Noninvasive identification of HER2-low-positive status by MRI-based deep learning radiomics predicts the disease-free survival of patients with breast cancer.

OBJECTIVE: This study aimed to establish a MRI-based deep learning radiomics (DLR) signature to predict the human epidermal growth factor receptor 2 (HER2)-low-positive status and further verified the difference in prognosis by the DLR model. METHODS: A total of 481 patients with breast cancer who underwent preoperative MRI were retrospectively recruited from two institutions. Traditional radiomics features and deep semantic segmentation feature-based radiomics (DSFR) features were extracted from segmented tumors to construct models separately. Then, the DLR model was constructed to assess the HER2 status by averaging the output probabilities of the two models. Finally, a Kaplan‒Meier survival analysis was conducted to explore the disease-free survival (DFS) in patients with HER2-low-positive status. The multivariate Cox proportional hazard model was constructed to further determine the factors associated with DFS. RESULTS: First, the DLR model distinguished between HER2-negative and HER2-overexpressing patients with AUCs of 0.868 and 0.763 in the training and validation cohorts, respectively. Furthermore, the DLR model distinguished between HER2-low-positive and HER2-zero patients with AUCs of 0.855 and 0.750, respectively. Cox regression analysis showed that the prediction score obtained using the DLR model (HR, 0.175; p = 0.024) and lesion size (HR, 1.043; p = 0.009) were significant, independent predictors of DFS. CONCLUSIONS: We successfully constructed a DLR model based on MRI to noninvasively evaluate the HER2 status and further revealed prospects for predicting the DFS of patients with HER2-low-positive status. CLINICAL RELEVANCE STATEMENT: The MRI-based DLR model could noninvasively identify HER2-low-positive status, which is considered a novel prognostic predictor and therapeutic target. KEY POINTS: • The DLR model effectively distinguished the HER2 status of breast cancer patients, especially the HER2-low-positive status. • The DLR model was better than the traditional radiomics model or DSFR model in distinguishing HER2 expression. • The prediction score obtained using the model and lesion size were significant independent predictors of DFS.

HRadNet: A Hierarchical Radiomics-Based Network for Multicenter Breast Cancer Molecular Subtypes Prediction.

Breast cancer is a heterogeneous disease, where molecular subtypes of breast cancer are closely related to the treatment and prognosis. Therefore, the goal of this work is to differentiate between luminal and non-luminal subtypes of breast cancer. The hierarchical radiomics network (HRadNet) is proposed for breast cancer molecular subtypes prediction based on dynamic contrast-enhanced magnetic resonance imaging. HRadNet fuses multilayer features with the metadata of images to take advantage of conventional radiomics methods and general convolutional neural networks. A two-stage training mechanism is adopted to improve the generalization capability of the network for multicenter breast cancer data. The ablation study shows the effectiveness of each component of HRadNet. Furthermore, the influence of features from different layers and metadata fusion are also analyzed. It reveals that selecting certain layers of features for a specified domain can make further performance improvements. Experimental results on three data sets from different devices demonstrate the effectiveness of the proposed network. HRadNet also has good performance when transferring to other domains without fine-tuning.

Radiologic Response Combined with Dermatologic Toxicities is the Most Robust Predictor of Survival Benefits in Patients with Inoperable Hepatocellular Carcinoma After Transarterial Chemoembolization Plus Sorafenib Therapy.

