article is to provide a comprehensive guide to the relevant imaging features for different types of malignant small bowel neoplasms. J. Med. Invest. 69 : 19-24, February, 2022.

Differential Diagnosis and Molecular Stratification of Gastrointestinal Stromal Tumors on CT Images Using a Radiomics Approach

Treatment planning of gastrointestinal stromal tumors (GISTs) includes distinguishing GISTs from other intra-abdominal tumors and GISTs’ molecular analysis. The aim of this study was to evaluate radiomics for distinguishing GISTs from other intra-abdominal tumors, and in GISTs, predict the c-KIT, PDGFRA, BRAF mutational status, and mitotic index (MI). Patients diagnosed at the Erasmus MC between 2004 and 2017, with GIST or non-GIST intra-abdominal tumors and a contrast-enhanced venous-phase CT, were retrospectively included. Tumors were segmented, from which 564 image features were extracted. Prediction models were constructed using a combination of machine learning approaches. The evaluation was performed in a 100 × random-split cross-validation. Model performance was compared to that of three radiologists. One hundred twenty-five GISTs and 122 non-GISTs were included. The GIST vs. non-GIST radiomics model had a mean area under the curve (AUC) of 0.77. Three radiologists had an AUC of 0.69, 0.76, and 0.84, respectively. The radiomics model had an AUC of 0.52 for c-KIT, 0.56 for c-KIT exon 11, and 0.52 for the MI. The numbers of PDGFRA, BRAF, and other c-KIT mutations were too low for analysis. Our radiomics model was able to distinguish GISTs from non-GISTs with a performance similar to three radiologists, but less observer dependent. Therefore, it may aid in the early diagnosis of GIST, facilitating rapid referral to specialized treatment centers. As the model was not able to predict any genetic or molecular features, it cannot aid in treatment planning yet.

Correlation of CT radiomic features for GISTs with pathological classification and molecular subtypes: preliminary and monocentric experience

PURPOSE

Our primary purpose was to search for computed tomography (CT) radiomic features of gastrointestinal stromal tumors (GISTs) that could potentially correlate with the risk class according to the Miettinen classification. Subsequently, assess the existence of features with possible predictive value in differentiating responder from non-responder patients to first-line therapy with Imatinib.

METHODS

A retrospective study design was carried out using data from June 2009 to December 2020. We analyzed all the preoperative CTs of patients undergoing surgery for GISTs. We segmented non-contrast-enhanced CT (NCECT) and contrast-enhanced venous CT (CECT) images obtained either on three different CT scans (heterogeneous cohort) or on a single CT scan (homogeneous cohort). We then divided the patients into two groups according to Miettinen classification criteria and based on the predictive value of response to first-line therapy with Imatinib.

RESULTS

We examined 54 patients with pathological confirmation of GISTs. For the heterogeneous cohort, we found a statistically significant relationship between 57 radiomic features for NCECT and 56 radiomic features for CECT using the Miettinen risk classification. In the homogeneous cohort, we found the same relationship between 8 features for the NCECT and 5 features for CECT, all included in the heterogeneous cohort. The various radiomic features are distributed with different values in the two risk stratification groups according to the Miettinen classification. We also found some features for groups predictive of response to first-line therapy with Imatinib.

CONCLUSIONS

We found radiomic features that correlate with statistical significance for both the Miettinen risk classification and the molecular subtypes of response. All features found in the homogeneous study cohort were also found in the heterogeneous cohort. CT radiomic features may be useful in assessing the risk class and prognosis of GISTs.

CT classification model of pancreatic serous cystic neoplasms and mucinous cystic neoplasms based on a deep neural network

BACKGROUND

At present, numerous challenges exist in the diagnosis of pancreatic SCNs and MCNs. After the emergence of artificial intelligence (AI), many radiomics research methods have been applied to the identification of pancreatic SCNs and MCNs.

PURPOSE

A deep neural network (DNN) model termed Multi-channel-Multiclassifier-Random Forest-ResNet (MMRF-ResNet) was constructed to provide an objective CT imaging basis for differential diagnosis between pancreatic serous cystic neoplasms (SCNs) and mucinous cystic neoplasms (MCNs).

MATERIALS AND METHODS

This study is a retrospective analysis of pancreatic unenhanced and enhanced CT images in 63 patients with pancreatic SCNs and 47 patients with MCNs (3 of which were mucinous cystadenocarcinoma) confirmed by pathology from December 2010 to August 2016. Different image segmented methods (single-channel manual outline ROI image and multi-channel image), feature extraction methods (wavelet, LBP, HOG, GLCM, Gabor, ResNet, and AlexNet) and classifiers (KNN, Softmax, Bayes, random forest classifier, and Majority Voting rule method) are used to classify the nature of the lesion in each CT image (SCNs/MCNs). Then, the comparisons of classification results were made based on sensitivity, specificity, precision, accuracy, F1 score, and area under the receiver operating characteristic curve (AUC), with pathological results serving as the gold standard.

