Objective To observe the value of digital breast tomosynthesis(DBT)-based radiomics nomogram for predicting Ki-67 expression levels in breast invasive ductal carcinoma(BIDC).Methods Data of 374 cases of BIDC were retrospectively analyzed and divided into high-expression group(n=224)and low-expression group(n=150)according to expression of Ki-67 as well as training set(n=271,162 cases in high-expression group and 109 cases in low-expression group)and test set(n=103,62 cases in high-expression group and 41 cases in low-expression group)at the ratio of 7∶3.Clinical characteristics and lesion's image manifestations were compared between groups,and radiomic features were extracted and filtered.Then imaging radiomics models were constructed with 8 classifiers,respectively,and the optimal classifier was selected.A nomogram model was subsequently developed through integrating image features and radiomics scores.Receiver operating characteristic curves were drawn,and the area under the curves(AUC)were calculated to evaluate the performance of the above models.Results Significant differences of the maximum diameter and spiculated margin of the lesions were found between groups(both P＜0.05),and AUC of multi-layer perceptron(MLP)image model constructed based on these indexes for predicting BIDC expression level of Ki-67 was 0.654 in training set and 0.715 in test set,of MLP radiomics model constructed based on 8 radiomics features was 0.802 in training set and 0.806 in test set,while of the nomogram model constructed based on image features and radiomics scores was 0.802 in training set and 0.806 in test set,respectively.Conclusion DBT-based radiomics nomogram could be used to effectively predict Ki-67 expression levels in BIDC.

Objective To observe the value of digital breast tomosynthesis(DBT)radiomics combined with clinical data for predicting molecular type of breast invasive ductal carcinoma(IDC).Methods Totally 309 female patients with Luminal type and 97 with non-Luminal type breast IDC were retrospectively enrolled and divided into training set(n=284)and test set(n=122)at a ratio of 7∶3.Multivariate logistic regression analysis was performed to screen independent impact factors related to Luminal expression of breast IDC based on clinical data and DBT manifestations,then a clinical-imaging model was constructed.The radiomics model was established based on mediolateral oblique position of DBT images,and a combined model was conducted combining with clinical-imaging model and radiomics model.Receiver operating characteristic curves were drawn,and the area under the curves(AUC)were calculated to evaluate the efficacy of each model for predicting molecular type of breast IDC.Results Histopathological grade and lesion margin showed on DBT was the clinical and imaging independent impact factor associated with Luminal expression of breast IDC,respectively(both P＜0.05).AUC of combined model for predicting molecular type of breast IDC in training set and test set was 0.821 and 0.715,respectively,both higher than clinical-imaging model(0.727 and 0.665)and radiomics model(0.776 and 0.663)(all P＜0.05),while no significant difference was found between clinical-imaging model and radiomics model(both P＞0.05).Conclusion DBT radiomics combined with clinical data was helpful for predicting molecular type of breast IDC.

Objective To observe the value of digital breast tomosynthesis(DBT)-based radiomics nomogram for predicting Ki-67 expression levels in breast invasive ductal carcinoma(BIDC).Methods Data of 374 cases of BIDC were retrospectively analyzed and divided into high-expression group(n=224)and low-expression group(n=150)according to expression of Ki-67 as well as training set(n=271,162 cases in high-expression group and 109 cases in low-expression group)and test set(n=103,62 cases in high-expression group and 41 cases in low-expression group)at the ratio of 7∶3.Clinical characteristics and lesion's image manifestations were compared between groups,and radiomic features were extracted and filtered.Then imaging radiomics models were constructed with 8 classifiers,respectively,and the optimal classifier was selected.A nomogram model was subsequently developed through integrating image features and radiomics scores.Receiver operating characteristic curves were drawn,and the area under the curves(AUC)were calculated to evaluate the performance of the above models.Results Significant differences of the maximum diameter and spiculated margin of the lesions were found between groups(both P＜0.05),and AUC of multi-layer perceptron(MLP)image model constructed based on these indexes for predicting BIDC expression level of Ki-67 was 0.654 in training set and 0.715 in test set,of MLP radiomics model constructed based on 8 radiomics features was 0.802 in training set and 0.806 in test set,while of the nomogram model constructed based on image features and radiomics scores was 0.802 in training set and 0.806 in test set,respectively.Conclusion DBT-based radiomics nomogram could be used to effectively predict Ki-67 expression levels in BIDC.

Objective To observe the value of digital breast tomosynthesis(DBT)radiomics combined with clinical data for predicting molecular type of breast invasive ductal carcinoma(IDC).Methods Totally 309 female patients with Luminal type and 97 with non-Luminal type breast IDC were retrospectively enrolled and divided into training set(n=284)and test set(n=122)at a ratio of 7∶3.Multivariate logistic regression analysis was performed to screen independent impact factors related to Luminal expression of breast IDC based on clinical data and DBT manifestations,then a clinical-imaging model was constructed.The radiomics model was established based on mediolateral oblique position of DBT images,and a combined model was conducted combining with clinical-imaging model and radiomics model.Receiver operating characteristic curves were drawn,and the area under the curves(AUC)were calculated to evaluate the efficacy of each model for predicting molecular type of breast IDC.Results Histopathological grade and lesion margin showed on DBT was the clinical and imaging independent impact factor associated with Luminal expression of breast IDC,respectively(both P＜0.05).AUC of combined model for predicting molecular type of breast IDC in training set and test set was 0.821 and 0.715,respectively,both higher than clinical-imaging model(0.727 and 0.665)and radiomics model(0.776 and 0.663)(all P＜0.05),while no significant difference was found between clinical-imaging model and radiomics model(both P＞0.05).Conclusion DBT radiomics combined with clinical data was helpful for predicting molecular type of breast IDC.

