Abstract:As a new generation of artificial intelligence technology, the deep neural network takes the cognitive ability of machine to a historical high level in natural language processing, learning ability and computer vision. At present, the application of deep neural network in medical imaging can be categorized into discovery of anomalies, quantitative measurement, and differential diagnosis. Medical imaging research based on deep neural network research has involved various medical imaging domains such as radiological imaging, pathological images, ultrasound imaging, and endoscopic imaging. In several tasks, deep neural network has demonstrated physician-level or even above-physician-level performance. In the context of rapid development of artificial intelligence in imaging medicine, physicians should adopt a more objective, scientific, and proactive attitude towards artificial intelligence technology, and become the masters of artificial intelligence technology and the creators of a futuristic medical world assisted by artificial intelligence technology.

Abstract:The incidence of rare diseases is extremely low, but the overall number of patients with rare diseases is quite large. The consequences of rare diseases are severe and impose a heavy burden on patients, their families and the entire society. Although there are many researches on gene sequencing technology and clinical decision support system (CDSS) combined with artificial intelligence technology to assist the diagnosis of rare diseases, the diagnosis of rare diseases remains a great challenge in clinical practice. In this paper, we briefly reviewed the CDSS for rare diseases, aiming to analyze the current technique status and challenges of artificial intelligence technology in rare diseases.

Abstract:Difficult inheritance and poor clinical service ability are two major problems limiting the research and development of traditional Chinese medicine (TCM). With the gradual maturity of artificial intelligence technology in the medical field, under the background of national strategy for promoting TCM development, the construction of comprehensive ecological TCM knowledge base and intelligent system will change and improve the traditional inheritance model of TCM and enhance the clinical service ability of TCM. Meanwhile, the formulation of relevant laws and regulations will promote the development of intelligent TCM and further enhance the service ability of TCM. In the program, we aim to develop a novel intelligent TCM system that covers the construction and analysis of TCM knowledge at the grassroots level, so as to provide a new model for the promotion of intelligent TCM in China.

Abstract:In recent years, minimally invasive surgery has rapidly emerged and flourished owing to the advancement of information technology and the incomparable advantages such as small trauma and rapid recovery, and has become the embryo for the development of surgical intelligence. The continuous improvement of medical demands, the advancement of surgical intelligence, and the gradual accumulation of new technology, especially the breakthrough of artificial intelligence technology, eventually led to the birth of intelligent surgery. Intelligent surgery can raise the level of automation in surgical practice to new heights, change the thinking mode of surgeons, and create new surgical models and new medical industries. Intelligent technology will change the service logic of surgeons, so the surgeons must have comprehensive ability.

Abstract:Lung cancer is a malignant tumor with the highest morbidity and mortality, which seriously threatens human health. It is important to improve the diagnosis and treatment efficiency of patients with lung cancer. Artificial intelligence technology provides novel promising strategies for the diagnosis and treatment of patients with lung cancer. Numerous studies have focused on the early screening, diagnosis, treatment and health management of lung tumor, and the development of computer-aided diagnosis system based on deep learning technology, and achieved remarkable results. In this paper, we systematically reviewed the progress of artificial intelligence technology in early screening based medical imaging, pathological diagnosis, prognostic evaluation, surgical navigation and immunotherapy of lung tumors. It is believed that artificial intelligence technology will bring new opportunities for the diagnosis and treatment of lung cancer, and improve the overall survival and quality of life of patients with lung cancer.

Abstract:Artificial intelligence technology based on pathological slice images promotes the development of medicine, and the development of artificial intelligence technology in pathological imaging benefits from the digital whole slide. The digitization of whole slide can provide a large amount of data that can be freely amplified and conveniently labeled, which is conducive to deep learning and clinical application. Digital whole slide is not only applied to human pathology, but also to animal and plant pathology. In this paper, we systematically discussed the role of digital whole slide combined with artificial intelligence technology in pathological recognition, feature extraction, animal models and plant morphology, aiming to provide new clues for the clinical practice of digital pathology.

Abstract:The modernization of traditional Chinese medicine (TCM) diagnosis and treatment technology relies on the development of modern science and technology. Based on the TCM theory of syndrome differentiation and treatment, supported by modern diagnostic technology of TCM, the data-based TCM diagnosis and treatment are informationized and intelligentized with the aid of artificial intelligence technology. Meanwhile, through the research on the diagnosis, treatment, and efficacy evaluation of TCM syndromes and disease, by combining the disease with TCM syndrome on diagnosis-treatment and by exchanging the data of TCM and western medicine, modern diagnostic technology and intelligent diagnosis-treatment system with the characteristics of TCM can be established, which can maximize the advantages of human-computer cooperation. Finally, an intelligent TCM diagnosis and treatment decision-making system with the connotation of syndrome differentiation is established to provide intelligent decision for TCM in clinic and to explore the innovative diagnosis-treatment mode of TCM diseases and syndromes. At the same time, the research of intelligent TCM diagnosis and treatment technology will also promote the rule of TCM diagnosis and treatment, accelerate the leap-forward development of TCM diagnosis and treatment technology, solve the main problems of the modernization of TCM diagnosis and treatment, and promote the modernization of TCM. In this paper, we summarized the current status and trends of the combination of TCM diagnosis and treatment and artificial intelligence technology.

