An [18F]FDG PET/3D-ultrashort echo time MRI-based radiomics model established by machine learning facilitates preoperative assessment of lymph node status in non-small cell lung cancer.

OBJECTIVES: To develop an [18F]FDG PET/3D-UTE model based on clinical factors, three-dimensional ultrashort echo time (3D-UTE), and PET radiomics features via machine learning for the assessment of lymph node (LN) status in non-small cell lung cancer (NSCLC). METHODS: A total of 145 NSCLC patients (training, 101 cases; test, 44 cases) underwent whole-body [18F]FDG PET/CT and chest [18F]FDG PET/MRI were enrolled. Preoperative clinical factors and 3D-UTE, CT, and PET radiomics features were analyzed. The Mann-Whitney U test, LASSO regression, and SelectKBest were used for feature extraction. Five machine learning algorithms were used to establish prediction models, which were evaluated by the area under receiver-operator characteristic (ROC), DeLong test, calibration curves, and decision curve analysis (DCA). RESULTS: A prediction model based on random forest, consisting of four clinical factors, six 3D-UTE, and six PET radiomics features, was used as the final model for PET/3D-UTE. The AUCs of this model were 0.912 and 0.791 in the training and test sets, respectively, which not only showed different degrees of improvement over individual models such as clinical, 3D-UTE, and PET (AUC-training = 0.838, 0.834, and 0.828, AUC-test = 0.756, 0.745, and 0.768, respectively) but also achieved the similar diagnostic efficacy as the optimal PET/CT model (AUC-training = 0.890, AUC-test = 0.793). The calibration curves and DCA indicated good consistency (C-index, 0.912) and clinical utility of this model, respectively. CONCLUSION: The [18F]FDG PET/3D-UTE model based on clinical factors, 3D-UTE, and PET radiomics features using machine learning methods could noninvasively assess the LN status of NSCLC. CLINICAL RELEVANCE STATEMENT: A machine learning model of 18F-fluorodeoxyglucose positron emission tomography/ three-dimensional ultrashort echo time could noninvasively assess the lymph node status of non-small cell lung cancer, which provides a novel method with less radiation burden for clinical practice. KEY POINTS: • The 3D-UTE radiomics model using the PLS-DA classifier was significantly associated with LN status in NSCLC and has similar diagnostic performance as the clinical, CT, and PET models. • The [18F]FDG PET/3D-UTE model based on clinical factors, 3D-UTE, and PET radiomics features using the RF classifier could noninvasively assess the LN status of NSCLC and showed improved diagnostic performance compared to the clinical, 3D-UTE, and PET models. • In the assessment of LN status in NSCLC, the [18F]FDG PET/3D-UTE model has similar diagnostic efficacy as the [18F]FDG PET/CT model that incorporates clinical factors and CT and PET radiomics features.

Predicting the grade of hepatocellular carcinoma based on non-contrast-enhanced MRI radiomics signature.

PURPOSE: This study was conducted in order to investigate the value of magnetic resonance imaging (MRI)-based radiomics signatures for the preoperative prediction of hepatocellular carcinoma (HCC) grade. METHODS: Data from 170 patients confirmed to have HCC by surgical pathology were divided into a training group (n = 125) and a test group (n = 45). The radiomics features of tumours based on both T1-weighted imaging (WI) and T2WI were extracted by using Matrix Laboratory (MATLAB), and radiomics signatures were generated using the least absolute shrinkage and selection operator (LASSO) logistic regression model. The predicted values of pathological HCC grades using radiomics signatures, clinical factors (including age, sex, tumour size, alpha fetoprotein (AFP) level, history of hepatitis B, hepatocirrhosis, portal vein tumour thrombosis, portal hypertension and pseudocapsule) and the combined models were assessed. RESULTS: Radiomics signatures could successfully categorise high-grade and low-grade HCC cases (p < 0.05) in both the training and test datasets. Regarding the performances of clinical factors, radiomics signatures and the combined clinical and radiomics signature (from the combined T1WI and T2WI images) models for HCC grading prediction, the areas under the curve (AUCs) were 0.600, 0.742 and 0.800 in the test datasets, respectively. Both the AFP level and radiomics signature were independent predictors of HCC grade (p < 0.05). CONCLUSIONS: Radiomics signatures may be important for discriminating high-grade and low-grade HCC cases. The combination of the radiomics signatures with clinical factors may be helpful for the preoperative prediction of HCC grade. KEY POINTS: • The radiomics signature based on non-contrast-enhanced MR images was significantly associated with the pathological grade of HCC. • The radiomics signatures based on T1WI or T2WI images performed similarly at predicting the pathological grade of HCC. • Combining the radiomics signature and clinical factors (including age, sex, tumour size, AFP level, history of hepatitis B, hepatocirrhosis, portal vein tumour thrombosis, portal hypertension and pseudocapsule) may be helpful for the preoperative prediction of HCC grade.