RESULTS

Multi-channel-ResNet (AUC 0.98) was superior to Manual-ResNet (AUC 0.91).CT image characteristics of lesions extracted by ResNet are more representative than wavelet, LBP, HOG, GLCM, Gabor, and AlexNet. Compared to the use of three classifiers alone and Majority Voting rule method, the use of the MMRF-ResNet model exhibits a better evaluation effect (AUC 0.96) for the classification of the pancreatic SCNs and MCNs.

CONCLUSION

The CT image classification model MMRF-ResNet is an effective method to distinguish between pancreatic SCNs and MCNs.

Building Radiomics Models Based on Triple-Phase CT Images Combining Clinical Features for Discriminating the Risk Rating in Gastrointestinal Stromal Tumors

We aimed to build radiomics models based on triple-phase CT images combining clinical features to predict the risk rating of gastrointestinal stromal tumors (GISTs). A total of 231 patients with pathologically diagnosed GISTs from July 2012 to July 2020 were categorized into a training data set (82 patients with high risk, 80 patients with low risk) and a validation data set (35 patients with high risk, 34 patients with low risk) with a ratio of 7:3. Four diagnostic models were constructed by assessing 20 clinical characteristics and 18 radiomic features that were extracted from a lesion mask based on triple-phase CT images. The receiver operating characteristic (ROC) curves were applied to calculate the diagnostic performance of these models, and ROC curves of these models were compared using Delong test in different data sets. The results of ROC analyses showed that areas under ROC curves (AUC) of model 4 [Clinic + CT value of unenhanced (CTU) + CT value of arterial phase (CTA) + value of venous phase (CTV)], model 1 (Clinic + CTU), model 2 (Clinic + CTA), and model 3 (Clinic + CTV) were 0.925, 0.894, 0.909, and 0.914 in the training set and 0.897, 0.866, 0,892, and 0.892 in the validation set, respectively. Model 4, model 1, model 2, and model 3 yielded an accuracy of 88.3%, 85.8%, 86.4%, and 84.6%, a sensitivity of 85.4%, 84.2%, 76.8%, and 78.0%, and a specificity of 91.2%, 87.5%, 96.2%, and 91.2% in the training set and an accuracy of 88.4%, 84.1%, 82.6%, and 82.6%, a sensitivity of 88.6%, 77.1%, 74.3%, and 85.7%, and a specificity of 88.2%, 91.2%, 91.2%, and 79.4% in the validation set, respectively. There was a significant difference between model 4 and model 1 in discriminating the risk rating in gastrointestinal stromal tumors in the training data set (Delong test, p < 0.05). The radiomic models based on clinical features and triple-phase CT images manifested excellent accuracy for the discrimination of risk rating of GISTs.

Preoperative CT-Based Deep Learning Model for Predicting Risk Stratification in Patients With Gastrointestinal Stromal Tumors

Objective

To develop and evaluate a deep learning model (DLM) for predicting the risk stratification of gastrointestinal stromal tumors (GISTs).

Methods

Preoperative contrast-enhanced CT images of 733 patients with GISTs were retrospectively obtained from two centers between January 2011 and June 2020. The datasets were split into training (n = 241), testing (n = 104), and external validation cohorts (n = 388). A DLM for predicting the risk stratification of GISTs was developed using a convolutional neural network and evaluated in the testing and external validation cohorts. The performance of the DLM was compared with that of radiomics model by using the area under the receiver operating characteristic curves (AUROCs) and the Obuchowski index. The attention area of the DLM was visualized as a heatmap by gradient-weighted class activation mapping.

Results

In the testing cohort, the DLM had AUROCs of 0.90 (95% confidence interval [CI]: 0.84, 0.96), 0.80 (95% CI: 0.72, 0.88), and 0.89 (95% CI: 0.83, 0.95) for low-malignant, intermediate-malignant, and high-malignant GISTs, respectively. In the external validation cohort, the AUROCs of the DLM were 0.87 (95% CI: 0.83, 0.91), 0.64 (95% CI: 0.60, 0.68), and 0.85 (95% CI: 0.81, 0.89) for low-malignant, intermediate-malignant, and high-malignant GISTs, respectively. The DLM (Obuchowski index: training, 0.84; external validation, 0.79) outperformed the radiomics model (Obuchowski index: training, 0.77; external validation, 0.77) for predicting risk stratification of GISTs. The relevant subregions were successfully highlighted with attention heatmap on the CT images for further clinical review.

Conclusion

The DLM showed good performance for predicting the risk stratification of GISTs using CT images and achieved better performance than that of radiomics model.

CT Radiomics Model for Discriminating the Risk Stratification of Gastrointestinal Stromal Tumors: A Multi-Class Classification and Multi-Center Study

Objective

To establish and verify a computed tomography (CT)-based multi-class prediction model for discriminating the risk stratification of gastrointestinal stromal tumors (GISTs).