Abstract:The rapid development of deep learning technology provides new methods and ideas for achieving the goal of assisting doctors in high-precision diagnosis. In this paper, we summarized the principles and characteristics of deep learning models that are commonly used in disease diagnosis, including convolutional neural networks, deep belief network, restricted Boltzmann machine and circulation neural network model. Then we introduced the application of deep learning technology in disease diagnosis of several typical diseases, such as lung cancer, breast cancer, and diabetic retinopathy. Finally, we proposed the future of deep learning considering the limitations of deep learning technology in disease diagnosis.

Abstract:Medical imaging technology plays an important role in the detection, diagnosis and treatment of diseases. Due to the instability of human expert experience, machine learning technology is expected to assist researchers and doctors to improve the accuracy of imaging diagnosis and treatment and reduce the imbalance of medical resources. In this paper, we systematically summarized the methods of deep learning technology, introduced the application of deep learning in medical imaging, and discussed the limitations of deep learning technology in medical imaging.

Abstract:Artificial intelligence (AI) technology has developed rapidly with widespread application in recent years. In the medical field, AI technology has made revolutionary advances in medical imaging, pathological diagnosis, health management, drug discovery and surgical navigation. Liver tumors are common diseases in China, and the research and application of AI technology in this field have broad prospect. In this paper, we summarized the research progress of AI technology in imaging and pathological diagnosis, prognosis, treatment and surgical assistance of liver tumor, and prospected the promotion of AI technology in the individualized and precise treatment of liver tumor.

Abstract:The discovery of new drugs is a systematic project with long cycle and low success rate. The traditional drug discovery is to find effective targets related to diseases, and then to screen and design effective small molecules (or large molecules) using various technologies. Artificial intelligence technology has made significant progress in the medical field. In the field of new drug discovery, artificial intelligence technology can integrate a large number of high-dimensional phenotype data, including high-throughput omics data, network pharmacology data and images, so as to effectively screen therapeutic targets and design drugs, saving the costs of drug discovery and shortening the time required for drug discovery. In this article, we explored the drug discovery process driven by new generation of artificial intelligence technology, hoping to provide a reference for the development of new drugs.

Abstract:The rapid development of "Internet Plus" and extensive application of big data technology has laid foundations for the development of artificial intelligence technology. Based on powerful deep learning theory and technology, artificial intelligence technology has made breakthroughs in different areas such as in aiding medical experts answering questions, cutting and classification of medical image of traditional Chinese medicine, and establishing objective four diagnostic methods of traditional Chinese medicine. There is an urgent need to improve overall efficiency in the inheritance and development of traditional Chinese medicine. Artificial intelligence technology has promoted the comprehensive development of traditional Chinese medicine in data mining, intelligence diagnosis and treatment, intelligence learning, and construction of diagnosis and treatment guidelines. How to further improvement in traditional Chinese medicine by artificial intelligence technology is an important issue that needs to be considered.

Abstract:Objective To automatically extract the characteristics of fundus cataract by deep learning, construct a automatic classifier for cataract, and visualize the layer-by-layer feature transformation process of the intermediate layer of deep network.Methods Based on the clinical fundus image, a deep convolutional neural network (CNN) was used to directly learn useful features from the original representation of input data, and then the features extracted by the CNN were compared with pre-defined features. The deconvolution neural network (DN) method was used to quantitatively analyze the characteristics of each intermediate layer of CNN, analyze the pixel sets that have the most contribution to the prediction performance of CNN in the input image, and explore the process in characterizing cataract by CNN.Results The classifier constructed by deep learning achieved an average accuracy of 0.818 6 in four-category tasks. Compared with the existing pre-defined feature set, the feature set automatically extracted by the deep CNN performed better in representing characteristics of cataract. The features of the intermediate layer of CNN hierarchically transformed from low-level abstraction to high-level abstraction, including changed from gradient to edge, then to the combination of edge-like divergent structures, and finally to the high-level abstraction of blood vessel and optic disc information, and this transformation process coincided with the clinical diagnostic criteria of cataract.Conclusion The classifier based on deep learning is superior to the existing classifier in terms of performance. In addition, this method has potential application in detecting other eye diseases.