[Quantitative analysis of hepatocellular carcinomas pathological grading in non-contrast magnetic resonance images].

In order to solve the pathological grading of hepatocellular carcinomas (HCC) which depends on biopsy or surgical pathology invasively, a quantitative analysis method based on radiomics signature was proposed for pathological grading of HCC in non-contrast magnetic resonance imaging (MRI) images. The MRI images were integrated to predict clinical outcomes using 328 radiomics features, quantifying tumour image intensity, shape and text, which are extracted from lesion by manual segmentation. Least absolute shrinkage and selection operator (LASSO) were used to select the most-predictive radiomics features for the pathological grading. A radiomics signature, a clinical model, and a combined model were built. The association between the radiomics signature and HCC grading was explored. This quantitative analysis method was validated in 170 consecutive patients (training dataset: n = 125; validation dataset, n = 45), and cross-validation with receiver operating characteristic (ROC) analysis was performed and the area under the ROC curve (AUC) was employed as the prediction metric. Through the proposed method, AUC was 0.909 in training dataset and 0.800 in validation dataset, respectively. Overall, the prediction performances by radiomics features showed statistically significant correlations with pathological grading. The results showed that radiomics signature was developed to be a significant predictor for HCC pathological grading, which may serve as a noninvasive complementary tool for clinical doctors in determining the prognosis and therapeutic strategy for HCC.

Radiologic and Clinicopathologic Findings of Inflammatory Myofibroblastic Tumor.

PURPOSE: The aim of the study was to describe the clinical, radiographic, and pathologic features of inflammatory myofibroblastic tumor (IMT) to enhance the recognition of this rare disease. MATERIALS AND METHODS: The clinical, imaging, and pathologic findings were retrospectively reviewed in 54 patients with IMT lesions, which were conformed by biopsy or surgical pathology. Of 54 patients, 51 had preoperative computed tomography (CT) examination and 13 had preoperative magnetic resonance imaging records. RESULTS: The clinical appearances of these 54 patients had some relationship with the locations of lesions. Of 54 IMT patients, 87.0% cases (47/54) had solitary lesion. The mean long diameter of the lesions located at the sites of chest, abdomen, and pelvic regions was bigger than that of other locations (F = 3.025, P = 0.038). On plain CT images, soft tissue mass was found in all IMT lesions, except for 3 lesions that arose in the intestine tract, appearing as focal or diffuse thickening in the bowel wall. After contrast administration, all lesions were persistently enhanced; 72.7% cases (24/33) demonstrated heterogeneous enhancement with various cystic regions. Comparing the CT features with different anatomic lesions, ill-defined margin on the plain CT images and calcification were seen more frequently in the lesions of the head and neck (P = 0.010 and 0.035); however, the other radiological findings had no significant differences (all P > 0.05). Twelve of 51 IMT patients showed invasion into adjacent structures. On magnetic resonance imaging, 92.3% lesions (12/13) showed soft tissue masses demonstrating isointense to hypointense contrast compared with skeletal muscle on T1-weighted images and heterogeneously high signals on T2-weighted images; 85.7%(6/7) of lesions were heterogeneously enhanced with cystic changes. Immunohistochemistry showed that the percentage of positive staining for SMA, vimentin, anaplastic lymphoma kinase, CD68, CD34, CD99, B-cell lymphoma/leukemia-2, cytokeratin, Desmin, and S-100 protein were 88.9%, 87.0%, 44.4%, 59.3%, 53.7%, 29.6%, 42.6%, 28.5%, 13.0%, and 24.1%, respectively. CONCLUSIONS: Inflammatory myofibroblastic tumor can involve any part of the body, and the clinical and radiological appearances are various owing to different anatomic sites. An ill-defined soft tissue mass heterogeneous enhancement with or without invasion into adjacent structures on computed tomographic or magnetic resonance images and positive staining for SMA and vimentin on immunohistochemical examination could suggest the diagnosis.

A pilot study using low-dose Spectral CT and ASIR (Adaptive Statistical Iterative Reconstruction) algorithm to diagnose solitary pulmonary nodules.

