Branchial cleft cyst arising in posterior mediastinum: A case report.

Branchial cleft cysts are congenital diseases of the neck caused by abnormal embryonic development of the first to fourth branchial clefts. Most branchial cleft cysts are found in the head and neck, but branchial cleft cysts arising in posterior mediastinum are rarely reported. We report a 44-year-old Chinese man who was found to have a right-posterior mediastinal mass on chest computed tomography (CT) during a physical examination. The size of the mass was about 30.6 mm * 25.1 mm and enhanced CT of the chest showed an occupying lesion in the right parietal esophagus of the upper-posterior mediastinum with no significant enhancement. The patient was considered to have a neurogenic tumor with cystic change and underwent posterior mediastinal tumor resection. Postoperatively, pathological examination confirmed the final diagnosis of bronchial cleft cyst. The patient was discharged on the 7th day after surgery. One year postsurgery, no obvious recurrence was found in reexamination.

Transformer for Gene Expression Modeling (T-GEM): An Interpretable Deep Learning Model for Gene Expression-Based Phenotype Predictions.

Deep learning has been applied in precision oncology to address a variety of gene expression-based phenotype predictions. However, gene expression data's unique characteristics challenge the computer vision-inspired design of popular Deep Learning (DL) models such as Convolutional Neural Network (CNN) and ask for the need to develop interpretable DL models tailored for transcriptomics study. To address the current challenges in developing an interpretable DL model for modeling gene expression data, we propose a novel interpretable deep learning architecture called T-GEM, or Transformer for Gene Expression Modeling. We provided the detailed T-GEM model for modeling gene-gene interactions and demonstrated its utility for gene expression-based predictions of cancer-related phenotypes, including cancer type prediction and immune cell type classification. We carefully analyzed the learning mechanism of T-GEM and showed that the first layer has broader attention while higher layers focus more on phenotype-related genes. We also showed that T-GEM's self-attention could capture important biological functions associated with the predicted phenotypes. We further devised a method to extract the regulatory network that T-GEM learns by exploiting the attributions of self-attention weights for classifications and showed that the network hub genes were likely markers for the predicted phenotypes.

Esophageal hiatal hernia: risk, diagnosis and management.

INTRODUCTION: Esophageal hiatal hernia involves abnormal abdominal entry into thoracic cavity. It is classified based on orientation between esophageal junction and diaphragm. Sliding hiatal hernia (Type-I) comprises the most frequent category, emanating from right crus of diaphragm. Type-II esophageal hernia engages both left and right muscular crura. Type-III and IV additionally include the left crus. Age and increased body mass index are key risk factors, and congenital skeletal aberrations trigger pathogenesis through intestinal malrotations. Familiar manifestations include gastric reflux, nausea, bloating, chest and epigastric discomfort, pharyngeal and esophageal expulsion and dysphagia. Weight loss and colorectal bleeding are severe symptoms. Areas covered: This review summarizes updated evidence of pathophysiology, risk factors, diagnosis and management of hiatal hernias. Laparoscopy and oesophagectomy procedures have been discussed as surgical procedures. Expert commentary: Endoscopy identifies untreatable gastric reflux; radiology is better for pre-operative assessments; manometry measures esophageal peristalsis, and CT scanning detects gastric volvulus and associated organ ruptures. Gastric reflux disease is mitigated using antacids and proton pump and histamine-2-receptor blockers. Severe abdominal penetration into chest cavity demands surgical approaches. Hence, esophagectomy has chances of post-operative morbidity, while minimally invasive laparoscopy entails fewer postoperative difficulties and better visualization of hernia and related vascular damages.

[The first and second cluster analysis of dual index grade sequence of IR fingerprint spectra of Guifudihuang pill and Jinkuishenqi pill samples and their quality evaluation].

In the present paper a method for identifying and evaluating traditional Chinese medicine (TCM) Jinkuishenqi pill and Guifudihuang pill samples is proposed. Mathematical statistics was applied to analyze the dual index sequences of infrared fingerprint spectra of components extracted from Jinkuishenqi pill and Guifudihuang pill samples with water, then the grade sequences were determined and the first and second clusters were performed based on these grade sequences. As a results, the most similar sample groups of different samples were also obtained. The Jinkuishenqi pill and Guifudihuang pill samples can be distinguished accurately with the most similar sample groups. The method put forward is new and effective to identify and evaluate combinational TCM quantitatively.