No abstract

PurposeTo develop and validate a clinical-radiomics nomogram based on radiomics features and clinical risk factors for identification of human epidermal growth factor receptor 2 (HER2) status in patients with breast cancer (BC).MethodsTwo hundred and thirty-five female patients with BC were enrolled from July 2018 to February 2022 and divided into a training group (from center I, 115 patients), internal validation group (from center I, 49 patients), and external validation group (from centers II and III, 71 patients). The preoperative MRI of all patients was obtained, and radiomics features were extracted by a free open-source software called 3D Slicer. The Least Absolute Shrinkage and Selection Operator regression model was used to identify the most useful features. The radiomics score (Rad-score) was calculated by using the radiomics signature-based formula. A clinical-radiomics nomogram combining clinical factors and Rad-score was developed through multivariate logistic regression analysis. The performance of the nomogram was evaluated using receiver operating characteristic (ROC) curve and decision curve analysis (DCA).ResultsA total of 2,553 radiomics features were extracted, and 21 radiomics features were selected as the most useful radiomics features. Multivariate logistic regression analysis indicated that Rad-score, progesterone receptor (PR), and Ki-67 were independent parameters to distinguish HER2 status. The clinical-radiomics nomogram, which comprised Rad-score, PR, and Ki-67, showed a favorable classification capability, with AUC of 0.87 [95% confidence internal (CI), 0.80 to 0.93] in the training group, 0.81 (95% CI, 0.69 to 0.94) in the internal validation group, and 0.84 (95% CI, 0.75 to 0.93) in the external validation group. DCA illustrated that the nomogram was useful in clinical practice.ConclusionsThe nomogram combined with Rad-score, PR, and Ki-67 can identify the HER2 status of BC.

BackgroundBrain metastasis (BM) is a serious neurological complication of cancer of different origins. The value of deep learning (DL) to identify multiple types of primary origins remains unclear.PurposeTo distinguish primary site of BM and identify the best DL models.Study TypeRetrospective.PopulationA total of 449 BM derived from 214 patients (49.5% for female, mean age 58 years) (100 from small cell lung cancer [SCLC], 125 from non‐small cell lung cancer [NSCLC], 116 from breast cancer [BC], and 108 from gastrointestinal cancer [GIC]) were included.Field Strength/SequenceA 3‐T, T1 turbo spin echo (T1‐TSE), T2‐TSE, T2FLAIR‐TSE, DWI echo‐planar imaging (DWI‐EPI) and contrast‐enhanced T1‐TSE (CE T1‐TSE).AssessmentLesions were divided into training (n = 285, 153 patients), testing (n = 122, 93 patients), and independent testing cohorts (n = 42, 34 patients). Three‐dimensional residual network (3D‐ResNet), named 3D ResNet6 and 3D ResNet 18, was proposed for identifying the four origins based on single MRI and combined MRI (T1WI + T2‐FLAIR + DWI, CE‐T1WI + DWI, CE‐T1WI + T2WI + DWI). DL model was used to distinguish lung cancer from non‐lung cancer; then SCLC vs. NSCLC for lung cancer classification and BC vs. GIC for non‐lung cancer classification was performed. A subjective visual analysis was implemented and compared with DL models. Gradient‐weighted class activation mapping (Grad‐CAM) was used to visualize the model by heatmaps.Statistical TestsThe area under the receiver operating characteristics curve (AUC) assess each classification performance.Results3D ResNet18 with Grad‐CAM and AIC showed better performance than 3DResNet6, 3DResNet18 and the radiologist for distinguishing lung cancer from non‐lung cancer, SCLC from NSCLC, and BC from GIC. For single MRI sequence, T1WI, DWI, and CE‐T1WI performed best for lung cancer vs. non‐lung cancer, SCLC vs. NSCLC, and BC vs. GIC classifications. The AUC ranged from 0.675 to 0.876 and from 0.684 to 0.800 regarding the testing and independent testing datasets, respectively. For combined MRI sequences, the combination of CE‐T1WI + T2WI + DWI performed better for BC vs. GIC (AUCs of 0.788 and 0.848 on testing and independent testing datasets, respectively), while the combined MRI approach (T1WI + T2‐FLAIR + DWI, CE‐T1WI + DWI) could not achieve higher AUCs for lung cancer vs. non‐lung cancer, SCLC vs. NSCLC. Grad‐CAM helped for model visualization by heatmaps that focused on tumor regions.Data ConclusionDL models may help to distinguish the origins of BM based on MRI data.Evidence Level3Technical EfficacyStage 2.