BACKGROUND: Lung cancer is the most common cancer which has the highest mortality rate. With the development of computed tomography (CT) techniques, the case detection rates of solitary pulmonary nodules (SPN) has constantly increased and the diagnosis accuracy of SPN has remained a hot topic in clinical and imaging diagnosis. The aim of this study was to evaluate the combination of low-dose spectral CT and ASIR (Adaptive Statistical Iterative Reconstruction) algorithm in the diagnosis of solitary pulmonary nodules (SPN). METHODS: 62 patients with SPN (42 cases of benign SPN and 20 cases of malignant SPN, pathology confirmed) were scanned by spectral CT with a dual-phase contrast-enhanced method. The iodine and water concentration (IC and WC) of the lesion and the artery in the image that had the same density were measured by the GSI (Gemstone Spectral Imaging) software. The normalized iodine and water concentration (NIC and NWC) of the lesion and the normalized iodine and water concentration difference (ICD and WCD) between the arterial and venous phases (AP and VP) were also calculated. The spectral HU (Hounsfield Unit ) curve was divided into 3 sections based on the energy (40-70, 70-100 and 100-140 keV) and the slopes (λHU) in both phases were calculated. The ICAP, ICVP, WCAP and WCVP, NIC and NWC, and the λHU in benign and malignant SPN were compared by independent sample t-test. RESULTS: The iodine related parameters (ICAP, ICVP, NICAP, NICVP, and the ICD) of malignant SPN were significantly higher than that of benign SPN (t = 3.310, 1.330, 2.388, 1.669 and 3.251, respectively, P <0.05). The 3 λHU values of venous phase in malignant SPN were higher than that of benign SPN (t = 3.803, 2.846 and 3.205, P <0.05). The difference of water related parameters (WCAP, WCVP, NWCAP, NWCVP and WCD) between malignant and benign SPN were not significant (t = 0.666, 0.257, 0.104, 0.550 and 0.585, P > 0.05). CONCLUSIONS: The iodine related parameters and the slope of spectral curve are useful markers to distinguish the benign from the malignant lung diseases, and its application is extremely feasible in clinical applications.

Radiologic and clinicopathologic findings of peripheral primitive neuroectodermal tumors.

BACKGROUND: Primitive neuroectodermal tumors (PNETs) constitute a rare type of malignant neuroectodermal tumors that have chromosomal translocations identical to Ewing's sarcoma (ES), and the characteristics of this disease remain unclear. PURPOSE: To describe the clinical, radiological, and pathological features of peripheral PNETs (pPNETs) to enhance their recognition. MATERIAL AND METHODS: The clinical, imaging, and pathologic findings of 35 patients with pPNETs were retrospectively reviewed, all being confirmed by biopsy or surgical pathology. All 35 patients had preoperative computed tomography (CT) examinations; 10 patients had preoperative magnetic resonance imaging (MRI) studies. RESULTS: Of 35 pPNET patients, 54.3% had a primary tumor in soft tissue, the others in bone. On plain CT images, 33 lesions demonstrated heterogeneous hypodense masses with multiple lamellar lower density, and with osteolytic destruction if the tumor originated in bone. Calcification was only found in five lesions arising in soft tissue. All lesions enhanced heterogeneously with varying areas of cystic changes, and all lesions in bone and 52.6% of lesions in soft tissue showed ill-defined margins after contrast administration. On MRI, these tumors appeared in conjunction with osteolytic bone destruction and irregular soft tissue masses iso- to hypointense to skeletal muscle on T1-weighted images and showed heterogeneously high intensity on T2-weighted images. All lesions enhanced heterogeneously with cystic changes. Homer-Wright rosettes were observed in 15 lesions, and 97.1% lesions were positive for CD99 in histopathological results. CONCLUSION: pPNETs can involve any part of the body, and a large, ill-defined, aggressive soft tissue mass and heterogeneous enhancement with or without osteolytic bone destruction on CT or MR images could suggest the diagnosis.

[The best noise index combined with ASIR weighting selection in low-dose chest scanning].

OBJECTIVE: To discuss the best noise index combined with ASIR weighting selection in low-dose chest scanning based on BMI. METHODS: 200 patients collected from May to December 2014 underwent non-contrast chest CT examinations, they were randomly assigned into standard dose group (Group A, NI15 combined with 30% ASIR) and low-dose groups (Group B, NI25 combined with 40% ASIR, Group C, NI30 combined with 50% ASIR, Group D, NI35 combined with 60% ASIR), 50 cases in each group; the patients were assigned into three groups based on BMI (kg/m2): BMI<18.5; 18.5≤BMI≤25; BMI>25. Signal-to-nosie ratio (SNR), contrast-to noise ratio (CNR), CT dose index volume (CTDIvol), dose-length product (DLP), effective dose (ED) and subjective scoring between the standard and low-dose groups were compared and analyzed statistically. Differences of SNR, CNR, CTDIvol, DLP and ED among groups were determined with ANOVA analysis and the consistency of diagnosis with Kappa test. RESULTS: SNR, CTDIvol, DLP and ED reduced with the increase of nosie index, the differences among the groups were statistically significant (P<0.05). Kappa value of the two reviewers were 0.888. Subjective scoring of four groups were all above 3 points in BMI<18.5 kg/m2 group; subjective scoring of ABC groups were all above 3 points in 18.5 kg/m2≤BMI≤25 kg/m2 group and subjective scoring of AB groups were all above 3 points in BMI>25 kg/m2 group. CONCLUSIONS: NI35 combined with 60% ASIR in BMI<18.5 kg/m2 group; NI30 combined with 50% ASIR in 18.5 kg/m2≤BMI≤25 kg/m2 group; NI25 combined with 40% ASIR in 18.5 kg/m2≤BMI≤25 kg/m2 group were the best parameters combination which both can significantly reduce the radiation dose and ensure the image quality.

