Purpose To establish and validate radiomics models for predicting the early efficacy (less than 3 months) of microwave ablation (MWA) in malignant lung tumors. Methods The study enrolled 130 malignant lung tumor patients (72 in the training cohort, 32 in the testing cohort, and 26 in the validation cohort) treated with MWA. Post-operation CT images were analyzed. To evaluate the therapeutic effect of ablation, three models were constructed by least absolute shrinkage and selection operator and logistic regression: the tumoral radiomics (T-RO), peritumoral radiomics (P-RO), and tumoral-peritumoral radiomics (TP-RO) models. Univariate and multivariate analyses were performed to identify clinical variables and radiomics features associated with early efficacy, which were incorporated into the combined radiomics (C-RO) model. The performance of the C-RO model was evaluated by the area under the receiver operating characteristic (ROC) curve (AUC), calibration curve, and decision curve analysis (DCA). The C-RO model was used to derive the best cutoff value of ROC and to distinguish the high-risk group (Nomo-score of C-RO model below than cutoff value) from the low-risk group (Nomo-score of C-RO model higher than cutoff value) for survival analysis of patients. Results Four radiomics features were selected from the region of interest of tumoral and peritumoral CT images, which showed good performance for evaluating prognosis and early efficacy in three cohorts. The C-RO model had the highest AUC value in all models, and the C-RO model was better than the P-RO model (AUC in training, 0.896 vs. 0.740; p = 0.036). The DCA confirmed the clinical benefit of the C-RO model. Survival analysis revealed that in the C-RO model, the low-risk group defined by best cutoff value had significantly better progression-free survival than the high-risk group (p<0.05). Conclusions CT-based radiomics models in malignant lung tumor patients after MWA could be useful for individualized risk classification and treatment.

No abstract

PurposeFocal lesions of the liver are usually detected by enhanced CT and further diagnosed by enhanced MR in clinical practice. The harmful effects of repeated contrast use in CT and MR, and the subjectivity of conventional imaging increase the risk of misdiagnosis. Our aim is to establish a radiomics nomogram based on T2-weighted imaging for differentiating hepatocellular carcinoma and benign liver lesions with rich blood supply and to estimate the enhancive value to the traditional imaging diagnosis.MethodsThe retrospective study analyzed the imaging and clinical data of 144 patients with hepatocellular carcinoma (n=101) and benign blood-rich lesions of the liver (n=43) pathologically confirmed. These patients were randomly assigned to the training cohort (n=100) and the validation cohort (n=44). We developed three prediction models - a radiomic model, a clinical model, and a fusion model that combined radiomics score (Rad-score) with clinical factors. Comparing the predictive performance of three models, we obtained the best prediction model, which was then compared with the diagnostic efficacy of junior and senior radiologists. The efficacy was evaluated using the area under receiver operating characteristic curve (ROC).ResultsFour radiomics features and three clinical factors (age, sex, lesion location) were chosen for construction of the radiomics model and the clinical model, respectively. Comparing to the radiomics model and the clinical model, the fusion model showed significant discrimination capability in the training set (AUC, 0.972; 95%CI 0.918 - 0.995) and the validation set (AUC, 0.943; 95%CI 0.829 - 0.990). And it was statistically better than the junior radiologist and the senior radiologist in the training cohort (p=0.007 and p=0.005, respectively).ConclusionsThe T2WI-based radiomics nomogram greatly complements the flaw of traditional imaging diagnosis and avoid the reuse of contrast agents. It might facilitate early clinical diagnosis and precision treatment with performed exceedingly favorable predictive efficacy in differentiating HCC and BLLs with rich blood supply.

Purpose: In this study, a non-invasive radiomics method was used to extract a large amount of valuable information from patient CT images (including pre-treatment and follow-up images), and the image features were effectively combined with treatment effect evaluation, aiming to solve the problem of lung adenocarcinoma in patients with EGFR mutation positive. The early efficacy evaluation of cancer EGFR-TKI targeted drugs provides a scientific basis for clinical drug selection and adjustment of treatment plans. Methods: It is planned to include 106 patients with EGFR-TKI targeted therapy for lung adenocarcinoma in Shaoxing People's Hospital (pathologically confirmed EGFR mutation positive), and randomly divide them into a training group and a validation group in a ratio of 7:3. The CT plain scan images of patients before and 2 months after treatment were collected, and the images were segmented by ITK-SNAP software. Perform feature screening and modeling, and draw nomograms, calibration curves, and decision curves. Results: After Lasso dimensionality reduction, three features with non-zero coefficients are extracted between groups, sumEntropy,LongRunLowGreyLevelEmphasis_angle0_offset1, LongRunLowGreyLevelEmphasis_angle0_offset4. The average area under the curve (AUC) of the logistic regression model in the training group and the validation group was 0.778 and 0.773, respectively; the Nomogram chart showed that after the three radiomics parameters were scored, the total score was 0-130, and the corresponding effective The risk was 0.1-0.9; the calibration curve showed that the predicted value and the actual clinical observation value approached a 45° slope, indicating that the model was well calibrated; while the decision curve radiomics model was used to evaluate EGFR-TKI in lung adenocarcinoma Targeted drugs have better clinical benefits when they have early efficacy. Conclusions: The radiomics model established based on the radiomics features of CT unenhanced images has good performance in evaluating the efficacy of EGFR-TKI targeted drugs in the treatment of lung adenocarcinoma, and the final Nomogram map can quantitatively evaluate the curative effect, providing individual patient information.