An imConvNet-based deep learning model for Chinese medical named entity recognition.

BACKGROUND: With the development of current medical technology, information management becomes perfect in the medical field. Medical big data analysis is based on a large amount of medical and health data stored in the electronic medical system, such as electronic medical records and medical reports. How to fully exploit the resources of information included in these medical data has always been the subject of research by many scholars. The basis for text mining is named entity recognition (NER), which has its particularities in the medical field, where issues such as inadequate text resources and a large number of professional domain terms continue to face significant challenges in medical NER. METHODS: We improved the convolutional neural network model (imConvNet) to obtain additional text features. Concurrently, we continue to use the classical Bert pre-training model and BiLSTM model for named entity recognition. We use imConvNet model to extract additional word vector features and improve named entity recognition accuracy. The proposed model, named BERT-imConvNet-BiLSTM-CRF, is composed of four layers: BERT embedding layer-getting word embedding vector; imConvNet layer-capturing the context feature of each character; BiLSTM (Bidirectional Long Short-Term Memory) layer-capturing the long-distance dependencies; CRF (Conditional Random Field) layer-labeling characters based on their features and transfer rules. RESULTS: The average F1 score on the public medical data set yidu-s4k reached 91.38% when combined with the classical model; when real electronic medical record text in impacted wisdom teeth is used as the experimental object, the model's F1 score is 93.89%. They all show better results than classical models. CONCLUSIONS: The suggested novel model (imConvNet) significantly improves the recognition accuracy of Chinese medical named entities and applies to various medical corpora.

Visualisation of Droplet Flow Induced by Ultrasonic Dental Cleaning.

INTRODUCTION: During dental treatment procedures ultrasonic scalers generate droplets containing microorganisms such as bacteria and viruses. Hence, it is necessary to study the dynamic properties of generated droplets in order to investigate the risks associated with the spread of infection. The aim of this study was to visualise the flow state of droplets and to evaluate the impact of droplets generated during the use of an ultrasonic scaler during an oral surgical procedure. METHODS: We studied the spatial flow of liquid droplets through a combination of imaging and numeric simulation of a simulated dental treatment processes. First, we photographed the real time images of the ultrasonic scaler and evaluated the images using image-processing software Image J to visualise the flow of liquid droplets. Finally we simulated the flow process of liquid droplets by using the initial velocity of droplet splashing and the angle of the obtained information using computerised fluid dynamics technology. RESULTS: Under different working conditions, the droplet particle splashing velocity, maximum height, and spray angle varied, but the particle trajectory was generally parabolic. The maximum droplet velocity varied between 3.56 and 8.56 m/s, and the splashing height was between 40 and 110 mm. CONCLUSIONS: During risk assessment of an ultrasonic scaler usage, difficulties arise due to the insufficient research on droplet velocity and distribution. This study aims to address this gap by visualising the flow trajectories of droplets generated by ultrasonic scalers. The obtained data will assist in developing more effective interventions based on spatial and temporal distribution of droplets. This provides a new approach for droplet particle research and offers new strategies for public health prevention and control.

Caerin 1.9-Titanium Plates Aid Implant Healing and Inhibit Bacterial Growth in New Zealand Rabbit Mandibles.

OBJECTIVES: With rising rates of maxillofacial fracture, postoperative infection following rigid internal fixation is an important issue that requires immediate resolution. It is important to explore an alternative antibacterial method apart from conventional antibiotics. A controlled experiment was conducted to evaluate the effectiveness of a caerin 1.9 peptide-coated titanium plate in reducing mandibular infection in New Zealand (NZ) rabbits, aiming to minimise the risk of post-metallic implantation infection. METHODS: Twenty-two NZ rabbits were randomly divided into 3 groups. The experiment group received caerin 1.9 peptide-coated titanium plates and mixed oral bacteria exposure. The control group received normal titanium plates with mixed oral bacteria exposure. The untreated group served as a control to prove that bacteria in the mouth can cause infection. Weight, temperature, hepatic function, and C-reactive protein levels were measured. Wound and bone conditions were evaluated. Further analysis included local infection, anatomic conditions, histology, and bacterial load. RESULTS: No significant differences were found in temperature, weight, blood alanine aminotransferase, and C-reactive protein levels amongst the 3 groups. The experiment group showed the lowest amount of bacterial RNA in wounds. Additionally, the experiment group had higher peripheral lymphocyte counts compared to the control group and lower neutrophil counts on the third and seventh day postoperatively. Histologic analysis revealed lower levels of inflammatory cell infiltration, bleeding, and areas of necrosis in the experimental group compared with the controls. CONCLUSIONS: A caerin 1.9-coated titanium plate is able to inhibit bacterial growth in a NZ rabbit mandibular mixed bacteria infection model and is worth further investigation.