Diagnostic performance of preoperative CT in differentiating between benign and malignant origin of suspicious gallbladder lesions

Discrimination between benign and malignant gallbladder lesions on enhanced CT imaging using radiomics

BACKGROUND

Gallbladder cancer is a rare but aggressive malignancy that is often diagnosed at an advanced stage and is associated with poor outcomes.

PURPOSE

To develop a radiomics model to discriminate between benign and malignant gallbladder lesions using enhanced computed tomography (CT) imaging.

MATERIAL AND METHODS

All patients had a preoperative contrast-enhanced CT scan, which was independently analyzed by two radiologists. Regions of interest were manually delineated on portal venous phase images, and radiomics features were extracted. Feature selection was performed using mRMR and LASSO methods. The patients were randomly divided into training and test groups at a ratio of 7:3. Clinical and radiomics parameters were identified in the training group, three models were constructed, and the models' prediction accuracy and ability were evaluated using AUC and calibration curves.

RESULTS

In the training group, the AUCs of the clinical model and radiomics model were 0.914 and 0.968, and that of the nomogram model was 0.980, respectively. There were statistically significant differences in diagnostic accuracy between nomograms and radiomics features (P <0.05). There was no significant difference in diagnostic accuracy between the nomograms and clinical features (P >0.05) or between the clinical features and radiomics features (P >0.05). In the testing group, the AUC of the clinical model and radiomics model were 0.904 and 0.941, and that of the nomogram model was 0.948, respectively. There was no significant difference in diagnostic accuracy between the three groups (P >0.05).

CONCLUSION

It was suggested that radiomics analysis using enhanced CT imaging can effectively discriminate between benign and malignant gallbladder lesions.

Optimal radiological gallbladder lesion characterization by combining visual assessment with CT-based radiomics

OBJECTIVES

Differentiating benign gallbladder diseases from gallbladder cancer (GBC) remains a radiological challenge because they can appear very similar on imaging. This study aimed at investigating whether CT-based radiomic features of suspicious gallbladder lesions analyzed by machine learning algorithms could adequately discriminate benign gallbladder disease from GBC. In addition, the added value of machine learning models to radiological visual CT-scan interpretation was assessed.

METHODS

Patients were retrospectively selected based on confirmed histopathological diagnosis and available contrast-enhanced portal venous phase CT-scan. The radiomic features were extracted from the entire gallbladder, then further analyzed by machine learning classifiers based on Lasso regression, Ridge regression, and XG Boosting. The results of the best-performing classifier were combined with radiological visual CT diagnosis and then compared with radiological visual CT assessment alone.

RESULTS

In total, 127 patients were included: 83 patients with benign gallbladder lesions and 44 patients with GBC. Among all machine learning classifiers, XG boosting achieved the best AUC of 0.81 (95% CI 0.72-0.91) and the highest accuracy rate of 73% (95% CI 65-80%). When combining radiological visual interpretation and predictions of the XG boosting classifier, the highest diagnostic performance was achieved with an AUC of 0.98 (95% CI 0.96-1.00), a sensitivity of 91% (95% CI 86-100%), a specificity of 93% (95% CI 90-100%), and an accuracy of 92% (95% CI 90-100%).

CONCLUSIONS

Machine learning analysis of CT-based radiomic features shows promising results in discriminating benign from malignant gallbladder disease. Combining CT-based radiomic analysis and radiological visual interpretation provided the most optimal strategy for GBC and benign gallbladder disease differentiation.

KEY POINTS

Radiomic-based machine learning algorithms are able to differentiate benign gallbladder disease from gallbladder cancer. Combining machine learning algorithms with a radiological visual interpretation of gallbladder lesions at CT increases the specificity, compared to visual interpretation alone, from 73 to 93% and the accuracy from 85 to 92%. Combined use of machine learning algorithms and radiological visual assessment seems the most optimal strategy for GBC and benign gallbladder disease differentiation.

