Multi-omics and Multi-VOIs to predict esophageal fistula in esophageal cancer patients treated with radiotherapy.

OBJECTIVE: This study aimed to develop a prediction model for esophageal fistula (EF) in esophageal cancer (EC) patients treated with intensity-modulated radiation therapy (IMRT), by integrating multi-omics features from multiple volumes of interest (VOIs). METHODS: We retrospectively analyzed pretreatment planning computed tomographic (CT) images, three-dimensional dose distributions, and clinical factors of 287 EC patients. Nine groups of features from different combination of omics [Radiomics (R), Dosiomics (D), and RD (the combination of R and D)], and VOIs [esophagus (ESO), gross tumor volume (GTV), and EG (the combination of ESO and GTV)] were extracted and separately selected by unsupervised (analysis of variance (ANOVA) and Pearson correlation test) and supervised (Student T test) approaches. The final model performance was evaluated using five metrics: average area under the receiver-operator-characteristics curve (AUC), accuracy, precision, recall, and F1 score. RESULTS: For multi-omics using RD features, the model performance in EG model shows: AUC, 0.817 ± 0.031; 95% CI 0.805, 0.825; p < 0.001, which is better than single VOI (ESO or GTV). CONCLUSION: Integrating multi-omics features from multi-VOIs enables better prediction of EF in EC patients treated with IMRT. The incorporation of dosiomics features can enhance the model performance of the prediction.

Protection of the marginal mandibular branches of the facial nerves by different surgical procedures in comprehensive cervical lymphadenectomy for locally advanced oral and oropharyngeal cancer: a multicenter experience.

OBJECTIVE: According to the different characteristics of patients and cervical lymph node metastasis of oral and oropharyngeal cancer, the marginal mandibular branches of facial nerves were treated by different surgical procedures, and the safety and protective effects of different surgical procedures were investigated. METHODS: One hundred ninety-seven patients with oral and oropharyngeal cancer satisfying the inclusion criteria were selected. According to the different characteristics of patients and cervical metastatic lymph nodes, three different surgical procedures were used to treat the marginal mandibular branches of the facial nerve: finding and exposing the marginal mandibular branches of the facial nerves at the mandibular angles of the platysma flaps, finding and exposing the marginal mandibular branches of facial nerves at the intersections of the distal ends of facial arteries and veins with the mandible, and not exposing the marginal mandibular branches of the facial nerves. The anatomical position, injury, and complications of the marginal mandibular branches of the facial nerves were observed. RESULTS: The marginal mandibular branches of the facial nerves were found and exposed at the mandibular angles of the platysma flaps in 102 patients; the marginal mandibular branches of facial nerves were found and exposed at the intersections of the distal ends of the facial arteries and veins with the mandibles in 64 patients; the marginal mandibular branches of facial nerves were not exposed in 31 patients; among them, four patients had permanent injury of the marginal mandibular branches of the facial nerves, and temporary injury occurred in seven patients. There were statistically significant differences in the protection of the mandibular marginal branch of the facial nerve among the three different surgical methods (P = 0.0184). The best protective effect was to find and expose the mandibular marginal branch of the facial nerve at the mandibular angle of the platysma muscle flap, and the injury rate was only 2.94%. CONCLUSION: The three different surgical procedures were all safe and effective in treating the marginal mandibular branches of the facial nerves, the best protective effect was to find and expose the mandibular marginal branch of the facial nerve at the mandibular angle of the platysma muscle flap.

Multi-Modality Fusion & Inductive Knowledge Transfer Underlying Non-Sparse Multi-Kernel Learning and Distribution Adaption.

With the development of sensors, more and more multimodal data are accumulated, especially in biomedical and bioinformatics fields. Therefore, multimodal data analysis becomes very important and urgent. In this study, we combine multi-kernel learning and transfer learning, and propose a feature-level multi-modality fusion model with insufficient training samples. To be specific, we firstly extend kernel Ridge regression to its multi-kernel version under the lp-norm constraint to explore complementary patterns contained in multimodal data. Then we use marginal probability distribution adaption to minimize the distribution differences between the source domain and the target domain to solve the problem of insufficient training samples. Based on epilepsy EEG data provided by the University of Bonn, we construct 12 multi-modality & transfer scenarios to evaluate our model. Experimental results show that compared with baselines, our model performs better on most scenarios.

Hybrid-Solvent Electrolytes for Enhanced Potassium-Oxygen Battery Performance.

Rechargeable potassium-oxygen batteries (KOB) are promising next-generation energy storage devices because of the highly reversible O2/O2- redox reactions during battery charge and discharge. However, the complicated cathode reaction processes seriously jeopardize the battery reaction kinetics and discharge capacity. Herein, we propose a hybrid-solvent strategy to effectively tune the K+ solvation structure, which demonstrates a critical influence on the charge-transfer kinetics and cathode reaction mechanism. The cosolvation of K+ by 1,2-dimethoxyethane (DME) and dimethyl sulfoxide (DMSO) could greatly decrease overpotentials for the cathode processes and increase the cathode discharge capacity. Furthermore, the Coulombic efficiency for the cathode could be significantly improved with the enhanced solution-mediated KO2 growth and stripping during cycling. This work provides a promising electrolyte design approach to improve the electrochemical performance of the KOB.

EEG Feature Selection via Stacked Deep Embedded Regression With Joint Sparsity.

In the field of brain-computer interface (BCI), selecting efficient and robust features is very seductive for artificial intelligence (AI)-assisted clinical diagnosis. In this study, based on an embedded feature selection model, we construct a stacked deep structure for feature selection in a layer-by-layer manner. Its promising performance is guaranteed by the stacked generalized principle that random projections added into the original features can help us to continuously open the manifold structure existing in the original feature space in a stacked way. With such benefits, the original input feature space becomes more linearly separable. We use the epilepsy EEG data provided by the University of Bonn to evaluate our model. Based on the EEG data, we construct three classification tasks. On each task, we use different feature selection models to select features and then use two classifiers to perform classification based on the selected features. Our experimental results show that features selected by our new structure are more meaningful and helpful to the classifier hence generates better performance than benchmarking models.