Materials and Methods

A total of 381 patients with GISTs were confirmed by surgery and pathology. Information on 213 patients were obtained from one hospital and used as training cohort, whereas the details of 168 patients were collected from two other hospitals and used as independent validation cohort. Regions of interest on CT images of arterial and venous phases were drawn, radiomics features were extracted, and dimensionality reduction processing was performed. Using a one-vs-rest method, a Random Forest-based GISTs risk three-class prediction model was established, and the receiver operating characteristic curve (ROC) was used to evaluate the performance of the multi-class classification model, and the generalization ability was verified using external data.

Results

The training cohort included 96 very low-risk and low-risk, 60 intermediate-risk and 57 high-risk patients. External validation cohort included 82 very low-risk and low-risk, 48 intermediate-risk and 38 high-risk patients. The GISTs risk three-class radiomics model had a macro/micro average area under the curve (AUC) of 0.84 and an accuracy of 0.78 in the training cohort. It had a stable performance in the external validation cohort, with a macro/micro average AUC of 0.83 and an accuracy of 0.80.

Conclusion

CT radiomics can discriminate GISTs risk stratification. The performance of the three-class radiomics prediction model is good, and its generalization ability has also been verified in the external validation cohort, indicating its potential to assist stratified and accurate treatment of GISTs in the clinic.

Computed-Tomography-Based Radiomics Model for Predicting the Malignant Potential of Gastrointestinal Stromal Tumors Preoperatively: A Multi-Classifier and Multicenter Study

Background: Our goal was to establish and verify a radiomics risk grading model for gastrointestinal stromal tumors (GISTs) and to identify the optimal algorithm for risk stratification. Methods: We conducted a retrospective analysis of 324 patients with GISTs, the presence of which was confirmed by surgical pathology. Patients were treated at three different hospitals. A training cohort of 180 patients was collected from the largest center, while an external validation cohort of 144 patients was collected from the other two centers. To extract radiomics features, regions of interest (ROIs) were outlined layer by layer along the edge of the tumor contour on CT images of the arterial and portal venous phases. The dimensionality of radiomic features was reduced, and the top 10 features with importance value above 5 were selected before modeling. The training cohort used three classifiers [logistic regression, support vector machine (SVM), and random forest] to establish three GIST risk stratification prediction models. The receiver operating characteristic curve (ROC) was used to compare model performance, which was validated by external data. Results: In the training cohort, the average area under the curve (AUC) was 0.84 ± 0.07 of the logistic regression, 0.88 ± 0.06 of the random forest, and 0.81 ± 0.08 of the SVM. In the external validation cohort, the AUC was 0.85 of the logistic regression, 0.90 of the random forest, and 0.80 of the SVM. The random forest model performed the best in both the training and the external validation cohorts and could be generalized. Conclusion: Based on CT radiomics, there are multiple machine-learning models that can predict the risk of GISTs. Among them, the random forest algorithm had the highest prediction efficiency and could be readily generalizable. Through external validation data, we assume that the random forest model may be used as an effective tool to guide preoperative clinical decision-making.

CT Texture Analysis for Preoperative Identification of Lymphoma from Other Types of Primary Small Bowel Malignancies

Objectives

To explore the application of computed tomography (CT) texture analysis in differentiating lymphomas from other malignancies of the small bowel.

Methods

Arterial and venous CT images of 87 patients with small bowel malignancies were retrospectively analyzed. The subjective radiological features were evaluated by the two radiologists with a consensus agreement. The region of interest (ROI) was manually delineated along the edge of the lesion on the largest slice, and a total of 402 quantified features were extracted automatically from AK software. The inter- and intrareader reproducibility was evaluated to select highly reproductive features. The univariate analysis and minimum redundancy maximum relevance (mRMR) algorithm were applied to select the feature subsets with high correlation and low redundancy. The multivariate logistic regression analysis based on texture features and radiological features was employed to construct predictive models for identification of small bowel lymphoma. The diagnostic performance of multivariate models was evaluated using receiver operating characteristic (ROC) curve analysis.

Results

The clinical data (age, melena, and abdominal pain) and radiological features (location, shape, margin, dilated lumen, intussusception, enhancement level, adjacent peritoneum, and locoregional lymph node) differed significantly between the nonlymphoma group and lymphoma group (p < 0.05). The areas under the ROC curve of the clinical model, arterial texture model, and venous texture model were 0.93, 0.92, and 0.87, respectively.

Conclusion

The arterial texture model showed a great diagnostic value and fitted performance in preoperatively discriminating lymphoma from nonlymphoma of the small bowel.

Evaluation of risk classifications for gastrointestinal stromal tumor using multi-parameter Magnetic Resonance analysis

BACKGROUND

Gastrointestinal stromal tumor (GIST) is the most common mesenchymal malignancy of the gastrointestinal tract. At present, it is generally believed that the prognosis of GIST is closely related to its risk classification. It may add value to correctly diagnose and evaluate the risk of invasion using a noninvasive imaging examination prior to surgery. MRI has the advantages of multiple parameters and high soft tissue resolution, which may be the potential method to preoperatively evaluate the risk of GIST.

PURPOSE

To retrospectively evaluate the diagnostic accuracy of multi-parameter MR analysis for preoperative risk classification of GIST.

