Visible-light-promoted trifluoromethylselenolation of ortho-hydroxyarylenaminones.

Development of an efficient process that employs easy to handle and shelf-stable reagents for the synthesis of trifluoromethylselenylated heterocyclics remains a daunting challenge in organic synthesis. Herein, we report a green and practical protocol using trifluoromethyl tolueneselenosulfonate and ortho-hydroxyarylenaminones to access a wide range of chromone derivatives under photocatalyst and oxidant free conditions. This reaction proceeded smoothly under photoirradiation conditions and various functional groups were tolerant of the reaction conditions.

Res-TransNet: A Hybrid deep Learning Network for Predicting Pathological Subtypes of lung Adenocarcinoma in CT Images.

This study aims to develop a CT-based hybrid deep learning network to predict pathological subtypes of early-stage lung adenocarcinoma by integrating residual network (ResNet) with Vision Transformer (ViT). A total of 1411 pathologically confirmed ground-glass nodules (GGNs) retrospectively collected from two centers were used as internal and external validation sets for model development. 3D ResNet and ViT were applied to investigate two deep learning frameworks to classify three subtypes of lung adenocarcinoma namely invasive adenocarcinoma (IAC), minimally invasive adenocarcinoma and adenocarcinoma in situ, respectively. To further improve the model performance, four Res-TransNet based models were proposed by integrating ResNet and ViT with different ensemble learning strategies. Two classification tasks involving predicting IAC from Non-IAC (Task1) and classifying three subtypes (Task2) were designed and conducted in this study. For Task 1, the optimal Res-TransNet model yielded area under the receiver operating characteristic curve (AUC) values of 0.986 and 0.933 on internal and external validation sets, which were significantly higher than that of ResNet and ViT models (p < 0.05). For Task 2, the optimal fusion model generated the accuracy and weighted F1 score of 68.3% and 66.1% on the external validation set. The experimental results demonstrate that Res-TransNet can significantly increase the classification performance compared with the two basic models and have the potential to assist radiologists in precision diagnosis.

An attention base U-net for parotid tumor autosegmentation.

A parotid neoplasm is an uncommon condition that only accounts for less than 3% of all head and neck cancers, and they make up less than 0.3% of all new cancers diagnosed annually. Due to their nonspecific imaging features and heterogeneous nature, accurate preoperative diagnosis remains a challenge. Automatic parotid tumor segmentation may help physicians evaluate these tumors. Two hundred eighty-five patients diagnosed with benign or malignant parotid tumors were enrolled in this study. Parotid and tumor tissues were segmented by 3 radiologists on T1-weighted (T1w), T2-weighted (T2w) and T1-weighted contrast-enhanced (T1wC) MR images. These images were randomly divided into two datasets, including a training dataset (90%) and an validation dataset (10%). A 10-fold cross-validation was performed to assess the performance. An attention base U-net for parotid tumor autosegmentation was created on the MRI T1w, T2 and T1wC images. The results were evaluated in a separate dataset, and the mean Dice similarity coefficient (DICE) for both parotids was 0.88. The mean DICE for left and right tumors was 0.85 and 0.86, respectively. These results indicate that the performance of this model corresponds with the radiologist's manual segmentation. In conclusion, an attention base U-net for parotid tumor autosegmentation may assist physicians to evaluate parotid gland tumors.

The use of MRI apparent diffusion coefficient (ADC) in monitoring the development of brain infarction.

BACKGROUND: To study the rules that apparent diffusion coefficient (ADC) changes with time and space in cerebral infarction, and to provide the evidence in defining the infarction stages. METHODS: 117 work-ups in 98 patients with cerebral infarction (12 hyperacute, 43 acute, 29 subacute, 10 steady, and 23 chronic infarctions) were imaged with both conventional MRI and diffusion weighted imaging. The average ADC values, the relative ADC (rADC) values, and the ADC values or rADC values from the center to the periphery of the lesion were calculated. RESULTS: The average ADC values and the rADC values of hyperacute and acute infarction lesion depressed obviously. rADC values in hyperacute and acute stage was minimized, and increased progressively as time passed and appeared as "pseudonormal" values in approximately 8 to 14 days. Thereafter, rADC values became greater than normal in chronic stage. There was positive correlation between rADC values and time (P < 0.01). The ADC values and the rADC values in hyperacute and acute lesions had gradient signs that these lesions increased from the center to the periphery. The ADC values and the rADC values in subacute lesions had adverse gradient signs that these lesions decreased from the center to the periphery. CONCLUSION: The ADC values of infarction lesions have evolution rules with time and space. The evolution rules with time and those in space can be helpful to decide the clinical stage, and to provide the evidence in guiding the treatment or judging the prognosis in infarction.

