Machine-learning radiomics to predict early recurrence in perihilar cholangiocarcinoma after curative resection.

BACKGROUND AND AIMS: Up to 40%-65% of patients with perihilar cholangiocarcinoma (PHC) rapidly progress to early recurrence (ER) even after curative resection. Quantification of ER risk is difficult and a reliable prognostic prediction tool is absent. We developed and validated a multilevel model, integrating clinicopathology, molecular pathology and radiology, especially radiomics coupled with machine-learning algorithms, to predict the ER of patients after curative resection in PHC. METHODS: In total, 274 patients who underwent contrast-enhanced CT (CECT) and curative resection at 2 institutions were retrospectively identified and randomly divided into training (n = 167), internal validation (n = 70) and external validation (n = 37) sets. A machine-learning analysis of 18,120 radiomic features based on multiphase CECT and 48 clinico-radiologic characteristics was performed for the multilevel model. RESULTS: Comprehensively, 7 independent factors (tumour differentiation, lymph node metastasis, pre-operative CA19-9 level, enhancement pattern, A-Shrink score, V-Shrink score and P-Shrink score) were built to the multilevel model and quantified the risk of ER. We benchmarked the gain in discrimination with the area under the curve (AUC) of 0.883, superior to the rival clinical and radiomic models (AUCs 0.792-0.805). The accuracy (ACC) of the multilevel model was 0.826, which was significantly higher than those of the conventional staging systems (AJCC 8th (0.641), MSKCC (0.617) and Gazzaniga (0.581)). CONCLUSION: The radiomics-based multilevel model demonstrated superior performance to rival models and conventional staging systems, and could serve as a visual prognostic tool to plan surveillance of ER and guide post-operative individualized management in PHC.

Value of susceptibility-weighted imaging in differentiating benign from malignant portal vein thrombosis.

Background: Many diseases are accompanied by portal vein thrombosis (PVT), and its nature is closely related to its prognosis and treatment. It is important to evaluate magnetic resonance imaging (MRI) parameters, including susceptibility-weighted imaging (SWI) and qualitative diffusion-weighted imaging (DWI), in the differentiation between benign and malignant PVT. Methods: In this retrospective study, we collected clinical imaging data from 140 patients with PVTs characterized as benign or malignant based on enhanced MRI between January 2011 and April 2016 and retrospectively analyzed PVTs using SWI and DWI. There were 37 benign and 103 malignant PVTs. Image review was performed by 2 radiologists blinded to clinical information. The signal intensity (SI) of PVTs was recorded on SWI. The apparent diffusion coefficient (ADC) and the ratio of signal intensity (SIR) on SWI (SIRSWI) and ADC (SIRADC) between the PVTs and the spinal cord were calculated. Finally, we generated receiver operating characteristic (ROC) curves to evaluate the efficacy of SIRSWI and SIRADC for distinguishing benign and malignant PVTs. Results: On SWI and DWI, 100.0% (36/36) and 80.5% (29/36) of benign PVTs were hypointense, respectively. For malignant PVTs on SWI and DWI, 99.0% (103/104) and 89.4% (93/104) were hyperintense, respectively. The SIRSWI values of benign and malignant PVTs were 0.58±0.13 and 0.88±0.06, respectively, representing a significant difference (P<0.001). The SIRADC values of benign and malignant PVTs were 0.72±0.32 and 0.62±0.17, respectively, representing a significant difference (P=0.034). The area under the ROC curve (AUROC) for SIRSWI [0.990; 95% confidence interval (CI): 0.971-1.000] was significantly higher than that for SIRADC (0.619; 95% CI: 0.500-0.737; P<0.001). The SIRSWI had a sensitivity of 100.0% and a specificity of 97.3% with a cutoff value of 0.749, while the SIRADC had a sensitivity of 45.9% and specificity of 83.3% with a cutoff value of 0.791. Conclusions: The diagnostic performance of SWI is superior to that of DWI in the differentiation of benign and malignant PVTs.

Evaluation of Preoperative Microvascular Invasion in Hepatocellular Carcinoma Through Multidimensional Parameter Combination Modeling Based on Gd-EOB-DTPA MRI.