MATERIALS AND METHODS

In this 6-year retrospective study, full MRI examination was performed on all 60 GIST cases confirmed classified by pathology, including 35 cases of very low-to-low-risk GIST and 25 cases of intermediate-to-high-risk GIST. Dynamic contrast-enhanced T1- and T2-weighted images, and apparent diffusion coefficient (ADC) maps were reviewed independently by two radiologists blinded to pathologic results. Volume, ADC ratio, three wash-in indexes (WII) were calculated and compared using t-test or Kruskal-Wallis nonparametric test. Sensitivity and specificity analyses were performed to calculate diagnostic accuracy using ROC analyses. Differences were considered significant at p < 0.05.

RESULTS

All GISTs were resected. Patient age, sex, tumor location and tumor shape did not differ significantly across the two groups (p = 0.798, 0.767, 0.822 and 0.096, respectively). GIST in the intermediate-to-high-risk group presented significantly greater volume (p = 0.0045), lower ADC ratio (p = 0.0125) and faster enhancement (for WII2, p < 0.0001; for WII3, p = 0.0358) than that of GIST in the very low-to-low-risk group. This combination of the volume, ADC ratio and WII2 provided sensitivity of 88%, specificity of 94.29%, and accuracy of 91.7% for the risk classification of GIST.

CONCLUSION

Multi-parameter MR analysis provides a preoperative imaging standard for accurately distinguishing very low-to-low-risk GIST from intermediate-to-high-risk GIST.

Primary Gastro-Intestinal Lymphoma and Gastro-Intestinal Adenocarcinoma: An Initial Study of CT Texture Analysis as Quantitative Biomarkers for Differentiation

BACKGROUND

To explore the potential role of computed tomography (CT) texture analysis and an imaging biomarker in differentiating primary gastro-intestinal lymphoma (PGIL) from gastro-intestinal adenocarcinoma (GIAC).

METHODS

A total of 131 patients with surgical pathologically PGIL and GIAC were enrolled in this study. Histogram parameters of arterial and venous phases extracted from contrast enhanced modified discrete cosine transform (MDCT) images were compared between PGIL and GIAC by Mann-Whitney U tests. The optimal parameters for differentiating these two groups were obtained through receiver operating characteristic (ROC) curves and the area under the curve (AUC) was calculated.

RESULTS

Compared with GIAC, in arterial phase, PGIL had statistically higher 5th, 10th percentiles (p = 0.003 and 0.011) and statistically lower entropy (p = 0.001). In the venous phase, PGIL had statistically lower mean, median, 75th, 90th, 95th percentiles, and entropy (p = 0.036, 0.029, 0.007, 0.001 and 0.001, respectively). For differentiating PGIL from GIAC, V-median + A-5th percentile was an optimal parameter for combined diagnosis (AUC = 0.746, p < 0.0001), and the corresponding sensitivity and specificity were 81.7 and 64.8%, respectively.

CONCLUSION

CT texture analysis could be useful for differential diagnosis of PGIL and GIAC.

CT texture analysis in histological classification of epithelial ovarian carcinoma

OBJECTIVES

The study aimed to compare the ability of morphological and texture features derived from contrast-enhanced CT in histological subtyping of epithelial ovarian carcinoma (EOC).

METHODS

Consecutive 205 patients with newly diagnosed EOC who underwent contrast-enhanced CT were included and dichotomised into high-grade serous carcinoma (HGSC) and non-HGSC. Clinical information including age and cancer antigen 125 (CA-125) was documented. The pre-treatment images were analysed using commercial software, TexRAD, by two independent radiologists. Eight qualitative CT morphological features were evaluated, and 36 CT texture features at 6 spatial scale factors (SSFs) were extracted per patient. Features' reduction was based on kappa score, intra-class correlation coefficient (ICC), univariate ROC analysis and Pearson's correlation test. Texture features with ICC ≥ 0.8 were compared by histological subtypes. Patients were randomly divided into training and testing sets by 8:2. Two random forest classifiers were determined and compared: model 1 incorporating selected morphological and clinical features and model 2 incorporating selected texture and clinical features.

RESULTS

HGSC showed specifically higher texture features than non-HGSC (p < 0.05). Both models performed highly in predicting histological subtypes of EOC (model 1: AUC 0.891 and model 2: AUC 0.937), and no statistical significance was found between the two models (p = 0.464).

CONCLUSION

CT texture analysis provides objective and quantitative metrics on tumour characteristics with HGSC demonstrating specifically high texture features. The model incorporating texture analysis could classify histology subtypes of EOC with high accuracy and performed as well as morphological features.

KEY POINTS

• A number of CT morphological and texture features showed good inter- and intra-observer agreements. • High-grade serous ovarian carcinoma showed specifically higher CT texture features than non-high-grade serous ovarian carcinoma. • CT texture analysis could differentiate histological subtypes of epithelial ovarian carcinoma with high accuracy.

Prediction of the World Health Organization Grade of rectal neuroendocrine tumors based on CT histogram analysis

OBJECTIVES

To investigate the diagnostic value of contrast-enhanced computed tomography (CECT) histogram analysis in predicting the World Health Organization (WHO) grade of rectal neuroendocrine tumors (R-NETs).