Diagnosis of Struma ovarii with medical imaging.

OBJECTIVE: To analyze the imaging features of Struma ovarii (SO), and to correlate the imaging results with the pathological findings so as to enhance the knowledge of the imaging diagnostics of this disease. METHODS: The clinical records, CT and MRI features of twelve patients with pathologically proved SO were retrospectively analyzed. Imaging features were compared with pathological results. RESULTS: Most tumors (n = 11, 91.7%) were unilateral. In CT and MRI images, the lesions presented as defined irregularly shaped masses, showing mainly cystic (n = 6, 50%) or cystic (n = 6, 50%). The cystic portions presented as well defined, multiple, various size, and a whole cyst wall with smooth inner wall. Eight cases of tumors (66.7%) showed a high attenuation lesion in the cyst portion of the mass on CT precontrast scans, in which two cases showed high signal on T1WI and low signal on T2WI. The solid portions, which distributed in the cyst showed irregular tissue density. After contrast administration, the cystic portions showed no enhancement, the solid portions marked enhancement, and the cyst wall demonstrated no, moderate, or marked enhancement. Eight cases of tumors (66.7%) showed stippled calcification in the cyst wall. Four cases of tumors (33.3%) accompany a great of abdominal dropsy and pleural fluid. CONCLUSIONS: In general, SO appeared as a smooth marginated multicystic mass with a high attenuation lesion on precontrast scans on CT scans, and signal intensities on T1-weighted images were partly intermediate to high, or high, and those on T2-weighted images were low. The CT and MRI characteristic findings of SO might be of great value for the diagnosis.

Deep Learning for Differentiating Benign From Malignant Parotid Lesions on MR Images.

PURPOSE/OBJECTIVESS: Salivary gland tumors are a rare, histologically heterogeneous group of tumors. The distinction between malignant and benign tumors of the parotid gland is clinically important. This study aims to develop and evaluate a deep-learning network for diagnosing parotid gland tumors via the deep learning of MR images. MATERIALS/METHODS: Two hundred thirty-three patients with parotid gland tumors were enrolled in this study. Histology results were available for all tumors. All patients underwent MRI scans, including T1-weighted, CE-T1-weighted and T2-weighted imaging series. The parotid glands and tumors were segmented on all three MR image series by a radiologist with 10 years of clinical experience. A total of 3791 parotid gland region images were cropped from the MR images. A label (pleomorphic adenoma and Warthin tumor, malignant tumor or free of tumor), which was based on histology results, was assigned to each image. To train the deep-learning model, these data were randomly divided into a training dataset (90%, comprising 3035 MR images from 212 patients: 714 pleomorphic adenoma images, 558 Warthin tumor images, 861 malignant tumor images, and 902 images free of tumor) and a validation dataset (10%, comprising 275 images from 21 patients: 57 pleomorphic adenoma images, 36 Warthin tumor images, 93 malignant tumor images, and 89 images free of tumor). A modified ResNet model was developed to classify these images. The input images were resized to 224x224 pixels, including four channels (T1-weighted tumor images only, T2-weighted tumor images only, CE-T1-weighted tumor images only and parotid gland images). Random image flipping and contrast adjustment were used for data enhancement. The model was trained for 1200 epochs with a learning rate of 1e-6, and the Adam optimizer was implemented. It took approximately 2 hours to complete the whole training procedure. The whole program was developed with PyTorch (version 1.2). RESULTS: The model accuracy with the training dataset was 92.94% (95% CI [0.91, 0.93]). The micro-AUC was 0.98. The experimental results showed that the accuracy of the final algorithm in the diagnosis and staging of parotid cancer was 82.18% (95% CI [0.77, 0.86]). The micro-AUC was 0.93. CONCLUSION: The proposed model may be used to assist clinicians in the diagnosis of parotid tumors. However, future larger-scale multicenter studies are required for full validation.

Deep Learning-Based Stage-Wise Risk Stratification for Early Lung Adenocarcinoma in CT Images: A Multi-Center Study.