Clinical study on the prediction of ALN metastasis based on intratumoral and peritumoral DCE-MRI radiomics and clinico-radiological characteristics in breast cancer.

Objective: To investigate the value of predicting axillary lymph node (ALN) metastasis based on intratumoral and peritumoral dynamic contrast-enhanced MRI (DCE-MRI) radiomics and clinico-radiological characteristics in breast cancer. Methods: A total of 473 breast cancer patients who underwent preoperative DCE-MRI from Jan 2017 to Dec 2020 were enrolled. These patients were randomly divided into training (n=378) and testing sets (n=95) at 8:2 ratio. Intratumoral regions (ITRs) of interest were manually delineated, and peritumoral regions of 3 mm (3 mmPTRs) were automatically obtained by morphologically dilating the ITR. Radiomics features were extracted, and ALN metastasis-related radiomics features were selected by the Mann-Whitney U test, Z score normalization, variance thresholding, K-best algorithm and least absolute shrinkage and selection operator (LASSO) algorithm. Clinico-radiological risk factors were selected by logistic regression and were also used to construct predictive models combined with radiomics features. Then, 5 models were constructed, including ITR, 3 mmPTR, ITR+3 mmPTR, clinico-radiological and combined (ITR+3 mmPTR+ clinico-radiological) models. The performance of models was assessed by sensitivity, specificity, accuracy, F1 score and area under the curve (AUC) of receiver operating characteristic (ROC), calibration curves and decision curve analysis (DCA). Results: A total of 2264 radiomics features were extracted from each region of interest (ROI), 3 and 10 radiomics features were selected for the ITR and 3 mmPTR, respectively. 5 clinico-radiological risk factors were selected, including lesion size, human epidermal growth factor receptor 2 (HER2) expression, vascular cancer thrombus status, MR-reported ALN status, and time-signal intensity curve (TIC) type. In the testing set, the combined model showed the highest AUC (0.839), specificity (74.2%), accuracy (75.8%) and F1 Score (69.3%) among the 5 models. DCA showed that it had the greatest net clinical benefit compared to the other models. Conclusion: The intra- and peritumoral radiomics models based on DCE-MRI could be used to predict ALN metastasis in breast cancer, especially for the combined model with clinico-radiological characteristics showing promising clinical application value.

Radiomics analysis of intratumoral and different peritumoral regions from multiparametric MRI for evaluating HER2 status of breast cancer: A comparative study.

Purpose: To investigate the potential of radiomics signatures (RSs) from intratumoral and peritumoral regions on multiparametric magnetic resonance imaging (MRI) to noninvasively evaluate HER2 status in breast cancer. Method: In this retrospective study, 992 patients with pathologically confirmed breast cancers who underwent preoperative MRI were enrolled. The breast cancer lesions were segmented manually, and the intratumor region of interest (ROIIntra) was dilated by 2, 4, 6 and 8 mm (ROIPeri2mm, ROIPeri4mm, ROIPeri6mm, and ROIPeri8mm, respectively). Quantitative radiomics features were extracted from dynamic contrast-enhanced T1-weighted imaging (DCE-T1), fat-saturated T2-weighted imaging (T2) and diffusion-weighted imaging (DWI). A three-step procedure was performed for feature selection, and RSs were constructed using a support vector machine (SVM) to predict HER2 status. Result: The best single-area RSs for predicting HER2 status were DCE_Peri4mm-RS, T2_Peri4mm-RS, and DWI_Peri4mm-RS, yielding areas under the curve (AUCs) of 0.716 (95% confidence interval (CI), 0.648-0.778), 0.706 (95% CI, 0.637-0.768), and 0.719 (95% CI, 0.651-0.780), respectively, in the test set. The optimal RSs combining intratumoral and peritumoral regions for evaluating HER2 status were DCE-T1_Intra + DCE_Peri4mm-RS, T2_Intra + T2_Peri6mm-RS and DWI_Intra + DWI_Peri4mm-RS, with AUCs of 0.752 (95% CI, 0.686-0.810), 0.754 (95% CI, 0.688-0.812) and 0.725 (95% CI, 0.657-0.786), respectively, in the test set. Combining three sequences in the ROIIntra, ROIPeri2mm, ROIPeri4mm, ROIPeri6mm and ROIPeri8mm areas, the optimal RS was DCE-T1_Peri4mm + T2_Peri4mm + DWI_Peri4mm-RS, achieving an AUC of 0.795 (95% CI, 0.733-0.849) in the test set. Conclusion: This study systematically explored the influence of the intratumoral region, different peritumoral sizes and their combination in radiomics analysis for predicting HER2 status in breast cancer based on multiparametric MRI and found the optimal RS.