PURPOSE: The survival benefits of patients with inoperable hepatocellular carcinoma (HCC) who undergo transarterial chemoembolization (TACE) and receive sorafenib therapy remain controversial. We aimed to identify clinical predictors in patients with inoperable HCC undergoing TACE and receiving sorafenib. METHODS: Between January 2014 and December 2017, 148 consecutive patients with inoperable HCC who were treated with TACE plus sorafenib were retrospectively analyzed. Critical clinical factors associated with overall survival (OS) were identified by Cox regression model analysis. Kaplan-Meier methods were used to calculate the survival times, which were compared with the log-rank test. RESULTS: Macrovascular invasion (MVI), radiologic response and sorafenib-related dermatologic toxicities were identified as independent factors associated with OS. MVI is a known prognostic factor before treatment. The median OS of patients with either radiologic response or dermatologic toxicities was significantly improved compared with that of patients without it (both 23.0 vs. 7.0 months, P < 0.001). The median OS of patients with a combination of radiologic response and dermatologic toxicities was significantly longer than that of patients with either radiologic response or dermatologic toxicities, as well as no response (25.0 vs. 14.0 vs. 6.0 months, respectively, P < 0.001), and the predictive value was confirmed across patients with different baseline characteristics in terms of MVI, α-fetoprotein level, performance status and liver function. CONCLUSION: The combination of radiologic response and sorafenib-related dermatologic toxicities is the most robust predictor of survival benefits for HCC patients after TACE plus sorafenib therapy. LEVEL OF EVIDENCE: Level 3.

Additive value of texture analysis based on breast MRI for distinguishing between benign and malignant non-mass enhancement in premenopausal women.

BACKGROUND: Non-mass enhancement (NME) is a diagnostic dilemma and highly reliant on the experience of the radiologists. Texture analysis (TA) could serve as an objective method to quantify lesion characteristics. However, it remains unclear what role TA plays in a predictive model based on routine MRI characteristics. The purpose of this study was to explore the value of TA in distinguishing between benign and malignant NME in premenopausal women. METHODS: Women in whom NME was histologically proven (n = 147) were enrolled (benign: 58; malignant: 89) was retrospective. Then, 102 and 45 patients were classified as the training and validation groups, respectively. Scanning sequences included Fat-suppressed T2-weighted and fat-suppressed contrast-enhanced T1-weighted which were acquired on a 1.5T MRI system. Clinical and routine MR characteristics (CRMC) were evaluated by two radiologists according to the Breast Imaging and Reporting and Data system (2013). Texture features were extracted from all post-contrast sequences in the training group. The combination model was built and then assessed in the validation group. Pearson's chi-square test and Mann-Whitney U test were used to compare categorical variables and continuous variables, respectively. Logistic regression analysis and receiver operating characteristic curve were employed to assess the diagnostic performance of CRMC, TA, and their combination model in NME diagnosis. RESULTS: The combination model showed superior diagnostic performance in differentiating between benign and malignant NME compared to that of CRMC or TA alone (AUC, 0.887 vs 0.832 vs 0.74). Moreover, compared to CRMC, the model showed high specificity (72.5% vs 80%). The results obtained in the validation group confirmed the model was promising. CONCLUSIONS: With the combined use of TA and CRMC could afford an improved diagnostic performance in differentiating between benign and malignant NME.

Radiation-induced occult insufficiency fracture or bone metastasis after radiotherapy for cervical cancer? The nomogram based on quantitative apparent diffusion coefficients for discrimination.

BACKGROUND: Radiation-induced insufficiency fractures (IF) is frequently occult without fracture line, which may be mistaken as metastasis. Quantitative apparent diffusion coefficient (ADC) shows potential value for characterization of benign and malignant bone marrow diseases. The purpose of this study was to develop a nomogram based on multi-parametric ADCs in the differntiation of occult IF from bone metastasis after radiotherapy (RT) for cervical cancer. METHODS: This study included forty-seven patients with cervical cancer that showed emerging new bone lesions in RT field during the follow-up. Multi-parametric quantitative ADC values were measured for each lesion by manually setting region of interests (ROIs) on ADC maps, and the ROIs were copied to adjacent normal muscle and bone marrow. Six parameters were calculated, including ADCmean, ADCmin, ADCmax, ADCstd, ADCmean ratio (lesion/normal bone) and ADCmean ratio (lesion/muscle). For univariate analysis, receiver operating characteristic curve (ROC) analysis was performed to assess the performance. For combined diagnosis, a nomogram model was developed by using a multivariate logistic regression analysis. RESULTS: A total of 75 bone lesions were identified, including 48 occult IFs and 27 bone metastases. There were significant differences in the six ADC parameters between occult IFs and bone metastases (p < 0.05), the ADC ratio (lesion/ muscle) showed an optimal diagnostic efficacy, with an area under ROC (AUC) of 0.887, the sensitivity of 95.8%, the specificity of 81.5%, respectively. Regarding combined diagnosis, ADCstd and ADCmean ratio (lesion/muscle) were identified as independent factors and were selected to generate a nomogram model. The nomogram model showed a better performance, yielded an AUC of 0.92, the sensitivity of 91.7%, the specificity of 96.3%, positive predictive value (PPV) of 97.8% and negative predictive value (NPV) of 86.7%, respectively. CONCLUSIONS: Multi-parametric ADC values demonstrate potential value for differentiating occult IFs from bone metastasis, a nomogram based on the combination of ADCstd and ADCmean ratio (lesion/muscle) may provide an improved classification performance.