This study aims to develop a deep neural network (DNN)-based two-stage risk stratification model for early lung adenocarcinomas in CT images, and investigate the performance compared with practicing radiologists. A total of 2393 GGNs were retrospectively collected from 2105 patients in four centers. All the pathologic results of GGNs were obtained from surgically resected specimens. A two-stage deep neural network was developed based on the 3D residual network and atrous convolution module to diagnose benign and malignant GGNs (Task1) and classify between invasive adenocarcinoma (IA) and non-IA for these malignant GGNs (Task2). A multi-reader multi-case observer study with six board-certified radiologists' (average experience 11 years, range 2-28 years) participation was conducted to evaluate the model capability. DNN yielded area under the receiver operating characteristic curve (AUC) values of 0.76 ± 0.03 (95% confidence interval (CI): (0.69, 0.82)) and 0.96 ± 0.02 (95% CI: (0.92, 0.98)) for Task1 and Task2, which were equivalent to or higher than radiologists in the senior group with average AUC values of 0.76 and 0.95, respectively (p > 0.05). With the CT image slice thickness increasing from 1.15 mm ± 0.36 to 1.73 mm ± 0.64, DNN performance decreased 0.08 and 0.22 for the two tasks. The results demonstrated (1) a positive trend between the diagnostic performance and radiologist's experience, (2) the DNN yielded equivalent or even higher performance in comparison with senior radiologists, and (3) low image resolution decreased model performance in predicting the risks of GGNs. Once tested prospectively in clinical practice, the DNN could have the potential to assist doctors in precision diagnosis and treatment of early lung adenocarcinoma.

Comparison and Fusion of Deep Learning and Radiomics Features of Ground-Glass Nodules to Predict the Invasiveness Risk of Stage-I Lung Adenocarcinomas in CT Scan.

For stage-I lung adenocarcinoma, the 5-years disease-free survival (DFS) rates of non-invasive adenocarcinoma (non-IA) is different with invasive adenocarcinoma (IA). This study aims to develop CT image based artificial intelligence (AI) schemes to classify between non-IA and IA nodules, and incorporate deep learning (DL) and radiomics features to improve the classification performance. We collect 373 surgical pathological confirmed ground-glass nodules (GGNs) from 323 patients in two centers. It involves 205 non-IA (including 107 adenocarcinoma in situ and 98 minimally invasive adenocarcinoma), and 168 IA. We first propose a recurrent residual convolutional neural network based on U-Net to segment the GGNs. Then, we build two schemes to classify between non-IA and IA namely, DL scheme and radiomics scheme, respectively. Third, to improve the classification performance, we fuse the prediction scores of two schemes by applying an information fusion method. Finally, we conduct an observer study to compare our scheme performance with two radiologists by testing on an independent dataset. Comparing with DL scheme and radiomics scheme (the area under a receiver operating characteristic curve (AUC): 0.83 ± 0.05, 0.87 ± 0.04), our new fusion scheme (AUC: 0.90 ± 0.03) significant improves the risk classification performance (p < 0.05). In a comparison with two radiologists, our new model yields higher accuracy of 80.3%. The kappa value for inter-radiologist agreement is 0.6. It demonstrates that applying AI method is an effective way to improve the invasiveness risk prediction performance of GGNs. In future, fusion of DL and radiomics features may have a potential to handle the classification task with limited dataset in medical imaging.

Characteristics of diffusion tensor imaging of central nervous system in children with tourette's disease.

To investigate the characteristics of diffusion tensor imaging (DTI) of the central nervous system in children with Tourette syndrome (TS).Fifteen children with TS (TS group) and 15 normal children (control group) were studied, and all of them underwent DTI. The apparent diffusion coefficient (ADC) and fractional anisotropy (FA) parameters were calculated using the DTIStudio software. The region of interest was delineated manually. The ADC and FA values of the bilateral caudate nucleus, bilateral globus pallidus, bilateral putamen, bilateral thalamus, and bilateral frontal lobe white matter were measured using the region of interest editor software. The differences of FA values and ADC values between the same brain areas were compared. The associations between ADC, FA values and Yale Global Tic Severity Scale (YGTSS) scores were evaluated by Pearson correlation analyses.The FA values of left globus pallidus and left thalamus were significantly lower in the TS group than in the control group (P < .05), while the ADC values of the right caudate nucleus and bilateral thalamus were significantly higher in the TS group than in the control group (P < .05). The decrease in FA in the left thalamus significantly correlated with the YGTSS score (r = 0.692; P < .05). No correlation was found between FA and ADC values in other brain regions and the YGTSS score (P > .05).After the DTI analyses, abnormalities were found in the left globus pallidus, right caudate nucleus, and bilateral thalamus in children with TS. Especially the changes in the left thalamus structure was crucial in the pathophysiological clock of TS.

Baseline SUVmax of 18F-FDG PET-CT indicates prognosis of extranodal natural killer/T-cell lymphoma.