Primary small bowel tumors have low incidence and contain predominantly solid components, and the lesions are similar and difficult to be detected and distinguished with multislice spiral CT (MSCT) plain scans. In this article we describe contrast-enhanced MSCT technique and imaging characteristics for solid small bowel tumors or small bowel tumors containing predominantly solid components, including the type and use of contrast agents. In contrast-enhanced MSCT, small bowel imaging with CT has the advantages of determining the true extent of intestinal wall lesions, the possible extent of wall penetration, the degree of mesenteric involvement, and distant metastases, as well as easiness to detect and identify the blood supply vessels of small bowel tumors and assessment of the corresponding complications. Contrast-enhanced MSCT has become the best noninvasive imaging technique for the diagnosis, evaluation, and staging of solid small bowel tumors or small bowel tumors containing predominantly solid components. CT texture analysis (CTTA) is a new research hotspot and can be useful for the correct diagnosis of primary small bowel tumors containing predominantly solid components.

As a non-invasive image analysis method, radiomics can deeply explore the clinical information hidden behind medical images, and has been widely used in medicine in recent years. Consolidation immunotherapy after concurrent chemoradiotherapy has become the standard treatment for locally advanced non-small cell lung cancer. The prediction and identification of treatment-associated adverse events radiation pneumonitis (RP) and immune checkpoint inhibitor-related pneumonitis (CIP) are of vital importance for the formulation of treatment plan and the selection of subsequent treatment. CT-based radiomics analysis shows great potential in predicting and identifying RP and CIP.

Objective To investigate the value of 18 F-fluorodeoxyglucose FDG) positron emission tomography (PET)-computed tomography (CT) in predicting the progression-free survival (PFS)and overall survival (OS) of patients with esophageal squamous cell carcinoma (ESCC) after threedimensional (3D) radiotherapy.Methods A retrospective analysis was performed on 98 ESCC patients,who underwent FDG PET-CT before 3D radiotherapy from 2004 to 2010,to investigate their 1-,3-,and 5-year PFS and OS rates.The relationship of maximum standard uptake value (SUVmax),mean SUV (SUVmean),metabolic target volume (MTV),length of primary tumor on PET-CT before radiotherapy,and number of tumors on PET with PFS and OS were analyzed.The SUVs and clinical data were analysed by independent samples t-test or Hotelling T2 test; the Kaplan-Meier method was used for calculating PFS and OS rates,and the Logrank test was used for survival difference analysis;the prognostic factors were analysed using the Cox proportional hazard model.Results The follow-up rate was 100％ ;56 patients were followed up for at least 3 years,and 27 for at 5 years.The SUVmax SUVmean and MTV of primary tumor,length of primary tumor on PET-CT before radiotherapy,and number of tumors on PET were correlated with PFS and OS (x2 =8.99-41.82,all P ＜ 0.01).The Cox regression analysis showed that PFS could be well predicted based on SUVmean (x2 =4.41,P =0.036,RR =1.398) and number of tumors on PET (x2 =6.79,P =0.009,RR =3.650) and that OS could be well predicted based on number of tumors on PET (x2 =5.03,P =0.025,RR =3.740).Conclusions When estimating the long-term response to precise radiotherapy in patients with ESCC,SUV mean and number of tumors on PET may be used to predict PFS,and number of tumors on PET may be used to predict OS.

Objective To characterize the CT features of primary mediastinal seninoma. Methods CT scans of 10 cases with pathologically proved primary mediastinal seminoma were retrospectively reviewed.Results 9 of the 10 cases with the maximal dimension of 6. 5 ～ 18.2 cm( mean = 12. 3 cm) were located in the anterior mediastinum and 1 tumor was located in the middle mediastinum. All of the 10 tumors were solid masses and none of cavity, loculus, fat, calcification or fluid was detected. Areas of low density in the center of tumors were seen in 9 cases. Tumors encircling macrovascular or pericardium were seen in 7 cases. Chest wall invasion was detected in 2 cases. Enlarged lymph nodes in supraclavicular or mediastinum were identified in 4 cases. Lung diseases were found in 5 cases. Bilateral pulmonary metastasis was seen in 1 case. Pleural effusions were detected in 4 cases. Pericardial effusions were evident in 5 cases. Conclusion Most of the primary mediastinal seminomas are solid and large mass without calcification or fat. The tumors are usually located in the anterior mediastinum and areas of low density are usually seen in the center of tumors. Some tumors involve in adjacent structures. Seminoma has some features on CT.

Objective To design a kind of bed board that is suitable for various medical image diagnostic equipments. It is used to make the body position of the person who is examined to be fixed and can be moved with whom. Methods The bed board is made of low density plastics; its structure is similar to "winnowing fan" in shape. In which, the pipes filled with "multifunction developer" are arranged in order. The negative pressure vacuum bag and belts for the body positioning are installed on the bed board. Volumetric reconstruction has been done with single mode images, then multi- mode images volumetric fusion depend on registration with coronal and arrowy mark of image pipeline in volumetric reconstruction. Results The bed board can not be showed in CT, MRI, PET and SPECT with the soft tissue window, but the pipes be used as the marker for image fusion is displayed clearly. The body position is all the same with each medical image examination. National patent of the product was declared. Conclusion With the designed bed board, the multimode image fusion is rapid, simple and accurate, is easily expansion and popularize.