MATERIALS AND METHODS

A total of 61 (35 G1, 12 G2, 10 G3, and 4 NECs) patients who underwent preoperative CECT and treated with surgery to be confirmed as R-NETs were included in this study from January 2014 to May 2019. We depicted ROIs and measured the CECT texture parameters (mean, median, 10th, 25th, 75th, 90th percentiles, skewness, kurtosis, and entropy) from arterial phase (AP) and venous phase (VP) images by two radiologists. We calculated intraclass correlation coefficient (ICC) and compared the histogram parameters between low-grade (G1) and higher grade (HG) (G2/G3/NECs) by applying appropriate statistical method. We obtained the optimal parameters to identify G1 from HG using receiver operating characteristic (ROC) curves.

RESULTS

The capability of AP and VP histogram parameters for differentiating G1 from HG was similar in several histogram parameters (mean, median, 10th, 25th, 75th, and 90th percentiles) (all p < 0.001). Skewness, kurtosis, and entropy on AP images showed no significant differences between G1 and HG (p = 0.853, 0.512, 0.557, respectively). Entropy on VP images was significantly different (p = 0.017) between G1 and HG, however, skewness and kurtosis showed no significant differences (p = 0.654, 0.172, respectively). ROC analysis showed a good predictive performance between G1 and HG, and the 75th (AP) generated the highest area under the curve (AUC = 0.871), followed by the 25th (AP), mean (VP), and median (VP) (AUC = 0.864). Combined the size of tumor and the 75th (AP) generated the highest AUC.

CONCLUSIONS

CECT histogram parameters, including arterial and venous phases, can be used as excellent indicators for predicting G1 and HG of rectal neuroendocrine tumors, and the size of the tumor is also an important independent predictor.

Development of a Malignancy Potential Binary Prediction Model Based on Deep Learning for the Mitotic Count of Local Primary Gastrointestinal Stromal Tumors

Objective

The mitotic count of gastrointestinal stromal tumors (GIST) is closely associated with the risk of planting and metastasis. The purpose of this study was to develop a predictive model for the mitotic index of local primary GIST, based on deep learning algorithm.

Materials and Methods

Abdominal contrast-enhanced CT images of 148 pathologically confirmed GIST cases were retrospectively collected for the development of a deep learning classification algorithm. The areas of GIST masses on the CT images were retrospectively labelled by an experienced radiologist. The postoperative pathological mitotic count was considered as the gold standard (high mitotic count, > 5/50 high-power fields [HPFs]; low mitotic count, ≤ 5/50 HPFs). A binary classification model was trained on the basis of the VGG16 convolutional neural network, using the CT images with the training set (n = 108), validation set (n = 20), and the test set (n = 20). The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated at both, the image level and the patient level. The receiver operating characteristic curves were generated on the basis of the model prediction results and the area under curves (AUCs) were calculated. The risk categories of the tumors were predicted according to the Armed Forces Institute of Pathology criteria.

Results

At the image level, the classification prediction results of the mitotic counts in the test cohort were as follows: sensitivity 85.7% (95% confidence interval [CI]: 0.834–0.877), specificity 67.5% (95% CI: 0.636–0.712), PPV 82.1% (95% CI: 0.797–0.843), NPV 73.0% (95% CI: 0.691–0.766), and AUC 0.771 (95% CI: 0.750–0.791). At the patient level, the classification prediction results in the test cohort were as follows: sensitivity 90.0% (95% CI: 0.541–0.995), specificity 70.0% (95% CI: 0.354–0.919), PPV 75.0% (95% CI: 0.428–0.933), NPV 87.5% (95% CI: 0.467–0.993), and AUC 0.800 (95% CI: 0.563–0.943).

Conclusion

We developed and preliminarily verified the GIST mitotic count binary prediction model, based on the VGG convolutional neural network. The model displayed a good predictive performance.

New advances in radiomics of gastrointestinal stromal tumors

Gastrointestinal stromal tumors (GISTs) are uncommon neoplasms of the gastrointestinal tract with peculiar clinical, genetic, and imaging characteristics. Preoperative knowledge of risk stratification and mutational status is crucial to guide the appropriate patients’ treatment. Predicting the clinical behavior and biological aggressiveness of GISTs based on conventional computed tomography (CT) and magnetic resonance imaging (MRI) evaluation is challenging, unless the lesions have already metastasized at the time of diagnosis. Radiomics is emerging as a promising tool for the quantification of lesion heterogeneity on radiological images, extracting additional data that cannot be assessed by visual analysis. Radiomics applications have been explored for the differential diagnosis of GISTs from other gastrointestinal neoplasms, risk stratification and prediction of prognosis after surgical resection, and evaluation of mutational status in GISTs. The published researches on GISTs radiomics have obtained excellent performance of derived radiomics models on CT and MRI. However, lack of standardization and differences in study methodology challenge the application of radiomics in clinical practice. The purpose of this review is to describe the new advances of radiomics applied to CT and MRI for the evaluation of gastrointestinal stromal tumors, discuss the potential clinical applications that may impact patients’ management, report limitations of current radiomics studies, and future directions.

