A biomedical event extraction method based on fine-grained and attention mechanism.

BACKGROUND: Biomedical event extraction is a fundamental task in biomedical text mining, which provides inspiration for medicine research and disease prevention. Biomedical events include simple events and complex events. Existing biomedical event extraction methods usually deal with simple events and complex events uniformly, and the performance of complex event extraction is relatively low. RESULTS: In this paper, we propose a fine-grained Bidirectional Long Short Term Memory method for biomedical event extraction, which designs different argument detection models for simple and complex events respectively. In addition, multi-level attention is designed to improve the performance of complex event extraction, and sentence embeddings are integrated to obtain sentence level information which can resolve the ambiguities for some types of events. Our method achieves state-of-the-art performance on the commonly used dataset Multi-Level Event Extraction. CONCLUSIONS: The sentence embeddings enrich the global sentence-level information. The fine-grained argument detection model improves the performance of complex biomedical event extraction. Furthermore, the multi-level attention mechanism enhances the interactions among relevant arguments. The experimental results demonstrate the effectiveness of the proposed method for biomedical event extraction.

A blind medical image denoising method with noise generation network.

BACKGROUND: In the process of medical images acquisition, the unknown mixed noise will affect image quality. However, the existing denoising methods usually focus on the known noise distribution. OBJECTIVE: In order to remove the unknown real noise in low-dose CT images (LDCT), a two-step deep learning framework is proposed in this study, which is called Noisy Generation-Removal Network (NGRNet). METHODS: Firstly, the output results of L0 Gradient Minimization are used as the labels of a dental CT image dataset to form a pseudo-image pair with the real dental CT images, which are used to train the noise generation network to estimate real noise distribution. Then, for the lung CT images of the LIDC/IDRI database, we migrate the real noise to the noise-free lung CT images, to construct a new almost-real noisy images dataset. Since dental images and lung images are all CT images, this migration can be achieved. The denoising network is trained to realize the denoising of real LDCT for dental images by using this dataset but can extend for any low-dose CT images. RESULTS: To prove the effectiveness of our NGRNet, we conduct experiments on lung CT images with synthetic noise and tooth CT images with real noise. For synthetic noise image datasets, experimental results show that NGRNet is superior to existing denoising methods in terms of visual effect and exceeds 0.13dB in the peak signal-to-noise ratio (PSNR). For real noisy image datasets, the proposed method can achieve the best visual denoising effect. CONCLUSIONS: The proposed method can retain more details and achieve impressive denoising performance.

Joint Extraction of Biomedical Events Based on Dynamic Path Planning Strategy and Hybrid Neural Network.

Biomedical event detection is a pivotal information extraction task in molecular biology and biomedical research, which provides inspiration for the medical search, disease prevention, and new drug development. The existing methods usually detect simple biomedical events and complex events with the same model, and the performance of the complex biomedical event extraction is relatively low. In this paper, we build different neural networks for simple and complex events respectively, which helps to promote the performance of complex event extraction. To avoid redundant information, we design dynamic path planning strategy for argument detection. To take full use of the information between the trigger identification and argument detection subtasks, and reduce the cascading errors, we build a joint event extraction model. Experimental results demonstrate our approach achieves the best F-score on the biomedical benchmark MLEE dataset and outperforms the recent state-of-the-art methods.

DriverGenePathway: Identifying driver genes and driver pathways in cancer based on MutSigCV and statistical methods.

Although computational methods for driver gene identification have progressed rapidly, it is far from the goal of obtaining widely recognized driver genes for all cancer types. The driver gene lists predicted by these methods often lack consistency and stability across different studies or datasets. In addition to analytical performance, some tools may require further improvement regarding operability and system compatibility. Here, we developed a user-friendly R package (DriverGenePathway) integrating MutSigCV and statistical methods to identify cancer driver genes and pathways. The theoretical basis of the MutSigCV program is elaborated and integrated into DriverGenePathway, such as mutation categories discovery based on information entropy. Five methods of hypothesis testing, including the beta-binomial test, Fisher combined p-value test, likelihood ratio test, convolution test, and projection test, are used to identify the minimal core driver genes. Moreover, de novo methods, which can effectively overcome mutational heterogeneity, are introduced to identify driver pathways. Herein, we describe the computational structure and statistical fundamentals of the DriverGenePathway pipeline and demonstrate its performance using eight types of cancer from TCGA. DriverGenePathway correctly confirms many expected driver genes with high overlap with the Cancer Gene Census list and driver pathways associated with cancer development. The DriverGenePathway R package is freely available on GitHub: https://github.com/bioinformatics-xu/DriverGenePathway.