Localization and evaluation of sentinel lymph node in breast cancer from computed tomographic lymphography.

OBJECTIVE: This study aimed to determine the accuracy of computed tomographic (CT) localization and CT-based diagnosis of sentinel lymph nodes (SLNs) metastasis. METHODS: Thirty-four patients with confirmed breast cancer underwent 40-row CT scanning, and the first one or several lymph node(s) in the lymphatic drainage pathway was/were defined as the SLN(s). Dye and γ probe-guided SLN biopsy was performed on all patients. To accurately localize the SLN, 19 patients (55.9%) underwent the percutaneous lymph node puncture procedure. The morphologic features of all the SLNs on CT scans were analyzed and compared with the SLN biopsy pathologic diagnosis. RESULTS: Sentinel lymph nodes were successfully identified for all patients without any significant adverse effects. All localized SLNs corresponded well with SLNs identified on SLN biopsy, with an accuracy of 89.5%. Accuracy increased to 100% when the CT scan technique was combined with the blue dye method. The size criteria for metastatic diagnosis had a sensitivity of 85%, which increased to 94.7% when long-to-short-axis ratio and margin characteristics were also considered. CONCLUSIONS: The CT lymphography combined with the blue dye method accurately localized the SLNs. The CT-based diagnostic criteria improved the diagnostic accuracy of SLN metastases and were useful for evaluating the axillary status in early stage breast cancer patients.

MiR-20a-5p functions as a potent tumor suppressor by targeting PPP6C in acute myeloid leukemia.

Acute myeloid leukemia (AML) is as a highly aggressive and heterogeneous hematological malignancy. MiR-20a-5p has been reported to function as an oncogene or tumor suppressor in several tumors, but the clinical significance and regulatory mechanisms of miR-20a-5p in AML cells have not been fully understood. In this study, we found miR-20a-5p was significantly decreased in bone marrow from AML patients, compared with that in healthy controls. Moreover, decreased miR-20a-5p expression was correlated with risk status and poor survival prognosis in AML patients. Overexpression of miR-20a-5p suppressed cell proliferation, induced cell cycle G0/G1 phase arrest and apoptosis in two AML cell lines (THP-1 and U937) using CCK-8 assay and flow cytometry analysis. Moreover, miR-20a-5p overexpression attenuated tumor growth in vivo by performing tumor xenograft experiments. Luciferase reporter assay and western blot demonstrated that protein phosphatase 6 catalytic subunit (PPP6C) as a target gene of miR-20a-5p was negatively regulated by miR-20a-5p in AML cells. Furthermore, PPP6C knockdown imitated, while overexpression reversed the effects of miR-20a-5p overexpression on AML cell proliferation, cell cycle G1/S transition and apoptosis. Taken together, our findings demonstrate that miR-20a-5p/PPP6C represent a new therapeutic target for AML and a potential diagnostic marker for AML therapy.

Radiomics Signatures Based on Multiparametric MRI for the Preoperative Prediction of the HER2 Status of Patients with Breast Cancer.

OBJECTIVES: The aim of our study was to preoperatively predict the human epidermal growth factor receptor 2 (HER2) status of patients with breast cancer using radiomics signatures based on single-parametric and multiparametric magnetic resonance imaging (MRI). METHODS: Three hundred six patients with invasive ductal carcinoma of no special type (IDC-NST) were retrospectively enrolled. Quantitative imaging features were extracted from fat-suppressed T2-weighted and dynamic contrast-enhanced T1 weighted (DCE-T1) preoperative MRI. Then, three radiomics signatures based on fat-suppressed T2-weighted images, DCE-T1 images and their combination were developed using a support vector machine (SVM) to predict the HER2-positive vs HER2-negative status of patients with breast cancer. The area under the curve (AUC), accuracy, sensitivity, and specificity were calculated to assess the predictive performances of the signatures. RESULTS: Twenty-eight quantitative radiomics features, namely, 14 texture features, 4 first-order features, 9 wavelet features, and 1 shape feature, were used to construct radiomics signatures. The performance of the radiomics signatures for distinguishing HER2-positive from HER2-negative breast cancer based on fat-suppressed T2-weighted images, DCE-T1 images, and their combination had an AUC of 0.74 (95% confidence interval [CI], 0.700 to 0.770), 0.71 (0.673 to 0.738), and 0.86 (0.832 to 0.882) in the primary cohort and 0.70 (0.666 to 0.744), 0.68 (0.650 to 0.726), and 0.81 (0.776 to 0.837) in the validation cohort, respectively. CONCLUSION: Radiomics signatures based on multiparametric MRI represent a potential and efficient alternative tool to evaluate the HER2 status in patients with breast cancer.