Radiomics Based on CECT in Differentiating Kimura Disease From Lymph Node Metastases in Head and Neck: A Non-Invasive and Reliable Method

Background: Kimura disease may be easily misdiagnosed as malignant tumors such as lymph node metastases based on imaging and clinical symptoms. The aim of this article is to investigate whether the radiomic features and the model based on the features on venous-phase contrast-enhanced CT (CECT) images can distinguish Kimura disease from lymph node metastases in the head and neck.

Methods: A retrospective analysis of 14 patients of head and neck Kimura disease (a total of 38 enlarged lymph nodes) and 39 patients with head and neck lymph node metastases (a total of 39 enlarged lymph nodes), confirmed by biopsy or surgery resection, was conducted. All patients accepted CECT within 10 days before biopsy or surgery resection. Radiomic features based on venous-phase CECT were generated automatically from Artificial-Intelligence Kit (AK) software. All lymph nodes were randomly divided into the training set (n = 54) and testing set (n = 23) in a ratio of 7:3. ANOVA + Mann–Whitney, Spearman correlation, least absolute shrinkage and selection operator, and Gradient Descent were introduced for the reduction of the highly redundant features. Binary logistic regression model was constructed based on the selected features. Receiver operating characteristic was used to evaluate the diagnostic performance of the features and the model. Finally, a nomogram was established for model application.

Results: Seven features were screened out at the end. Significant difference was found between the two groups for all the features with area under the curves (AUCs) ranging from 0.759 to 0.915. The AUC of the model's identification performance was 0.970 in the training group and 0.977 in the testing group. The disease discrimination efficiency of the model was better than that of any single feature.

Conclusions: The radiomic features and the model based on these features on venous-phase CECT images had very good performance for the discrimination between Kimura disease and lymph node metastases in the head and neck.

Computed Tomography-Based Tumor Heterogeneity Analysis Reveals Differences in a Cohort with Advanced Pancreatic Carcinoma under Palliative Chemotherapy

Purpose

Imaging in pancreatic cancer is a challenge, especially regarding therapy response evaluation. Tumor size, attenuation, and perfusion are widely used as parameters for computed tomography (CT) examinations, but are often limited due to blurry tumor borders and missing qualitative parameters. To improve monitoring of therapy response, we tested a new CT-based approach of tumor heterogeneity feature analysis.

Methods

A total of 13 patients with pancreatic adenocarcinoma undergoing abdominal CT according to standard as baseline imaging with clinical follow-up and imaging (median time span 64 days) under systematic therapy (FOLFIRINOX/gemcitabine) were retrospectively analyzed. Progression was defined as new lesions and local tumor spread. Tumor heterogeneity analysis was performed using mintLesion®. Seven different image features referring to image heterogeneity were analyzed. Statistical analysis was performed with Spearman's rank correlation and Mann-Whitney U test.

Results

During follow-up, tumor volume did not significantly change between our groups with overall progression (local and systemic) and progression-free patients (p = 0.661). Mean positivity of pixel values were significantly higher in patients without progression compared to patients with progression (p = 0.030). There was a significant negative correlation between changes in kurtosis and time to local tumor spread (p = 0.008) or systemic progression (p = 0.017).

Conclusions

Results suggest that analysis of tumor heterogeneity might provide valuable information from routine-acquired images regarding therapy response evaluation. This might help adjusting therapy regimes and could be easily integrated in clinical workflows. Furthermore, this procedure might possibly predict therapy response and, hence could lead the way to find a potential marker for progression-free survival.

Computed tomography-based radiomics model for discriminating the risk stratification of gastrointestinal stromal tumors

PURPOSE

The pathological risk degree of gastrointestinal stromal tumors (GISTs) has become an issue of great concern. Computed tomography (CT) is beneficial for showing adjacent tissues in detail and determining metastasis or recurrence of GISTs, but its function is still limited. Radiomics has recently shown a great potential in aiding clinical decision-making. The purpose of our study is to develop and validate CT-based radiomics models for GIST risk stratification.

METHODS

Three hundred and sixty-six patients clinically suspected of primary GISTs from January 2013 to February 2018 were retrospectively enrolled, among which data from 140 patients were eventually analyzed after exclusion. Data from patient CT images were partitioned based on the National Institutes of Health Consensus Classification, including tumor segmentation, radiomics feature extraction and selection. A radiomics model was then proposed and validated.

RESULTS

The radiomics signature demonstrated discriminative performance for advanced and nonadvanced GISTs with an area under the curve (AUC) of 0.935 [95% confidence interval (CI) 0.870-1.000] and an accuracy of 90.2% for validation cohort. The radiomics signature demonstrated favorable performance for the risk stratification of GISTs with an AUC of 0.809 (95% CI 0.777-0.841) and an accuracy of 67.5% for the validation cohort. Radiomics analysis could capture features of the four risk categories of GISTs. Meanwhile, this CT-based radiomics signature showed good diagnostic accuracy to distinguish between nonadvanced and advanced GISTs, as well as the four risk stratifications of GISTs.