Predicting anticancer drug sensitivity on distributed data sources using federated deep learning.

Drug sensitivity prediction plays a crucial role in precision cancer therapy. Collaboration among medical institutions can lead to better performance in drug sensitivity prediction. However, patient privacy and data protection regulation remain a severe impediment to centralized prediction studies. For the first time, we proposed a federated drug sensitivity prediction model with high generalization, combining distributed data sources while protecting private data. Cell lines are first classified into three categories using the waterfall method. Focal loss for solving class imbalance is then embedded into the horizontal federated deep learning framework, i.e., HFDL-fl is presented. Applying HFDL-fl to homogeneous and heterogeneous data, we obtained HFDL-Cross and HFDL-Within. Our comprehensive experiments demonstrated that (i) collaboration by HFDL-fl outperforms private model on local data, (ii) focal loss function can effectively improve model performance to classify cell lines in sensitive and resistant categories, and (iii) HFDL-fl is not significantly affected by data heterogeneity. To summarize, HFDL-fl provides a valuable solution to break down the barriers between medical institutions for privacy-preserving drug sensitivity prediction and therefore facilitates the development of cancer precision medicine and other privacy-related biomedical research.

Double enhanced residual network for biological image denoising.

With the achievements of deep learning, applications of deep convolutional neural networks for the image denoising problem have been widely studied. However, these methods are typically limited by GPU in terms of network layers and other aspects. This paper proposes a multi-level network that can efficiently utilize GPU memory, named Double Enhanced Residual Network (DERNet), for biological-image denoising. The network consists of two sub-networks, and U-Net inspires the basic structure. For each sub-network, the encoder-decoder hierarchical structure is used for down-scaling and up-scaling feature maps so that GPU can yield large receptive fields. In the encoder process, the convolution layers are used for down-sampling to obtain image information, and residual blocks are superimposed for preliminary feature extraction. In the operation of the decoder, transposed convolution layers have the capability to up-sampling and combine with the Residual Dense Instance Normalization (RDIN) block that we propose, extract deep features and restore image details. Finally, both qualitative experiments and visual effects demonstrate the effectiveness of our proposed algorithm.

Clinical characteristics of remote Zeus robot-assisted laparoscopic cholecystectomy: a report of 40 cases.

AIM: To summarize the performing essentials and analyze the characteristics of remote Zeus robot-assisted laparoscopic cholecystectomy. METHODS: Robot-assisted laparoscopic cholecystectomy was performed in 40 patients between May 2004 and July 2005. The operating procedures and a variety of clinical parameters were recorded and analyzed. RESULTS: Forty laparoscopic cholecystectomy procedures were successfully completed with Zeus robotic system. And there were no post-operative complications. Total operating time, system setup time and performing time were 100.3 +/- 18.5 min, 27.7 +/- 8.8 min and 65.6 +/- 18.3 min, respectively. The blood loss and post-operative hospital stay were 30.6 +/- 10.2 mL and 2.8 +/- 0.8 d, respectively. Camera clearing times and time used for operative field adjustment were 1.1+/- 1.0 min and 2.0 +/- 0.8 min, respectively. The operative error was 7.5%. CONCLUSION: Robot-assisted laparoscopic cholecystectomy following the principles of laparoscopic operation has specific performing essentials. It preserves the benefits of minimally invasive surgery and offers enhanced ability of controlling operation field, precise and stable operative manipulations.

Zeus robot-assisted laparoscopic cholecystectomy in comparison with conventional laparoscopic cholecystectomy.