Annotation-efficient deep learning for automatic medical image segmentation.

Automatic medical image segmentation plays a critical role in scientific research and medical care. Existing high-performance deep learning methods typically rely on large training datasets with high-quality manual annotations, which are difficult to obtain in many clinical applications. Here, we introduce Annotation-effIcient Deep lEarning (AIDE), an open-source framework to handle imperfect training datasets. Methodological analyses and empirical evaluations are conducted, and we demonstrate that AIDE surpasses conventional fully-supervised models by presenting better performance on open datasets possessing scarce or noisy annotations. We further test AIDE in a real-life case study for breast tumor segmentation. Three datasets containing 11,852 breast images from three medical centers are employed, and AIDE, utilizing 10% training annotations, consistently produces segmentation maps comparable to those generated by fully-supervised counterparts or provided by independent radiologists. The 10-fold enhanced efficiency in utilizing expert labels has the potential to promote a wide range of biomedical applications.

[Value of diffusion weighted imaging (DWI) in evaluating early response to neoadjuvant chemotherapy in locally advanced breast cancer].

OBJECTIVE: To evaluate the role and the performance of diffusion weighted imaging (DWI) for predicting the early response to neoadjuvant chemotherapy (NAC) in local advanced breast cancer (LABC) and to assess the accuracy of DWI in evaluating residual lesion after NAC. METHODS: 88 women with LABC (89 lesions) underwent DWI before and after the first and final cycle of NAC. For each patient, the apparent diffusion coefficient (ADC) values were compared between the baseline and follow-up to predict the early response to NAC. The residual tumor volumes were obtained using 3D maximum intensity projections (MIP) of DWI map, and were compared with pathological findings to assess the accuracy of DWI in detecting and measuring residual tumor. All results were proved or analyzed comparing with the data from histopathology. RESULTS: There were 68 lesions responding to NAC, while 21 non-responders. The baseline ADC values of responders and non-responders were (1.049 +/- 0.135) x 10(-3) mm(2)/s and (1.171 +/- 0.134) x 10(-3)mm(2)/s, respectively, with a significant difference (t = -2.731, P = 0.009 < 0.01). The ADC value measured prior to treatment was (1.087 +/- 0.146) x 10(-3)mm(2)/s, and the degree of the changes in tumor volume after NAC was (70.4% +/- 55.1)%. A negative correlation was observed (r = -0.430, P = 0.025 < 0.05). In the response group, there was a significant difference in ADC value between prior to NAC and 1st cycle of NAC, the final cycle of NAC, respectively (P < 0.001). While no significant differences were found in non-responders during NAC (P > 0.05). The tumor volume correlation coefficient between DWI and pathology measurements was very high (r = 0.749, P < 0.01). CONCLUSION: DWI appears to provide functional information regarding changes in ADC value of tumors due to NAC. DWI may be useful in monitoring the early pathological response of tumor after the initiation of treatment and in evaluating the residual tumor after NAC.

[Value of dynamic contrast-enhanced MRI in assessment of early response to neoadjuvant chemotherapy in breast cancer].