The preservation and application of the submandibular gland in oral squamous cell carcinoma (STROBE).

This prospective study aimed to investigate metastases of oral squamous cell carcinoma (OSCC) to cervical lymph nodes and submandibular glands and to analyze the safety and feasibility of preserving and using submandibular glandular flaps to repair postoperative OSCC defects.Overall, 330 patients with OSCC who met the inclusion criteria were enrolled in the study from January 2014 to July 2018. OSCC metastasis to cervical lymph nodes and submandibular glands was investigated using intraoperative frozen section and postoperative pathological observation. Fifteen patients who underwent repair of postoperative OSCC defects with submandibular glandular flaps were monitored for postoperative wound healing, complications, pathology, and appearance satisfaction and were followed up long term.Among the 330 patients with OSCC, the most common type was tongue cancer (138/330); 204 patients were node negative and 126 were node positive. Of 363 samples of the submandibular gland, 7 were metastatic with stage IV lesion, 5 were directly invaded by the primary tumor, and 2 were metastatic with extranodal extension in level Ib. None of the submandibular gland samples showed intraglandular hematogenous and nodal metastases. The repair of OSCC defects with submandibular glandular flaps was successful in all 15 patients, including 11 males and 4 females, with an age range of 19-74 years and a mean of 51 years. Of 15 patients, some had complications with heart and cerebrovascular diseases, diabetes, or chronic respiratory disease. All the submandibular glandular flaps survived postoperatively without any complications, and the mucosification on the surface was adequate. Follow-ups (median duration: 14 months) revealed the satisfactory recovery of shape and function without any local recurrences or distant metastases.Direct invasion is the main form of OSCC metastasis to the submandibular gland, while hematogenous and nodal metastases are uncommon. Preservation of the submandibular gland is oncologically safe. Thus, repair of postoperative OSCC defects with submandibular gland flaps is a feasible and promising procedure.

Clinicopathological and prognostic significance of octamer-binding transcription factor 4 in patients with gastric cancer: a systematic review and meta-analysis.

AIM: The role of octamer-binding transcription factor 4 (Oct4) in gastric cancer (GC) progression is still under debate and reported results are inconsistent. Therefore, we conducted a meta-analysis to evaluate the clinicopathological and prognostic significance of Oct4 expression in patients with GC. MATERIALS & METHODS: Relevant articles were retrieved from a diverse number of databases, and meta-analysis was completed using STATA software 12.0. RESULTS: Total of 21 studies were included in this analysis (3209 samples). Expression of Oct4 was associated with incidence, tumor size, lymph node metastasis, histological differentiation, pTNM stage, tumor depth of infiltration, vascular invasion and distal metastasis. Additionally, Oct4 expression was correlated with poor overall survival rate. CONCLUSION: The Oct4 overexpression suggested aggressive biological behaviors and imply that Oct4 may be a useful prognostic biomarker in gastric cancers.

DNA Melting Analysis with Optofluidic Lasers Based on Fabry-Pérot Microcavity.

We conduct DNA high-resolution melting (HRM) analysis using optofluidic lasers based on a Fabry-Pérot microcavity. Compared to the fluorescence-based HRM, the laser-based HRM has advantages of higher emission intensity for better signal-to-noise ratio and sharper transition for better temperature resolution. In addition, the melting temperature can be lowered by optimizing the laser conditions such as external pump and cavity Q-factor. In this work, we first theoretically analyze the laser-based HRM. Then experiments are performed on three long DNA sequences as model systems, one being 99 bases and the other two being 130 bases long but with different GC contents. We show that the laser-based HRM is able to distinguish the target and the single-base mismatched DNA as long as 130 bases and with nearly 50% GC content. The dependence of laser threshold on the temperature for each DNA sample is first experimentally investigated and by optimizing the external pump, the melting temperature is reduced by more than 10 °C, compared to the fluorescence-based HRM for long DNA sequences up to 130 bases. Finally, we demonstrate an alternative method of using the laser-based HRM for rapid DNA screening that does not exist for the fluorescence-based HRM, in which laser excitation is scanned at a fixed temperature to distinguish the target and the base-mismatched DNA sequences. It is shown that the 130-bases-long DNA with nearly 50% GC content can have as much as 20% difference in the laser threshold and 40% difference in the laser output slope between the target and the single-base mismatched sequences, despite only 0.5 °C difference in their melting temperature, indicating that the laser-excitation-scanning method can also be suitable for long DNA sequences with higher GC content.

A systematic analysis of the association between Notch1 expression and the patients with digestive tract cancers.

AIM: Notch1 expression remains controversial on digestive tract cancers. This meta-analysis was performed to assess the clinicopathological significance of Notch1 expression in individuals with digestive tract cancers, mainly involving esophageal squamous cell carcinoma (ESCC), gastric cancer (GC), pancreatic cancer (PC) and colorectal cancer (CRC). METHODS: Available articles were searched from the online databases, and the meta-analysis was done using Review Manager software 5.3. RESULTS: 35 studies were included in this analysis (6187 samples). Notch1 is downregulated in esophageal squamous cell carcinoma (p < 0.00001), Notch1 expression at high levels was detected in GC (p = 0.02) and CRC (p < 0.001), and no significant difference exists between PC and normal tissue (p = 0.76). CONCLUSION: Notch1 overexpression in GC and CRC suggested aggressive biological behaviors, and Notch1 may be a biomarker in digestive tract cancers.