Optimizing perioperative treatment for potentially resectable stage III squamous cell lung carcinoma: promising results of a condensed four-cycle regimen with tislelizumaband chemotherapy.

BACKGROUND: The standard care for resectable non-small cell lung cancer (NSCLC) involves perioperative therapy combining chemotherapy and immune checkpoint inhibitors, typically lasting 6 to 12 months. However, the optimal treatment strategies for potentially resectable squamous cell lung carcinoma (SCC) remain unclear. This Phase 2 trial aimed to assess the efficacy and safety of a condensed four-cycle perioperative treatment regimen with tislelizumab combined with chemotherapy in patients with potentially resectable stage III SCC. METHODS: Patients with potentially resectable stage IIIA-IIIB (N2) SCC received intravenous tislelizumab, albumin-bound paclitaxel, and carboplatin for up to four cycles. The primary endpoints were major pathologic response (MPR) and incidence of treatment-related adverse events. Safety and potential biomarkers for efficacy prediction were also assessed. RESULTS: Among 35 enrolled patients, 32 underwent surgery with R0 resection achieved in all cases. MPR was achieved in 24 patients and pathological complete response (pCR) in 14 patients. Radiographic objective response was observed in 31 patients. The 12-month and 24-month event-free survival rate was 85.7 and 61.0%, respectively. Four patients experienced grade 3 or 4 adverse events. Tumor tissue based next-generation sequencing revealed the potential associations between several biomarkers and pathological response, including tumor neoantigen burden score, 18-gene expression profile score, CD8 + T cells, M1/M2 macrophages ratio and interferon-gamma expression level. Besides, circulating tumor DNA (ctDNA) dynamics and concentration were also associated with pathological response and the presence of ctDNA at postoperative month 1 was a strong predictor for disease relapse. Furthermore, metagenomic sequencing in bronchoalveolar lavage fluid demonstrated Streptococcus was the most abundant genus in the pCR group. CONCLUSIONS: A condensed four-cycle perioperative treatment regimen of tislelizumab combined with chemotherapy demonstrated promising efficacy and manageable toxicities in potentially resectable stage III SCC. Specific biomarkers showed potential for predicting treatment efficacy and the mechanism of superior antitumor response of pCR patients was preliminarily and indirectly explored. TRIAL REGISTRATION: ClinicalTrials.gov, NCT05024266. Registered August 27, 2021.

[Dynamic changes of chest CT imaging in patients with COVID-19].

OBJECTIVE: To analyze the dynamic changes of chest CT images of patients with coronavirus disease 2019 (COVID-19). METHODS: Fifty-two cases of COVID-19 were admitted in the First Affiliated Hospital of Zhejiang University School of Medicine. The consecutive chest CT scans were followed up for all patients with an average of 4 scans performed per patient during the hospitalization. The shortest interval between each scan was 2 days and the longest was 7 days. The shape, number and distribution of lung shadows, as well as the characteristics of the lesions on the CT images were reviewed. RESULTS: The obvious shadows infiltrating the lungs were shown on CT images in 50 cases, for other 2 cases there was no abnormal changes in the lungs during the first CT examination. Ground-glass opacities (GGO) were found in 48 cases (92.3%), and 19 cases (36.5%) had patchy consolidation and sub-consolidation, which were accompanied with air bronchi sign in 17 cases (32.7%). Forty one cases (78.8%) showed a thickened leaflet interval, 4 cases (7.6%) had a small number of fibrous stripes. During hospitalization, GGO lesions in COVID-19 patients gradually became rare,the fibrous strip shadows increased and it became the most common imaging manifestation. The lesions rapidly progressed in 39 cases (75.0%) within 6-9 days after admission. On days 10-14 of admission, the lesions distinctly resolved in 40 cases (76.9%). CONCLUSIONS: The chest CT images of patients with COVID-19 have certain characteristics with dynamic changes, which are of value for monitoring disease progress and clinical treatment.