BACKGROUND: The robotic surgical system overcomes many technological obstacles of conventional laparoscopic surgery, and possesses enormous clinical applied potential. The aim of this study was to compare the efficacy of Zeus robot-assisted laparoscopic cholecystectomy with conventional laparoscopic cholecystectomy. METHODS: Forty patients undergoing elective cholecystectomy were randomly divided into two groups. Patients in group A (n=20) underwent Zeus robot-assisted laparoscopic cholecystectomy, and patients in group B (n=20) received conventional laparoscopic cholecystectomy. The parameters on operative field, operative time, the number of actions, the rate of operative errors and minimal trauma were evaluated and compared between the two groups. RESULTS: The number of clearing camera (1.1+/-1.0 times) and the time of adjusting the operative field (2.2+/-0.7 minutes) in group A were significantly less than those (4.5+/-1.5 times) and (7.5+/-1.2 minutes) in group B. The number of dissection actions (337+/-86 times) and the rate of operative errors (10%) in group A were less than those (389+/-94 times), (25%) in group B. The total operation time (104.9+/-20.5 minutes) and setup time (29.5+/-9.8 minutes) in group A were significantly longer than those (78.6+/-17.1 minutes), (12.6+/-2.5 minutes) in group B. Blood loss and postoperative hospitalization were similar. No postoperative complications occurred in both groups, and open cholecystectomy was performed in each group. CONCLUSIONS: Zeus robot-assisted cholecystectomy inherits the benefits of minimally invasive surgery. The Zeus robotic surgical system is better than conventional laparoscopic technique in controlling the operative field and can be manipulated precisely and stably though it requires more operative time.

[Laparoscopic cholecystectomy with remote zeus surgical robotic system: report of 16 cases].

OBJECTIVE: To explore the methodology and operative essentials of laparoscopic cholecystectomy with remote Zeus surgical robotic system. METHODS: Based on strict training and successful experiment in animal model of swine, laparoscopic cholecystectomy using Zeus robotic system was performed on 16 patients with biliary diseases, including choledocholithiasis, cholelithiasis, polyposis of gallbladder, and chronic cholecystitis, 10 males and 16 females, aged 33 (14 approximately 27), 26 April to 31 August 2004. The general data, preoperative preparation time, operation time, amount of bleeding, complications, and hospitalization time were analyzed. RESULTS: All operations were performed without event. Along with the accumulation of experience the preoperative preparation time was shortened from 90 min to 30 min with an average of 41.7 min, and the operation time from 120 min to 30 min with an average of 64.4 min. The average amount of bleeding was 27.7 ml, and the average postoperative hospitalization time was 2.4 d. A telephone follow-up 30 days after operation showed no abnormality. CONCLUSION: Laparoscopic cholecystectomy with Zeus surgical robotic system is feasible and reliable with the advantages of clearer images in the field of operation, more precise handling, and remote surgery or education.

Clinical practice of video-assisted thoracoscopic surgery in children.

OBJECTIVE To estimate the value of video-assisted thoracoscopic surgery (VATS) in diagnosis and treatment of children with chest diseases. METHODS From May 1997 to October 2001, forty-one children (25 boys and 16 girls) with chest diseases received VATS under general anesthesia in our hospital. Their average age was 6.9 years (range 9 days to 16 years) and their median body weight was 22.5 kg (2.8-54.0 kg). Operative procedures included fibrinopurulent empyema with debridement and decortication in 15 children, biopsy and(or) resection of mediastinal tumor in 11, bullectomy and cystectomy of the lung in 6, lobectomy with huge cyst of the lung or sequestration in 5, clearance of hemothorax in 2, and exploration, and repair of diaphragmatic hernia in 2. RESULTS The mean operative time was 74 minutes (range 30 to 220 minutes). The lost blood volume was 33 ml (range 10 to 150 ml). The mean duration of chest drainage and hospital stay after surgery was 2.4 days and 7.0 days (range 4 to 15 days) respectively. One infant born after 9 days with congenital diaphragmatocele died of respiratory failure due to left pulmonary hypoplasia 10 days after operation. Postoperative morbidity was 7.3% (3 patients). Forty patients were followed up for an average of 15.6 months, 38 patients lived and developed normally, and 2 received chemotherapy. CONCLUSION Video-assisted thoracoscopy is a safe and effective diagnostic and therapeutic procedure for children with chest disease, and this approach has an important place in pediatric thoracic surgery.