A Comparative Evaluation of Mediastinal Nodal SUVmax and Derived Ratios from 18F-FDG PET/CT Imaging to Predict Nodal Metastases in Non-Small Cell Lung Cancer

18F-FDG positron emission tomography with computed tomography (PET/CT) is a standard imaging modality for the nodal staging of non-small cell lung cancer (NSCLC). To improve the accuracy of pre-operative staging, we compare the staging accuracy of mediastinal lymph node (LN) standard uptake values (SUV) with four derived SUV ratios based on the SUV values of primary tumours (TR), the mediastinal blood pool (MR), liver (LR), and nodal size (SR). In 2015-2017, 53 patients (29 women and 24 men, mean age 67.4 years, range 53-87) receiving surgical resection have pre-operative evidence of mediastinal nodal involvement (cN2). Among these, 114 mediastinal nodes are resected and available for correlative PET/CT analysis. cN2 status accuracy is low, with only 32.5% of the cN2 cases confirmed pathologically. Using receiver operating characteristic (ROC) curve analyses, a SUVmax of N2 LN performs well in predicting the presence of N2 disease (AUC, 0.822). Based on the respective selected thresholds for each ROC curve, normalisation of LN SUVmax to that for mediastinum, liver and tumour improved sensitivities of LN SUVmax from 68% to 81.1-89.2% while maintaining acceptable specificity (68-70.1%). In conclusion, normalised SUV ratios (particularly LR) improve current pre-operative staging performance in detecting mediastinal nodal involvement.

The value of CT findings combined with inflammatory indicators for preoperative differentiation of benign and malignant gallbladder polypoid lesions

BACKGROUND

The study aimed to explore the value of CT findings and inflammatory indicators in differentiating benign and malignant gallbladder polypoid lesions before surgery.

METHODS

The study comprised a total of 113 pathologically confirmed gallbladder polypoid lesions with a maximum diameter ≥ 1 cm (68 benign and 45 malignant), all of which were enhanced CT-scanned within 1 month before surgery. The CT findings and inflammatory indicators of the patients were analyzed by univariate and multivariate logistic regression analysis to identify independent predictors of gallbladder polypoid lesions, and then a nomogram distinguishing benign and malignant gallbladder polypoid lesions was developed by combining these characteristics. The receiver operating characteristic (ROC) curve and decision curve were plotted to assess the performance of the nomogram.

RESULTS

Base status of the lesion (p < 0.001), plain CT value (p < 0.001), neutrophil-lymphocyte ratio (NLR) (p = 0.041), and monocyte-lymphocyte ratio (MLR) (p = 0.022) were independent predictors of malignant polypoid lesions of the gallbladder. The nomogram model established by incorporating the above factors had good performance in differentiating and predicting benign and malignant gallbladder polypoid lesions (AUC = 0.964), with sensitivity and specificity of 82.4% and 97.8%, respectively. The DCA demonstrated the important clinical utility of our nomogram.

CONCLUSION

CT findings combined with inflammatory indicators can effectively differentiate benign and malignant gallbladder polypoid lesions before surgery, which is valuable for clinical decision-making.

The Value of Deep Learning in Gallbladder Lesion Characterization

BACKGROUND

The similarity of gallbladder cancer and benign gallbladder lesions brings challenges to diagnosing gallbladder cancer (GBC). This study investigated whether a convolutional neural network (CNN) could adequately differentiate GBC from benign gallbladder diseases, and whether information from adjacent liver parenchyma could improve its performance.

METHODS

Consecutive patients referred to our hospital with suspicious gallbladder lesions with histopathological diagnosis confirmation and available contrast-enhanced portal venous phase CT scans were retrospectively selected. A CT-based CNN was trained once on gallbladder only and once on gallbladder including a 2 cm adjacent liver parenchyma. The best-performing classifier was combined with the diagnostic results based on radiological visual analysis.

RESULTS

A total of 127 patients were included in the study: 83 patients with benign gallbladder lesions and 44 with gallbladder cancer. The CNN trained on the gallbladder including adjacent liver parenchyma achieved the best performance with an AUC of 0.81 (95% CI 0.71-0.92), being >10% better than the CNN trained on only the gallbladder (p = 0.09). Combining the CNN with radiological visual interpretation did not improve the differentiation between GBC and benign gallbladder diseases.

CONCLUSIONS

The CT-based CNN shows promising ability to differentiate gallbladder cancer from benign gallbladder lesions. In addition, the liver parenchyma adjacent to the gallbladder seems to provide additional information, thereby improving the CNN's performance for gallbladder lesion characterization. However, these findings should be confirmed in larger multicenter studies.