Clinical findings in 111 cases of influenza A (H7N9) virus infection.

BACKGROUND: During the spring of 2013, a novel avian-origin influenza A (H7N9) virus emerged and spread among humans in China. Data were lacking on the clinical characteristics of the infections caused by this virus. METHODS: Using medical charts, we collected data on 111 patients with laboratory-confirmed avian-origin influenza A (H7N9) infection through May 10, 2013. RESULTS: Of the 111 patients we studied, 76.6% were admitted to an intensive care unit (ICU), and 27.0% died. The median age was 61 years, and 42.3% were 65 years of age or older; 31.5% were female. A total of 61.3% of the patients had at least one underlying medical condition. Fever and cough were the most common presenting symptoms. On admission, 108 patients (97.3%) had findings consistent with pneumonia. Bilateral ground-glass opacities and consolidation were the typical radiologic findings. Lymphocytopenia was observed in 88.3% of patients, and thrombocytopenia in 73.0%. Treatment with antiviral drugs was initiated in 108 patients (97.3%) at a median of 7 days after the onset of illness. The median times from the onset of illness and from the initiation of antiviral therapy to a negative viral test result on real-time reverse-transcriptase-polymerase-chain-reaction assay were 11 days (interquartile range, 9 to 16) and 6 days (interquartile range, 4 to 7), respectively. Multivariate analysis revealed that the presence of a coexisting medical condition was the only independent risk factor for the acute respiratory distress syndrome (ARDS) (odds ratio, 3.42; 95% confidence interval, 1.21 to 9.70; P=0.02). CONCLUSIONS: During the evaluation period, the novel H7N9 virus caused severe illness, including pneumonia and ARDS, with high rates of ICU admission and death. (Funded by the National Natural Science Foundation of China and others.).

[CT and MRI findings in patients with autoimmune pancreatitis].

OBJECTIVE: To evaluate computed tomography (CT) and magnetic resonance imaging (MRI) findings in patients with autoimmune pancreatitis (AIP). METHODS: The imaging findings of pancreas and extra-pancreas in 24 patients with AIP were retrospectively reviewed. Among them, CT scan was performed in 18 patients, MRI in 11, and bGth CT and MRI in 10. RESULTS: The pancreas showed diffuse enlargement (25%, 6/24), focal enlargement (37. 5%, 9/24), combined enlargement (25%, 6/24) ,and no enlargement (12. 5%, 9/24). Unenhanced CT showed hypoattenuation in AIP area (n = 2) . After intravenous injection of contrast medium, 17 patients showed abnormal contrast enhancement in the affected pancreatic parenchyma, including hypoattenuation during the arterial phase (50%, 9/18) and hyper attenuation during the delayed phase (94. 4%, 17/18). Precontrast MRI showed abnormal signal intense (n =9), including hypointense on T1-weight images (T1 WI) (n = 7), hyperintense (n = 7) and hypointense (n = 2) on T2-weight images (TIWI). Enhanced MRI demonstrated abnormal contrast enhancement within lesions (n = 11), including hypoattenuation during the arterial phase (81. 8%, 9/11) and good enhancement during the delayed phase (100%, 11111). A capsule-like rim was seen around pancreas (37. 5%, 9/24), among which CT detected in 6 out of 18 patients and MRI found in 7 out of 11 patients.The main pancreatic duct lumen within lesions has no visualization (100%, 24/24) and upstream dilation of the main pancreatic duct (n = 8) , ranging from 2. 2 to 4. 5 mm(mean 3. 1 0. 47 mm) in diameter. Narrowing of the common bile duct was shown in 14 patients. Miscellaneous findings were: infiltration of extrapancreatic vein (n = 9) and artery (n = 1); mild fluid collection around pancreas (n = 2); pseudocysts (n = 3). Fourteen patients also presented one or more of the following extrapancreatic imaging findings: narrowing of the intra-hepatic bile duct or hilar duct (n = 5); thickening of gallbladder wall (n = 5); fibrosis in mesenteric (n = 2), in retroperitoneal (n = 2) and in ligamentum teres hepatis (n = 1); renal involvement (n = 3); peri-pancreatic or para-aortic lymphadenopathy (n = 10); and ulcerative colitis (n = 3). CONCLUSION: AIP display some characteristic CT and MRI imaging features: sausage-like change of the pancreas; capsule-like rims around lesions; delayed contrast enhancement in the affected pancreatic parenchyma; segment or diffuse pancreatic duct stenosis but mild upstream dilation and extrapancreatic organs involvement. CT and MRI findings combining with serological tests and pancreas biopsy can assist physicians to make accurate and timely diagnosis.