OBJECTIVE: To assess the value of dynamic contrast-enhanced MRI (DMRI) in predicting early response to neoadjuvant chemotherapy (NAC) in patients with locally advanced breast cancer (LABC) and to assess the accuracy of MRI in evaluation of residual disease after NAC. METHODS: Forty-three women with LABC (44 lesions, all were invasive ductal carcinoma) underwent DMRI before, after the first and final cycles of NAC. For each patient, the tumor volume, early enhancement ratio (E1), maximum enhancement ratio (Emax), and maximum enhancement time (Tmax), dynamic signal intensity-time curve were obtained during treatment. The residual tumor volumes obtained by DMRI were compared with pathological findings to assess the accuracy of DMRI. RESULTS: After the first cycle of NAC, the mean volume of responders decreased insignificantly (P = 0.055), but after NAC, mean volume of residual tumor decreased significantly (P = 0.000). Morphological changes: 29 cases showed a concentric shrinkage pattern while 7 cases showed a dendritic shrinkage pattern. Significant differences were found in E1, Emax and Tmax between responders and non-responders (P < 0.05). After the first cycle of NAC, E1, Emax and Tmax of responders changed significantly (P < 0.001), while there was no significant change in non-responders (P > 0.05). After NAC, the dynamic signal intensity-time types were changed in responders, and tended to be significantly flattening, while no significant change was found in non-responders. The residual tumor volume correlation coefficient between MRI and pathology measurements was very high (r = 0.866, P < 0.01). CONCLUSION: DMRI is useful to evaluate the early response to NAC in LABC. The presence and volume of residual tumor in LABC patients treated with NAC can be accurately evaluated by DMRI.

Preoperative Prediction of Axillary Lymph Node Metastasis in Breast Carcinoma Using Radiomics Features Based on the Fat-Suppressed T2 Sequence.

RATIONALE AND OBJECTIVES: To investigate the value of radiomics method based on the fat-suppressed T2 sequence for preoperative predicting axillary lymph node (ALN) metastasis in breast carcinoma. MATERIALS AND METHODS: The data of 329 invasive breast cancer patients were divided into the primary cohort (n = 269) and validation cohort (n = 60). Radiomics features were extracted from the fat-suppressed T2-weighted images on breast MRI, and ALN metastasis-related radiomics feature selection was performed using Mann-Whitney U-test and support vector machines with recursive feature elimination; then a radiomics signature was constructed by linear support vector machine. The predictive models were constructed using a linear regression model based on the clinicopathologic factors and radiomics signature, and nomogram was used for a visual prediction of the combined model. The predictive performances are evaluated with the sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve. RESULTS: A total of 647 radiomics features were extracted from each patient. About 23 ALN metastasis-related radiomics features were selected to construct the radiomics signature, including 17 texture features, 5 first-order statistical features, and one shape feature; patient age, tumor size, HER2 status, and vascular cancer thrombus accompanied or not were selected to construct the cilinicopathologic feature model. The sensitivity, specificity, accuracy, and are under the curve value of radiomics signature, clinicopathologic feature model, and the nomogram were 65.22%, 81.08%, 75.00%, and 0.819 (95% confidence interval [CI]: 0.776-0.861), 30.44%, 81.08%, 61.67%, and 0.605 (95% CI: 0.571-0.624) and 60.87%, 89.19%, 78.33%, and 0.810 (95% CI: 0.761-0.855), respectively. CONCLUSION: Radiomics methods based on the fat-suppressed T2 sequence and the nomogram are helpful for preoperative accurate predicting ALN metastasis.

Mammography-based radiomics nomogram: a potential biomarker to predict axillary lymph node metastasis in breast cancer.

OBJECTIVE: To establish a radiomics nomogram by integrating clinical risk factors and radiomics features extracted from digital mammography (MG) images for pre-operative prediction of axillary lymph node (ALN) metastasis in breast cancer. METHODS: 216 patients with breast cancer lesions confirmed by surgical excision pathology were divided into the primary cohort (n = 144) and validation cohort (n = 72). Radiomics features were extracted from craniocaudal (CC) view of mammograms, and radiomics features selection were performed using the methods of ANOVA F-value and least absolute shrinkage and selection operator; then a radiomics signature was constructed with the method of support vector machine. Multivariate logistic regression analysis was used to establish a radiomics nomogram based on the combination of radiomics signature and clinical factors. The C-index and calibration curves were derived based on the regression analysis both in the primary and validation cohorts. RESULTS: 95 of 216 patients were confirmed with ALN metastasis by pathology, and 52 cases were diagnosed as ALN metastasis based on MG-reported criteria. The sensitivity, specificity, accuracy and AUC (area under the receiver operating characteristic curve of MG-reported criteria were 42.7%, 90.8%, 24.1% and 0.666 (95% confidence interval: 0.591-0.741]. The radiomics nomogram, comprising progesterone receptor status, molecular subtype and radiomics signature, showed good calibration and better favorite performance for the metastatic ALN detection (AUC 0.883 and 0.863 in the primary and validation cohorts) than each independent clinical features (AUC 0.707 and 0.657 in the primary and validation cohorts) and radiomics signature (AUC 0.876 and 0.862 in the primary and validation cohorts). CONCLUSION: The MG-based radiomics nomogram could be used as a non-invasive and reliable tool in predicting ALN metastasis and may facilitate to assist clinicians for pre-operative decision-making. ADVANCES IN KNOWLEDGE: ALN status remains among the most important breast cancer prognostic factors and is essential for making treatment decisions. However, the value of detecting metastatic ALN by MG is very limited. The studies on pre-operative ALN metastasis prediction using the method of MG-based radiomics in breast cancer are very few. Therefore, we studied whether MG-based radiomics nomogram could be used as a predictive biomarker for the detection of metastatic ALN.