Background and Aims: The study established and compared the efficacy of the clinicoradiological model, radiomics model and clinicoradiological-radiomics hybrid model in predicting the microvascular invasion (MVI) of hepatocellular carcinoma (HCC) using gadolinium ethoxybenzyl diethylene triaminepentaacetic acid (Gd-EOB-DTPA) enhanced MRI. Methods: This was a study that enrolled 602 HCC patients from two institutions. Least absolute shrinkage and selection operator (Lasso) method was used to screen for the most important clinicoradiological and radiomics features that predict MVI pre-operatively. Three machine learning algorithms were used to establish the clinicoradiological, radiomics, and clinicoradiological-radiomics hybrid models. Area under the curve (AUC) of receiver operating characteristic (ROC) curves and Delong's test were used to compare and quantify the predictive performance of the models. Results: The AUCs of the clinicoradiological model in training and validation cohorts were 0.793 and 0.701, respectively. The radiomics signature of arterial phase (AP) images alone achieved satisfying predictive efficacy for MVI, with AUCs of 0.671 and 0.643 in training and validation cohort, respectively. The combination of clinicoradiological factors and fusion radiomics signature of AP and VP images achieved AUCs of 0.824 and 0.801 in training and validation cohorts, 0.812 and 0.805 in prospective validation and external validation cohorts, respectively. The hybrid model provided the best prediction results. The results of the Delong test revealed that there were statistically significant differences among the clinicoradiological-radiomics hybrid model, clinicoradiological model, and radiomics model (p<0.05). Conclusions: The combination of clinicoradiological factors and fusion radiomics signature of AP and VP images based on Gd-EOB-DTPA-enhanced MRI can effectively predict MVI.

Associating Preoperative MRI Features and Gene Expression Signatures of Early-stage Hepatocellular Carcinoma Patients using Machine Learning.

BACKGROUND AND AIMS: The relationship between quantitative magnetic resonance imaging (MRI) imaging features and gene-expression signatures associated with the recurrence of hepatocellular carcinoma (HCC) is not well studied. METHODS: In this study, we generated multivariable regression models to explore the correlation between the preoperative MRI features and Golgi membrane protein 1 (GOLM1), SET domain containing 7 (SETD7), and Rho family GTPase 1 (RND1) gene expression levels in a cohort study including 92 early-stage HCC patients. A total of 307 imaging features of tumor texture and shape were computed from T2-weighted MRI. The key MRI features were identified by performing a multi-step feature selection procedure including the correlation analysis and the application of RELIEFF algorithm. Afterward, regression models were generated using kernel-based support vector machines with 5-fold cross-validation. RESULTS: The features computed from higher specificity MRI better described GOLM1 and RND1 gene-expression levels, while imaging features computed from lower specificity MRI data were more descriptive for the SETD7 gene. The GOLM1 regression model generated with three features demonstrated a moderate positive correlation (p<0.001), and the RND1 model developed with five variables was positively associated (p<0.001) with gene expression levels. Moreover, RND1 regression model integrating four features was moderately correlated with expressed RND1 levels (p<0.001). CONCLUSIONS: The results demonstrated that MRI radiomics features could help quantify GOLM1, SETD7, and RND1 expression levels noninvasively and predict the recurrence risk for early-stage HCC patients.

Association of radiomic imaging features and gene expression profile as prognostic factors in pancreatic ductal adenocarcinoma.

In this study, we investigated whether radiomic features of CT image data can accurately predict HMGA2 and C-MYC gene expression status and identify the patient survival time using a machine learning approach in pancreatic ductal adenocarcinoma (PDAC). A cohort of 111 patients with PDAC was enrolled in our study. Radiomic features were extracted using conventional (shape and texture analysis) and deep learning approaches following to segmentation of preoperative CT data. To predict patient survival time, significant radiomic features were identified using a log-rank test. After surgical resection, level of HMGA2 and C-MYC gene expressions of PDAC tumor regions were classified using a support vector machines method. The model was evaluated in terms of accuracy, sensitivity, specificity, and area under the curve (AUC). Besides, inter-reader reliability analysis was used to demonstrate the robustness of the proposed features. The identified features consistently achieved good performance in survival prediction and classification of gene expression status, on images segmented by different radiologists. Using CT data from 111 patients, six features in the segmented region of images were highly correlated with survival time. Using extracted deep features of excised lesions from 47 patients, we observed an average AUC score of 0.90 with an accuracy of 95% in C-MYC prediction (sensitivity: 92% and specificity: 98%). In HGMA2 group, using shape features, the average AUC score was measured as 0.91 with an accuracy of 88% (sensitivity: 89% and specificity: 88%). In conclusion, the radiomic features of CT image can accurately predict the expression status of HMGA2 and C-MYC genes and identify the survival time of PDAC patients.

[The clinical evaluate external route surgery of retinal detachment under surgical microscope].

PURPOSE: To evaluate the methods and the outcome of the retinal detachment surgery under surgical microscope. METHODS: Twenty-one consecutive cases of the rhegmatogenous retinal detachment were enrolled in this study. All received retinal detachment surgery under microscope. Location the retinal breaks, the drainage of subretinal fluid with diathermy acupuncture and retinal cryopexy through sclera were performed under surgical microscope. Buckling and or circling were performed in six cases. All the patients were followed up three to nine months. RESULTS: The reaction of retinal cryotherapy was seen clearly under the microscope. And the observation of cryotherapy and the location of retinal breaks were not affected by mild media opacities. No severe complication appeared. Twenty cases got complete retinal reattachment. One eye got reattachment after an additional scleral buckling. The visual acuity were improved in all cases, The visual acuity was better than 0.3 in 15 eyes (71.42%). CONCLUSIONS: It appeared to be simple, convenient, easy observable to perform retinal reattachment surgery under microscope.