The utility of repeated computed tomography to track a foreign body penetrating the esophagus to the level of the thyroid gland.

OBJECTIVE: Foreign body (FB) ingestion is a common problem in otolaryngology. One uncommon complication of FB ingestion is penetration to the level of the thyroid gland. To our knowledge, only 21 such cases have been reported in the literature. Here, we report a case of an esophageal FB penetrating to the level of the right thyroid gland. CASE REPORT: The patient was a 38-year-old woman in whom an esophageal FB penetrated to the level of the right thyroid gland. We traced the path to the thyroid gland using repeated computed tomography (CT) scans and demonstrated the importance of multiplanar reconstruction in locating the FB and formulating a precise surgical plan. CONCLUSIONS: To our knowledge, this is the first report of repeat CT scans being used to demonstrate the migratory route, over time, of a FB penetrating through the esophagus to the level of the thyroid gland. Our results suggest that multiplanar reconstruction may play a key role in the precise diagnosis of a FB at the level of the thyroid gland and may help surgeons choose the best approach for removal.

Poor outcome of comprehensive therapy in a case of laryngeal synovial sarcoma.

BACKGROUND: Synovial sarcoma is common in the extremities. Our search revealed only 17 cases of synovial sarcoma of the larynx in the English-language literature. CASE REPORT: We report an additional case of a 37-year-old man with primary laryngeal synovial sarcoma who underwent positron emission tomography/computed tomography (PET/CT) following the treatment. Although the patient received comprehensive therapy including surgery, radiotherapy, repeated chemotherapies, and targeted therapies, he had an unfavourable outcome and died of distant metastases. CONCLUSIONS: In synovial sarcoma of the larynx, PET/CT can detect recurrence and metastasis. PET/CT can also predict the treatment effect in patients with synovial sarcoma.

Efficacy and safety of ripretinib in Chinese patients with advanced gastrointestinal stromal tumors: a real-world, multicenter, observational study.

Introduction: Mutations in KIT proto-oncogene, receptor tyrosine kinase (KIT) and platelet-derived growth factor receptor-α (PDGFRA) render the available tyrosine kinase inhibitors (TKI) ineffective in treating advanced gastrointestinal stromal tumors (GIST). Ripretinib, a broad-spectrum switch-control kinase inhibitor, has shown increased efficacy and manageable safety, but real-world evidence remains scarce. This study evaluates the efficacy and safety of ripretinib among Chinese patients in a real-world setting. Methods: Advanced GIST patients (N=23) receiving ripretinib following progression on previous lines of TKI treatment were enrolled to determine the efficacy [progression-free survival (PFS) and overall survival (OS)]. Safety was assessed by the incidence and severity of adverse events (AEs). All statistical analyses were performed using SPSS version 20.0 and a p-value of <0.05 was considered significant. Results: The median PFS (mPFS) of efficacy analysis set (EAS) (N=21) was 7.1 months. mPFS of patients receiving ripretinib following ≤2 lines of previous TKI treatment and ≥3 prior lines of therapy were 7.1 and 9.2 months, respectively. The median OS (mOS) was 12.0 months and shorter interval between the end of the latest TKI and ripretinib therapy was correlated with longer median PFS and OS (p=0.054 and p=0.046), respectively. Alopecia and asthenia were the most common AEs observed. Conclusion: Compared to previous lines of TKI in advanced GIST patients, ripretinib showed superior efficacy with clinically manageable AEs. Real-world results are comparable to that of phase III INVICTUS study and its Chinese bridging study. Hence, ripretinib can be used for the clinical management of advanced GIST patients.