Radiological and clinical findings in sclerosing adenosis of the breast.

To study the imaging and clinical features of breast sclerosing adenosis (SA), and to enhance the recognition of this disease, as well as to help the clinic to give a correct diagnosis.Imaging findings were retrospectively reviewed in 47 women with SA lesions confirmed by pathology (including 39 cases of mammography, 40 cases of ultrasound [US], and 34 cases magnetic resonance imaging [MRI]).Of 47 patients confirmed with SA, 18 cases were pure SA, and 29 cases coexist with other proliferative lesions and malignancies; the maximum diameter of SA lesions was 0.5 to 3.5 cm with an average of 1.6 cm. On the mammogram of 39 SA cases, the percentage of architectural distortion, calcifications, mass/nodular, asymmetric density, and mass combining with calcifications were 30.8%, 23.1%, 17.9%, 12.8%, and 7.7%, respectively; and 3 cases had no abnormal findings. On the sonogram (excluding 5 normal finding cases), the majority of lesions showed regular shaped (57.1%), well defined margined (60.0%), heterogenous low echoed (71.4%) nodulus. 85.3% lesions showed high signal on T2-weighted images, and all lesions were enhanced markedly, including 82.4% lesions appearing mass-like enhancement (17 star-shaped enhanced masses included); and the percentage of the time-signal intensity curve in type 1, type 2, and type 3 were 52.9%, 41.2%, and 5.9%, respectively. If the category breast imaging-reporting and data system ≥4b was considered to be a suspicious malignant lesion, the misdiagnostic rates of mammography, US, and MRI would be 17.9%, 17.5%, and 35.3%, respectively.The SA lesions are small and can occur with other diseases histologically. The majority of SA lesions showed distortion or calcifications on mammograms, low echo-level nodules with heterogenous echo on US and mass-like lesion with or without star shape on enhanced MRI.

Evaluating the HER-2 status of breast cancer using mammography radiomics features.

PURPOSE: The aim of our study was to evaluate the HER-2 status in breast cancer patients using mammography (MG) radiomics features. METHODS: A total of 306 Chinese female patients with invasive ductal carcinoma of no special type (IDC-NST) enrolled from January 2013 to July 2018 were divided into a training set (n = 244) and a testing set (n = 62). One hundred and eighty-six radiomics features were extracted from digital MG images based on the training set. The least absolute shrinkage and selection operator (LASSO) method was used to select the optimal predictive features for HER-2 status from the training set. Both support vector machine (SVM) and logistic regression models were employed based on the selected features. The area under the receiver operating characteristic (ROC) curves (AUCs) of the training set and testing set were used to evaluate the predictive performance of the models. RESULTS: Compared with the SVM model, the performance of the logistic regression model using a combination of cranial caudal (CC) and mediolateral oblique (MLO) MG views was optimal. In the training set, the sensitivity, specificity, accuracy and area under the curve (AUC) values of the logistic regression model for evaluating HER-2 status based on quantitative radiomics features were 87.29%, 58.73%, 80.00% and 0.846 (95% confidence interval (CI), 0.800-0.887), respectively, and in the testing set, the values were 73.91%, 68.75%, 77.00% and 0.787 (95% CI, 0.673-0.885), respectively. CONCLUSIONS: Radiomics features could be an efficient tool for the preoperative evaluation of HER-2 status in patients with breast cancer.

Fully Convolutional DenseNet with Multiscale Context for Automated Breast Tumor Segmentation.

Breast tumor segmentation plays a crucial role in subsequent disease diagnosis, and most algorithms need interactive prior to firstly locate tumors and perform segmentation based on tumor-centric candidates. In this paper, we propose a fully convolutional network to achieve automatic segmentation of breast tumor in an end-to-end manner. Considering the diversity of shape and size for malignant tumors in the digital mammograms, we introduce multiscale image information into the fully convolutional dense network architecture to improve the segmentation precision. Multiple sampling rates of atrous convolution are concatenated to acquire different field-of-views of image features without adding additional number of parameters to avoid over fitting. Weighted loss function is also employed during training according to the proportion of the tumor pixels in the entire image, in order to weaken unbalanced classes problem. Qualitative and quantitative comparisons demonstrate that the proposed algorithm can achieve automatic tumor segmentation and has high segmentation precision for various size and shapes of tumor images without preprocessing and postprocessing.