CONCLUSION

Our findings highlight the potential of a quantitative radiomics analysis as a complementary tool to achieve an accurate diagnosis for GISTs.

Histogram analysis with computed tomography angiography for discriminating soft tissue sarcoma from benign soft tissue tumor

To investigate the feasibility of histogram analysis with computed tomography angiography (CTA) in distinguishing between soft tissue sarcomas and benign soft tissue tumors. Fourty nine patients (23 men, mean age = 44.3 years, age range = 25-64) with pathologically-confirmed soft tissue sarcoma (n = 24) or benign soft tissue tumors (n = 25) in the lower extremities undergoing CTA for tumor evaluation were retrospectively analyzed. Two radiologists separately performed histogram analyses of CT density with CTA images by drawing a region of interest (ROI). The 10th (P10), 25th (P25), 50th (P50), 75th (P75), 90th percentiles (P90), mean, and standard deviations (SD) of measured tumor density were obtained along with measurements of the absolute value of kurtosis (AVK), absolute value of skewness (AVS), and inhomogeneity for each tumor. Intra-class correlation coefficients (ICC) were calculated to determine inter- and intra-reader variability in parameter measurements. The Mann-Whitney U test was used to compare histogram parameters between soft tissue sarcomas and benign soft tissue tumors. Receiver operator characteristic (ROC) curves were constructed to evaluate the accuracy of tumor discrimination. ICC was greater than 0.7 for AVS, AVK, and inhomogeneity, and >0.9 for mean, SD, and all percentile measures. There was no significant difference in P10, P25, P50, P75, P90, mean, or SD between soft tissue sarcomas and benign tumors (P > .05). AVS, AVK, and inhomogeneity were significantly higher in soft tissue sarcomas (P < .05). Areas under the curve (AUC) were 0.81, 0.83, and 0.84 for AVS, AVK, and inhomogeneity respectively. AUC were below 0.6 for mean, SD, and all percentiles.Skewness, kurtosis, and inhomogeneity measurements derived from histogram analysis from CTA distinguish between soft tissue sarcomas and benign soft tissue tumors.

Current and Potential Applications of Artificial Intelligence in Gastrointestinal Stromal Tumor Imaging

The most common mesenchymal tumors are gastrointestinal stromal tumors (GISTs), which have malignant potential and can occur anywhere along the gastrointestinal system. Imaging methods are important and indispensable of GISTs in diagnosis, risk staging, therapy, and follow-up. The recommended imaging method for staging and follow-up is computed tomography (CT) according to current guidelines. Artificial intelligence (AI) applies and elaborates theses, procedures, modes, and utilization systems for simulating, enlarging, and stretching the intellectual capacity of humans. Recently, researchers have done a few studies to explore AI applications in GIST imaging. This article reviews the present AI studies in GISTs imaging, including preoperative diagnosis, risk stratification and prediction of prognosis, gene mutation, and targeted therapy response.

Computed tomography textural analysis for the differentiation of chronic lymphocytic leukemia and diffuse large B cell lymphoma of Richter syndrome

OBJECTIVE

To test the hypothesis that both indolent and aggressive chronic lymphocytic leukemia (CLL) can be differentiated from diffuse large B cell lymphoma (DLBCL) of Richter syndrome (RS) by CT texture analysis (CTTA) of involved lymph nodes.

MATERIAL AND METHODS

We retrospectively included 52 patients with indolent CLL (26/52), aggressive CLL (8/52), and DLBCL of RS (18/52), who underwent standardized contrast-enhanced CT. In main lymphoma tissue, VOIs were generated from which CTTA features including first-, second-, and higher-order textural features were extracted. CTTA features were compared between the entire CLL group, the indolent CLL subtype, the aggressive CLL subtype, and DLBCL using a Kruskal-Wallis test. All p values were adjusted after the Bonferroni correction. ROC analyses for significant CTTA features were performed to determine cut-off values for differentiation between the groups.

RESULTS

Compared with DLBCL of RS, CTTA of the entire CLL group showed significant differences of entropy heterogeneity (p < 0.001), mean intensity (p < 0.001), mean average (p = 0.02), and number non-uniformity gray-level dependence matrix (NGLDM) (p = 0.03). Indolent CLL significantly differed for entropy (p < 0.001), uniformity of heterogeneity (p = 0.02), mean intensity (p < 0.001), and mean average (p = 0.01). Aggressive CLL showed significant differences in mean intensity (p = 0.04). For differentiation between CLL and DLBCL of RS, cut-off values for mean intensity and entropy of heterogeneity were defined (e.g., 6.63 for entropy heterogeneity [aggressive CLL vs. DLBCL]; sensitivity 0.78; specificity 0.63).

CONCLUSIONS

CTTA features of ultrastructure and vascularization significantly differ in CLL compared with that in DLBCL of Richter syndrome, allowing complementary to visual features for noninvasive differentiation by contrast-enhanced CT.