Development and validation of a meta-learning-based multi-modal deep learning algorithm for detection of peritoneal metastasis.

PURPOSE: The existing medical imaging tools have a detection accuracy of 97% for peritoneal metastasis(PM) bigger than 0.5 cm, but only 29% for that smaller than 0.5 cm, the early detection of PM is still a difficult problem. This study is aiming at constructing a deep convolution neural network classifier based on meta-learning to predict PM. METHOD: Peritoneal metastases are delineated on enhanced CT. The model is trained based on meta-learning, and features are extracted using multi-modal deep Convolutional Neural Network(CNN) with enhanced CT to classify PM. Besides, we evaluate the performance on the test dataset, and compare it with other PM prediction algorithm. RESULTS: The training datasets are consisted of 9574 images from 43 patients with PM and 67 patients without PM. The testing datasets are consisted of 1834 images from 21 testing patients. To increase the accuracy of the prediction, we combine the multi-modal inputs of plain scan phase, portal venous phase and arterial phase to build a meta-learning-based multi-modal PM predictor. The classifier shows an accuracy of 87.5% with Area Under Curve(AUC) of 0.877, sensitivity of 73.4%, specificity of 95.2% on the testing datasets. The performance is superior to routine PM classify based on logistic regression (AUC: 0.795), a deep learning method named ResNet3D (AUC: 0.827), and a domain generalization (DG) method named MADDG (AUC: 0.834). CONCLUSIONS: we proposed a novel training strategy based on meta-learning to improve the model's robustness to "unseen" samples. The experiments shows that our meta-learning-based multi-modal PM predicting classifier obtain more competitive results in synchronous PM prediction compared to existing algorithms and the model's improvements of generalization ability even with limited data.

A Clinical-Radiomics Nomogram for Preoperative Prediction of Lymph Node Metastasis in Gallbladder Cancer.

OBJECTIVES: The aim of the current study was to develop and validate a nomogram based on CT radiomics features and clinical variables for predicting lymph node metastasis (LNM) in gallbladder cancer (GBC). METHODS: A total of 353 GBC patients from two hospitals were enrolled in this study. A Radscore was developed using least absolute shrinkage and selection operator (LASSO) logistic model based on the radiomics features extracted from the portal venous-phase computed tomography (CT). Four prediction models were constructed based on the training cohort and were validated using internal and external validation cohorts. The most effective model was then selected to build a nomogram. RESULTS: The clinical-radiomics nomogram, which comprised Radscore and three clinical variables, showed the best diagnostic efficiency in the training cohort (AUC = 0.851), internal validation cohort (AUC = 0.819), and external validation cohort (AUC = 0.824). Calibration curves showed good discrimination ability of the nomogram using the validation cohorts. Decision curve analysis (DCA) showed that the nomogram had a high clinical utility. CONCLUSION: The findings showed that the clinical-radiomics nomogram based on radiomics features and clinical parameters is a promising tool for preoperative prediction of LN status in patients with GBC.

A radiomics model that predicts lymph node status in pancreatic cancer to guide clinical decision making: A retrospective study.