To evaluate the prognostic value of the baseline SUVmax of F-FDG PET-CT in extranodal natural killer/T-cell lymphoma (NKTCL) patients.From January 2010 to December 2015, 141 extranodal NKTCL patients with staging F-FDG PET-CT scan were divided into two group based on SUVmax cutoff value obtained from operating characteristic (ROC) curves. All the patients received radiotherapy, chemotherapy or chemoradiation. Survival analysis was performed on the basis of SUVmax.The median baseline SUVmax of the tumors was 11.67 (range 2.6-34.6). The ROC curves showed that the optimal cutoff of the baseline SUVmax was 9.65. The patients were divided into two groups: low SUV group (SUVmax < 9.65) and high SUV group (SUVmax ≥ 9.65). Patients in high SUV group were more likely to have invasive disease outside the nasal cavity (P < .001), poorer ECOG scores (P = .012) and higher LDH levels (P = .034). The univariate survival analyses indicated that high SUVmax was a poor prognostic factor for overall survival (OS, P = .038), progression free survival (PFS, P = .006) and distant relapse free survival (DRFS, P = .001), but not for local recurrence free survival (LRFS, P > .05). These results were consistent with that of the survival analyses using the Kaplan-Meier method. The multivariate survival analyses showed that the baseline SUVmax was no longer a prognostic factor for OS (HR 1.99, 95% CI 0.81-4.88, P = .135), but it still indicated worse PFS (HR 2.6, 95% CI 1.24-5.46, P = .012) and DRFS (HR 4.58, 95% CI 1.83-11.46, P = .001) independent of other variables.For extranodal NKTCL patients, a higher baseline SUVmax of F-FDG PET-CT was associated with more aggressive clinical features. An SUVmax ≥ 9.65 was an independent poor prognostic factor for DRFS and PFS. Thus, the baseline SUVmax may be a valuable tool to help identify patients with a high risk of disease progression.

Preoperative evaluation of neurovascular relationship in trigeminal neuralgia by three-dimensional fast low angle shot (3D-FLASH) and three-dimensional constructive interference in steady-state (3D-CISS) MRI sequence.

OBJECTIVE: The purpose of the study was to evaluate the value of high-resolution three-dimensional fast low angle shot (3D-FLASH) and three-dimensional constructive interference in steady-state (3D-CISS) MRI sequence solely or the combination of both in the visualization of neurovascular relationship in patients with trigeminal neuralgia (TN). METHODS: 65 patients with unilateral TN underwent 3D-FLASH and 3D-CISS imaging were retrospectively studied. Neurovascular relationship at the intracisternal segment of trigeminal nerve was reviewed by two experienced neuroradiologist, who was blinded to the clinical details. The imaging results were compared with the operative findings in all patients. RESULTS: The accuracy and positive rates of the 3D-FLASH + CISS imaging (98.46, 92.31%) in judging the symptomatic side according to the presence of vascular contacts were higher than those of 3D-CISS (90.77%, 84.62) or 3D-FLASH (89.23, 83.08%) sequence. In addition, the statistical analysis showed the sensitivity and accuracy of 3D-FLASH + CISS imaging was higher than that of 3D-FLASH (p < 0.05). The 3D-FLASH + CISS imaging was more accurate in determining the type of offending vessel than 3D-CISS or 3D-FLASH imaging. CONCLUSION: The retrospective study demonstrates that the combination of 3D-FLASH with 3D-CISS sequence well delineates the relationship between intracisternal segment of trigeminal nerve and adjacent vessels in terms of increased positive rates and accuracy. Advances in knowledge: The study firstly dealt with the combination of 3D-CISS and 3D-FLASH imaging in TN.

[Imaging manifestation and clinical value in herniation pit of femoral neck].

OBJECTIVE: To discuss the imaging manifestation and clinical value in herniation pit of femoral neck. METHODS: One case proved by operation and pathology and twenty cases with typical imaging manifestation described by Pitt were reviewed retrospectively. There were 17 males and 4 females with an average age of 53 years old(ranging from 30 to 85 years). All cases were examined by X-ray films and CT, and 13 cases were performed with MRI. RESULTS: Twenty-nine lesions were found in the 21 cases, 9 cases were in right side, 8 cases were in left side, 4 cases were in both sides. The lesions were all located in the superior lateral part of the femoral neck and anterior lateral base of femoral head. The lesions were round or oval, and most of their greatest diameter was less than 16 mm. X-ray films showed a central radiolucency with a thin clear sclerotic rim or simple sclerotic loop. CT scans showed a well-defined lesion of soft-tissue attenuation with sclerotic margin. The lesions had focal cortical perforation. On MRI images,most lesions showed uniformly long T1 and long T2 fluid signal intensity. CONCLUSION: Herniation pit of femoral neck have some specific imaging features, CT can make accurate diagnosis. X-ray and MRI are helpful to diagnosis.