KEY POINTS

• Richter transformation of CLL into DLBCL results in structural changes in lymph node architecture and vascularization that can be detected by CTTA. • First-order CT textural features including intensity and heterogeneity significantly differ between both indolent CLL and aggressive CLL and DLBCL of Richter syndrome. • CT texture analysis allows for noninvasive detection of Richter syndrome which is of prognostic value.

Can texture analysis in ultrashort echo-time MRI distinguish primary graft dysfunction from acute rejection in lung transplants? A multidimensional assessment in a mouse model

BACKGROUND

Differentiation of early postoperative complications affects treatment options after lung transplantation.

PURPOSE

To assess if texture analysis in ultrashort echo-time (UTE) MRI allows distinction of primary graft dysfunction (PGD) from acute transplant rejection (ATR) in a mouse lung transplant model.

STUDY TYPE

Longitudinal.

ANIMAL MODEL

Single left lung transplantation was performed in two cohorts of six mice (strain C57BL/6) receiving six syngeneic (strain C57BL/6) and six allogeneic lung transplants (strain BALB/c (H-2K^d )).

FIELD STRENGTH/SEQUENCE

4.7T small-animal MRI/eight different UTE sequences (echo times: 50-5000 μs) at three different postoperative timepoints (1, 3, and 7 days after transplantation).

ASSESSMENT

Nineteen different first- and higher-order texture features were computed on multiple axial slices for each combination of UTE and timepoint (24 setups) in each mouse. Texture features were compared for transplanted (graft) and contralateral native lungs between and within syngeneic and allogeneic cohorts. Histopathology served as a reference.

STATISTICAL TESTS

Nonparametric tests and correlation matrix analysis were used.

RESULTS

Pathology revealed PGD in the syngeneic and ATR in the allogeneic cohort. Skewness and low-gray-level run-length features were significantly different between PGD and ATR for all investigated setups (P < 0.03). These features were significantly different between graft and native lung in ATR for most setups (minimum of 20/24 setups; all P < 0.05). The number of significantly different features between PGD and ATR increased with elapsing postoperative time. Differences in significant features were highest for an echo-time of 1500 μs.

DATA CONCLUSION

Our findings suggest that texture analysis in UTE-MRI might be a tool for the differentiation of PGD and ATR in the early postoperative phase after lung transplantation.

LEVEL OF EVIDENCE

1 Technical Efficacy: Stage 3 J. Magn. Reson. Imaging 2020;51:108-116.

Prognosis assessment in metastatic gastrointestinal stromal tumors treated with tyrosine kinase inhibitors based on CT-texture analysis

PURPOSE

Identification of prognostic CT-textural features in patients with gastrointestinal stromal tumors undergoing tyrosine kinase inhibitor (TKI) therapy.

METHODS AND MATERIALS

We identified 25 GIST patients (mean age, 70.58 ± 9.7 years; range, 41.25-84.08 years; 20 males, 5 females) with a total of 123 scans, each examined with a standardized CT protocol between 1/2014-7/2018. 92 texture features, based on pyradiomics library, were extracted and correlated to response categories; evaluated with help of modified Choi criteria. All patients underwent therapy with imatinib in the first line and different tyrosine kinase inhibitors after disease progression. KIT and PDGFR-mutations were registered in all patients as well as the number of previous treatment regimens, patient's age as well as gender and the presence of contrast enhancement (vitality) in tumor. The lesion with the largest diameter was chosen and contoured using the spherical VOI tool. Inter-rater testing was performed by a second experienced radiologist. Regression and AUC analysis was performed.

RESULTS

Ten variables could be confirmed to be significantly associated with disease progression. Of them, four textural parameters were significantly positively associated with disease progression and negatively with progression free survival (Glcm Id [grey-level co-occurrence matrix inverse difference], p = 0.012, HR 3.83; 95% CI 1.697-8.611, Glcm Idn [grey-level co-occurrence matrix inverse difference normalized], p = 0.045, HR 2.06, 95% CI 1.015-4.185, Glrlm [grey-level run length matrix] normalized, p = 0.005, HR 3.181; 95% CI 1.418-7.138 and Ngtdm [neighboring grey-tone difference matrix] coarseness, p < 0.001, HR 3.156, 95% CI 1.554-6.411). Single variables were shown to be significantly inferior to the combination of all variables. After 6 months, 90% of patients with 0-1 risk factors (group 1), 64.4% with 2-3 risk variables and 38.1% of patients presenting > 3 structural risk variables showed stable disease. Gclm Id, Gclm Idn and Glrlm non-uniformity were associated with the number of previous treatments, Glrlm non-uniformity also with tumor vitality (enhancement), whereas Gclm Idn and Ngtdm coarseness were associated with the number of tumor mutations.

CONCLUSION

Some of the CT-textural features correlate with disease progression and the progressive free survival as well as with the number of gene mutations and the number of treatment regimens the patients were exposed to as well as with the tumor enhancement. All these features reflect tumor homogeneity.