Case report: A tracheobronchial schwannoma in a child.

Neurogenic tumors of the tracheobronchial tree are extremely rare, and these include neurofibroma and schwannoma. The rare schwannoma most frequently is reported in adults. We will report an endobronchial schwannoma in an 11-year-old boy.

Pelvic insufficiency fracture or bone metastasis after radiotherapy for cervical cancer? The added value of DWI for characterization.

OBJECTIVES: We sought to determine the added value of diffusion-weighted magnetic resonance imaging (DWI) in the differentiation of pelvic insufficiency fracture (PIF) from bone metastasis after radiotherapy in cervical cancer patients. METHODS: In the present study, 42 cervical cancer patients after radiotherapy with 61 bone lesions (n = 40, PIFs; n = 21, bone metastasis) were included. Conventional MRI and DWI were performed in all patients. For qualitative imaging diagnosis, two sets of images were reviewed independently by three observers, including a conventional MRI set (unenhanced T1-weighted, T2-weighted, and enhanced T1-weighted images) and a DWI set (conventional MRIs, DW images, and ADC maps). The mean ADC value of each lesson was measured on ADC maps. The diagnostic performance was assessed by using the area under the receiver operating characteristic curve (Az), and sensitivity and specificity were determined. RESULTS: For all observers, the Az value and sensitivity of the DWI set showed improvement compared with the conventional MRI set. The observer who had the least experience (3 years) demonstrated significant improvement in diagnostic performance with the addition of DWI; Az value increased from 0.804 to 0.915 (p = 0.042) and sensitivity increased from 75.0 to 92.5% (p = 0.035). The mean ADCs of the PIFs were significantly higher than the bone metastases (p < 0.001); ADC values > 0.97 × 10-3 mm2/s yielded an Az of 0.887, a sensitivity of 92.5%, and a specificity of 76.2%. CONCLUSIONS: The addition of DWI to conventional MRI improved the differentiation of PIF from bone metastasis after RT in patients with cervical cancer. KEY POINTS: • DWI showed additive value to conventional MRI in the differentiation of PIF from bone metastasis after RT. • For qualitative diagnosis, the addition of DWI can improve diagnostic performance compared with conventional MRI alone and can particularly improve the sensitivity. • Quantitative ADC assessment showed potential value for identifying PIF from bone metastasis.

The Utility of Texture Analysis Based on Breast Magnetic Resonance Imaging in Differentiating Phyllodes Tumors From Fibroadenomas.