Purpose: To construct a radiomics-based model for predicting lymph node (LN) metastasis status in pancreatic ductal adenocarcinoma (PDAC) before therapy and to evaluate its prognostic clinical value. Materials and Methods: We retrospectively collected preoperative CT scans of 130 PDAC patients who underwent original tumor resection and LN dissection in the entire cohort between January 2014 and December 2017. Radiomics features were systematically extracted and analyzed from CT scans of 89 patients in the primary cohort. To construct a radiomics signature, the least absolute shrinkage and selection operator methods were employed with LN metastasis status as classification labels. Pathological analysis of LN status which were assessed by experienced pathologists was used as the evaluation label. We subjected the clinical nomogram to multivariable logistic regression analysis and conducted performance evaluation based on its discrimination, calibration, and clinical value. The model was tested and validated in 41 patients with PDAC in a separate validation cohort. Results: Four radiomics features closely associated with LN metastasis were selected in the primary and validation cohorts (P < 0.01). Following the integration of CT-reported results and radiomics signatures into the radiomics nomogram, we reported better performance in the primary (area under the curve, 0.80) and validation (area under the curve, 0.78) cohorts. Conclusion: The noninvasive tool constructed from the portal venous phase CT based on radiomics showed better performance for LN metastasis prediction than traditional approaches in pancreatic cancer. It may assist surgeons in crafting detailed procedures before treatment, this subsequently improves tumor staging and resection of patients.

[Value of 18F-FDG-PET/CT in diagnosis of classic fever of unknown origin].

OBJECTIVE: To evaluate the application of (18)F-FDG PET/CT in diagnosis of classic fever of unknown origin. METHODS: A total of 27 consecutive patients with classic fever of unknown origin (FUO) (19 men, 8 women; aged 24-82 y) underwent (18)F-FDG PET/CT scans. The images were interpreted by visual inspection and semiquantitative analysis(standardized uptake value, SUV). Final diagnosis was based on histopathology or clinical follow-up. RESULTS: The cause of FUO was confirmed by followed investigations in 21 of 27 cases after PET/CT scan, including 10 cases of infection, 4 of noninfectious inflammation, 4 of malignancies and 3 of miscellaneous disorders; and remaining 6 cases were still confirmed FUO. The sensitivity, specificity, positive predictive value, negative predictive value and accuracy were 100.0 %, 83.3%, 83.3%, 100.0% and 96.3%, respectively. CONCLUSION: For patients with fever of unknown origin, (18)F-FDG-PET/CT can be a sensitive, reliable imaging modality. It is suggested that (18)F-FDG-PET/CT should be considered earlier in detecting the causes of FUO, which is difficultly diagnosed by conventional modalities.

A Deep Learning System to Screen Novel Coronavirus Disease 2019 Pneumonia.

The real-time reverse transcription-polymerase chain reaction (RT-PCR) detection of viral RNA from sputum or nasopharyngeal swab had a relatively low positive rate in the early stage of coronavirus disease 2019 (COVID-19). Meanwhile, the manifestations of COVID-19 as seen through computed tomography (CT) imaging show individual characteristics that differ from those of other types of viral pneumonia such as influenza-A viral pneumonia (IAVP). This study aimed to establish an early screening model to distinguish COVID-19 from IAVP and healthy cases through pulmonary CT images using deep learning techniques. A total of 618 CT samples were collected: 219 samples from 110 patients with COVID-19 (mean age 50 years; 63 (57.3%) male patients); 224 samples from 224 patients with IAVP (mean age 61 years; 156 (69.6%) male patients); and 175 samples from 175 healthy cases (mean age 39 years; 97 (55.4%) male patients). All CT samples were contributed from three COVID-19-designated hospitals in Zhejiang Province, China. First, the candidate infection regions were segmented out from the pulmonary CT image set using a 3D deep learning model. These separated images were then categorized into the COVID-19, IAVP, and irrelevant to infection (ITI) groups, together with the corresponding confidence scores, using a location-attention classification model. Finally, the infection type and overall confidence score for each CT case were calculated using the Noisy-OR Bayesian function. The experimental result of the benchmark dataset showed that the overall accuracy rate was 86.7% in terms of all the CT cases taken together. The deep learning models established in this study were effective for the early screening of COVID-19 patients and were demonstrated to be a promising supplementary diagnostic method for frontline clinical doctors.