Background: The preoperative diagnosis of phyllodes tumors (PTs) of the breast is critical to appropriate surgical treatment. However, reliable differentiation between PT and fibroadenoma (FA) remains difficult in daily clinical practice. The purpose of this study was to investigate the utility of breast MRI texture analysis for differentiating PTs from FAs. Materials and Methods: Forty-two PTs and 42 FAs were enrolled in this retrospective study. Clinical and conventional MRI features (CCMF) and MRI texture analysis were used to distinguish between PT and FA. Texture features were extracted from the axial short TI inversion recovery T2-weighted (T2W-STIR), T1-weighted pre-contrast, and two contrast-enhanced series (first contrast and third contrast). The Mann-Whitney U test was used to select statistically significant features of texture analysis and CCMF. Using a linear discriminant analysis, the most discriminative features were determined from statistically significant features. The K-nearest neighbor classifier and ROC curve were applied to evaluate the diagnostic performance. Results: With a higher classification accuracy (89.3%) and an AUC of 0.89, the texture features on T2W-STIR outperformed the texture features on other MRI sequences and CCMF. The AUC of the combination of CCMF with texture features on T2W-STIR was significantly higher than that of CCMF or texture features on T2W-STIR alone (p < 0.05). Based on the result of the classification accuracy (95.2%) and AUC (0.95), the diagnostic performance of the combination strategy performed better than texture features on T2W-STIR or CCMF separately. Conclusions: Texture features on T2W-STIR showed better diagnostic performance compared to CCMF for the distinction between PTs and FAs. After further validation of multi-institutional large datasets, MRI-based texture features may become a potential biomarker and be a useful medical decision tool in clinical trials having patients with breast fibroepithelial neoplasms.

Conservative Management of Retroperitoneal Ectopic Pregnancy by Computed Tomographic-guided Methotrexate Injection in the Gestational Sac: 2 Case Reports and Literature Review.

Retroperitoneal ectopic pregnancy (REP) is an extremely rare type of ectopic pregnancy. Currently, surgery is the most widely used treatment method although it involves a high likelihood of intraoperative hemorrhage. In this case report, we describe a safe and effective alternative method for managing epigastric REP. We conducted a retrospective analysis of the clinical data of 2 patients with REP in the epigastrium who were treated at our hospital using our nonsurgical method. The treatment involved conservative management by computed tomographic-guided methotrexate injection in the gestational sac. We also present a literature review of 26 case reports and discuss the clinical features and various methods for treating REP. Our experience with the successful treatment of 2 patients suggests that the novel approach of computed tomographic-guided methotrexate injection in the gestational sac may be a safe and effective approach to manage REP. Further studies are warranted to confirm our findings.

[Imaging and pathological features of gastrointestinal stromal tumors].

OBJECTIVE: To investigate the imaging features and pathological manifestations of gastrointestinal stromal tumors (GISTs). METHODS: The imaging characteristics and pathological manifestations of 26 surgically treated patients with histologically confirmed GISTs were retrospectively analyzed. RESULTS: The tumors were found to originate from the small bowel (n=10), stomach (n=8), colon (n=6), mesentery (n=1) and omentum (n=1). The imaging and pathological features of GISTs were (1) most of GISTs were well-defined and exophytic (n=19, 73.1% ), which usually compressed the adjacent tissues but no invasion. The tumor diameter ranged from 5.1 to 23.5 cm with a mean diameter of 11.6 +/- 5.9 cm, (2)most tumors had an inhomogenous density or signal intensity due to necrosis(n=21, 80.8%), hemorrhage (n=15, 57.7%) or calcification (n=3, 11.5%) within the tumor, (3) on the CT or MR images, heterogeneous enhancement pattern presented as peripheral or intra-tumor patchy enhancement was common (n=21, 80.8%). Furthermore, enhanced striped vessels were seen in 12 patients. However, homogenous enhancement pattern was rare (n=5, 19.2%), (4) the most common site where GIST metastasized to was the liver (n=7), followed by the peritoneum (n=4), but rarely to lymph nodes, (5) of these 26 patients, spindle-cell type was observed in 69.2% (n=18), epithelioid-cell type in 23.1% (n=6), and mixed cell type in 7.7% (n=2). Immunohistochemical studies showed positive CD117 expression in all of these 26 patients, but positive CD34 expression in only 22 patients. CONCLUSION: Gastrointestinal stromal tumor usually presents as a exophytic, well-defined large tumor, with internal striped vessels, necrosis or hemorrhage within. It usually metastasizes to the liver or the peritoneum but rarely to lymph nodes. Pathologically, most of gastrointestinal stromal tumors consist of spindle-cells, while a small portion of the tumors are composed of epithelioid-cells or mixed ones. Both CT and MRI play an important role in the diagnosis of gastrointestinal stromal tumors.