Excision of tumors in the parapharyngeal space using an endoscopically assisted transoral approach: a case series and literature review.

OBJECTIVE: Magnetic resonance imaging (MRI) provides important information regarding tumors in the parapharyngeal space (PPS), revealing their origin, whether they are benign or malignant, and their relationships with surrounding structures. METHODS: Twelve tumors in the PPS were completely excised using an endoscopically assisted transoral approach (EATA). The MRI features were analyzed. RESULTS: Ten pleomorphic adenomas confirmed on postoperative pathological examination had the parotid pedicle sign. A fat space between the tumor and parotid gland may distinguish such a tumor from a tumor arising from a minor salivary gland in the prestyloid space and a tumor arising from the deep lobe of the parotid gland. Both the jugular vein and carotid artery were displaced posteriorly in all 10 cases of pleomorphic adenomas. The principal features of the two schwannomas confirmed on postoperative pathological examination were separation of the internal carotid artery and internal jugular vein and anteromedial displacement of the internal carotid artery, suggesting that the tumors originated in the poststyloid space. In this review, 95 tumors were excised by the EATA in the English-language literature. CONCLUSIONS: MRI renders differential diagnosis possible. PPS tumors may be completely excised via an EATA guided by tumor features evident on preoperative MRI.

Collision carcinoma of squamous cell carcinoma and small cell neuroendocrine carcinoma of the larynx: A case report and review of the literature.

BACKGROUND: Collision carcinoma is rare in clinical practice, especially in the head and neck region. In this paper, we report a case of squamous cell carcinoma (SCC) and neuroendocrine carcinoma (NEC) colliding in the larynx and review 12 cases of collision carcinoma in the head and neck to further understand collision carcinoma, including its definition, diagnosis, and treatment. CASE SUMMARY: A 61-year-old man presented with a 1-year history of hoarseness. Contrast-enhanced magnetic resonance imaging of the larynx revealed that the right vocal cord had a nodule-like thickening with obvious enhancement. Laryngoscopy revealed a neoplasm on the right vocal cord, and a malignant tumor was initially considered. A frozen section of right vocal cord was performed under general anesthesia. The pathological result showed a malignant tumor in the right vocal cord. The tumor was excised with a CO2 laser (Vc type). Routine postoperative pathology showed moderately differentiated SCC with small cell NEC in the right vocal cord. No metastatic lymph nodes or distant metastases were found on postoperative positron emission tomography/computed tomography. Because of the coexistence of SCC and NEC, the patient received adjuvant chemotherapy and radiotherapy. The patient was followed for 8 mo, and no recurrence or distant metastasis was found. CONCLUSION: The treatment of collision carcinoma in the head and neck region is uncertain due to the small number of cases.

Role of fluorodeoxyglucose-PET versus fluorodeoxyglucose-PET/computed tomography in detection of unknown primary tumor: a meta-analysis of the literature.

The objectives of this study were to evaluate the diagnostic accuracy of fluorodeoxyglucose (FDG)-PET and FDG-PET/computed tomography (CT) in the detection of primary tumors in patients presenting with carcinoma of unknown primary (CUP) unidentified by conventional workup, and to compare the statistical difference between the FDG-PET and FDG-PET/CT. Twenty-eight studies (involving a total of 910 patients) published between 1990 and 2007 were reviewed. These studies evaluated the role of FDG-PET and FDG-PET/CT in the detection of unknown primary tumors after physical examination and conventional workup failed to detect a primary tumor. Systematic methods were used to identify, select, and evaluate the methodological quality of the studies as well as to summarize the overall findings of sensitivity, specificity, and detection capacity of the primary tumor. The overall sensitivity and specificity of FDG-PET in detecting unknown primary tumors were 0.78 [95% confidence interval (CI): 0.72-0.84)] and 0.79 (95% CI: 0.74-0.83), respectively. Furthermore, FDG-PET detected 28.54% of tumors that were not apparent after CUP failed to be detected by conventional workup. Data were collected on the locations of primary tumors detected by FDG-PET in 17 studies and detected by FDG-PET/CT in seven studies. Tumors from the base of the tongue accounted for 20.7% (six of 29) of all false-positive FDG-PET scans, corresponding to a false-positive rate of 28.6% (six of 29), much higher than tumors from the others. FDG-PET exhibited a lower sensitivity with respect to the tumors at the base of the tongue and tonsils, which was 68.2 and 76.7%, respectively. In the eight studies with 430 patients diagnosed with CUP by FDG-PET/CT, 31.4% (n=135) of primary tumors were detected. The pooled sensitivity and specificity were 0.81 (95% CI: 0.74-0.87) and 0.83 (95% CI: 0.78-0.87), respectively. FDG-PET and FDG-PET/CT can detect primary tumors that went undetected by physical examination and conventional workup. FDG-PET exhibited lower sensitivity with respect to the tumors at the base of the tongue and the tonsils.

Palatine tonsil schwannoma: correlation between clinicopathology and computed tomography features.

We describe the case of a 37-year old male who presented with a schwannoma in his right palatine tonsil. The tumour and tonsil were surgically removed transorally. Tonsillar schwannomas are slow growing, and the main symptoms were a foreign body feeling in the throat, dysphagia and snoring. The presence of a round, well-defined mass lesion with heterogeneous density on computed tomography (CT) may aid in differentiation from other tonsillar tumours. The CT findings in the reported case were different to those of schwannomas in other locations; the particular CT manifestations of tonsillar schwannomas seem to relate to the various pathological changes seen within the tumour.

SD-CNN: A shallow-deep CNN for improved breast cancer diagnosis.

Breast cancer is the second leading cause of cancer death among women worldwide. Nevertheless, it is also one of the most treatable malignances if detected early. Screening for breast cancer with full field digital mammography (FFDM) has been widely used. However, it demonstrates limited performance for women with dense breasts. An emerging technology in the field is contrast-enhanced digital mammography (CEDM), which includes a low energy (LE) image similar to FFDM, and a recombined image leveraging tumor neoangiogenesis similar to breast magnetic resonance imaging (MRI). CEDM has shown better diagnostic accuracy than FFDM. While promising, CEDM is not yet widely available across medical centers. In this research, we propose a Shallow-Deep Convolutional Neural Network (SD-CNN) where a shallow CNN is developed to derive "virtual" recombined images from LE images, and a deep CNN is employed to extract novel features from LE, recombined or "virtual" recombined images for ensemble models to classify the cases as benign vs. cancer. To evaluate the validity of our approach, we first develop a deep-CNN using 49 CEDM cases collected from Mayo Clinic to prove the contributions from recombined images for improved breast cancer diagnosis (0.85 in accuracy, 0.84 in AUC using LE imaging vs. 0.89 in accuracy, 0.91 in AUC using both LE and recombined imaging). We then develop a shallow-CNN using the same 49 CEDM cases to learn the nonlinear mapping from LE to recombined images. Next, we use 89 FFDM cases from INbreast, a public database to generate "virtual" recombined images. Using FFDM alone provides 0.84 in accuracy (AUC = 0.87), whereas SD-CNN improves the diagnostic accuracy to 0.90 (AUC = 0.92).

The diagnostic role of diffusion-weighted magnetic resonance imaging in hypopharyngeal carcinoma.

The aim of the present study was to assess the role of diffusion-weighted magnetic resonance imaging (DWI) and apparent diffusion coefficient (ADC) values in hypopharyngeal carcinoma. A total of 40 hypopharyngeal carcinoma tissues and 15 benign lesion tissues were retrospectively analyzed. DWI, and T1- and T2-weighted magnetic resonance imaging (MRI) was performed. The sensitivity, specificity and accuracy of conventional MRI were 97.5, 66.7, and 89.1%, respectively. The mean ADC value [diffusion sensitive factor (b)=1,000× sec/mm2) for hypopharyngeal carcinomas was (1.0285±0.0328)×10-3 mm2/sec, which was significantly lower than the mean ADC value for benign lesions [(1.5333±0.1061)×10-3 mm2/sec; P<0.001]. Receiver operating characteristic (ROC) curve analysis revealed that the area under the curve (AUC) was 0.921 while the optimal threshold for the cut-off point of the ADC was 1.075×10-3 mm2/sec. The mean ADC value of the metastatic nodes was (0.9184±0.0538)×10-3 mm2/sec, lower than the mean value for the benign nodes [(1.2538±0.1145)×10-3 mm2/sec; P=0.005]. Two groups were created according to the mean of the ADC value of hypopharyngeal carcinomas [≤(1.0285±0.0328)×10-3 mm2/sec vs. >(1.0285±0.0328)×10-3 mm2/sec]. The 2-year survival rates of the two groups were 55.6 and 100.0%, respectively (P=0.024). ADC values may aid in distinguishing hypopharyngeal carcinomas from benign lesions and differentiating metastatic lymph nodes of hypopharyngeal squamous cell carcinomas from reactive cervical lymph nodes. In conclusion, mean ADC values may be useful prognostic factors in univariate analysis of hypopharyngeal carcinoma.

[Study of paroxysmal kinesigenic dyskinesia with fMRI].

OBJECTIVE: To detect the differences in subcortical structures between patients with paroxysmal kinesigenic dyskinesia (PKD) and normal subjects during movement preparation and execution. METHODS: The PKD patients performed a movement task, in which a CUE signal (preparation) indicated the movement sequence prior to the appearance of an imperative GO signal (execution). Event-related functional magnetic resonance imaging (fMRI) and 3dDeconvolve program of AFNI were used to estimate the hemodynamic response function and to generate activation maps. RESULT: During movement preparation, the activated brain areas in PKD patients were less than those of normal subject, and there was no activation in basal ganglia in PKD patients. During execution, the activation was also less in PKD patients except in bilateral M1. CONCLUSION: During intermission, abnormalities of the brain still exist in PKD patients when during preparing or performing movement. The movement circuit in the brain displays an unusual state. The attack may be caused by reducing of inhibition in brain areas.

Evaluation of selective hepatic angiography and embolization in patients with massive hemobilia.

BACKGROUND: Massive hemobilia is a rare but potentially life-threatening cause of upper gastrointestinal hemorrhage. Transarterial embolization is considered the first line of intervention to stop the bleeding for most causes of hemobilia. This study was conducted to evaluate selective hepatic angiography and embolization in the diagnosis and treatment of patients with massive hemobilia. METHODS: The clinical data of 16 patients with massive hemobilia were analyzed retrospectively. These patients underwent emergency celiac and selective right or left hepatic artery angiography and treated by embolization using Gelfoam particles and/or coils. RESULTS: Hepatic artery angiography revealed hepatic artery pseudoaneurysms in 6 patients, cystic artery pseudoaneurysms in 2, diffuse hemorrhage of hepatic artery branches in 5, and right hepatic artery-bile duct fistulae in 3. The patients were diagnosed rapidly by angiography and treated successfully by embolization of the hepatic artery branch proximal to the bleeding point, and hemorrhage was stopped immediately. Two patients were embolized the second time for rebleeding. Neither recurrence of bleeding nor serious complication was found during the follow-up for 3 months to 2 years. The other 2 patients whose hemorrhage failed to be controlled died several days later. CONCLUSION: Being safe, reliable and minimally invasive, selective hepatic artery angiography and embolization are effective in the diagnosis and treatment of massive hemobilia.