Abstract:Objective To explore the feasibility and value of neural network combined with micro-hyperspectral imaging in identifying breast cancer tissue.Methods The micro-hyperspectral imaging technology was used to collect image data of breast cancer tissue, and the micro-hyperspectral breast tissue image analysis method based on neural network was used to realize the automatic classification and regional division of breast cancer tissue. Meanwhile, data preprocessing method was proposed to improve the signal to noise ratio of the image, and map information was trained by neural network to identify breast tissue lesions and highlight them for visualization.Results The micro-hyperspectral breast tissue image analysis method based on neural network utilized two characteristics of the images at the same time, and it was better than traditional color pathological images in identifying breast tissue.Conclusion The micro-hyperspectral breast tissue image analysis method based on neural network can provide more characteristic sample information compared with traditional color pathology images, and may serve as an effective complement to traditional color pathological images. With the support of neural network, the micro-hyperspectral imaging technology has prospects in analyzing breast cancer tissue.

Abstract:Objective To propose a scoring framework grid-based motion statistics (SF-GMS) feature matching algorithm to improve the poor real-time ability and inaccurate matching in the process of target recognition for medical service robots.Methods The feature point neighborhoods were segmented by SF-GMS algorithm using the grids, and the number of feature points in each neighborhood was counted and the scoring frame function was set to judge the feature matching accuracy according to the number of neighborhood feature points and the scoring threshold.Results and conclusion Compared with random sample consensus algorithm, SF-GMS algorithm effectively improved the successful matching rate, and had better real-time performance. SF-GMS algorithm had better stability to the changes of illumination view, occlusion, affine, scale and rotation, and could meet the demand of autonomous navigation in simulating hospital ward scenario for medical service robots.

Abstract:Objective To propose a classification method for small sample tongue images based on transfer learning and fully connected neural network, so as to solve the problems of large amount of data, high requirement of training equipment and long training time of deep learning in the classification of tongue images.Methods Effective features such as tongue points and lines of tongue images were extracted by the convolution Inception_v3 network after training on the massive data set of ImageNet. The above features were classified by the fully connected neural network, and the image knowledge acquired by the deep learning network was transferred to the tongue image recognition task, and then the tongue data set were used to train and test the efficiency of the network.Results Compared with the typical tongue image classification method such as K-nearest neighbor (KNN) algorithm, support vector machine (SVM) algorithm and convolutional neural network (CNN) deep learning method, the two methods (Inception_v3+2NN and Inception_v3+3NN) in our experiment had higher classification rates for tongue images, with the accuracy rates being 90.30% and 93.98%, respectively, and had shorter training time for the sample.Conclusion Compared with KNN algorithm, SVM algorithm and CNN deep learning method, the tongue image classification method based on transfer learning and fully connected neural network can effectively improve the accuracy rate of tongue image classification and shorten the training time.

Abstract:Objective To recognize cancer regions by using segmentation algorithm for pathological slices of gastric cancer based on deep learning.Methods The U-net network was used as the basic structure to design a deeper segmentation algorithm deeper U-Net (DU-Net) for gastric cancer pathological slices. The datasets were segmented into several small blocks by the region overlapping segmentation method. Then the blocks were firstly segmented by the pre-trained DU-Net model, and the new samples were re-synthesized using the image classifier to remove false positive samples. The new samples were repeatedly trained by repeated learning methods, and the results of segmentation were processed with fully connected conditional random field (CRF). Finally, the segmentation pictures of gastric cancer were obtained and validated.Results After 3 times of repeated learning, the mean accuracy of the DU-Net model for pathological slices of gastric cancer was 91.5%, and the mean intersection over union coefficient (IoU) was 88.4%. Compared with the basic DU-Net model without repeated learning, the mean accuracy and mean IoU of the DU-Net network were increased by 2.9% and 5.6%, respectively.Conclusion The segmentation algorithm for pathological slices of gastric cancer based on deep learning can accurately recognize cancer regions, improve the generalization ability and robustness of the model, and can be used for computer-assisted diagnosis of gastric cancer.

Abstract:Objective To evaluate the bone age of children using deep convolutional neural network based on feature extraction combined with key features and demographic information.Methods Left hand X-ray images were automatically recognized and preprocessed, and then the 17 key region features of bone age in the left hand joint were automatically extracted by X-ray image analysis method based on deep convolutional neural network. The image features of bone age were combined with clinical data (population statistics and gender) to train and test the bone age assessment model.Results The feature region extraction method based on deep learning had better efficiency in extracting feature information than traditional image analysis method, and the feature information combined with clinical information supplemented the information of bone age from another dimension. The average absolute error measured by bone age assessment model based on multi-dimensional data feature fusion was 0.455, which was better than traditional methods and only end-to-end deep learning method.Conclusion Compared with traditional machine learning methods, the deep convolutional neural network based on feature extraction has better performance, and can improve the predicting accuracy of image-based bone age by combining with population information such as gender and age.

Abstract:Based on artificial intelligence technology, the intelligent medical image recognition refers to the analysis and process of medical images scanned by medical imaging technologies such as X-ray films, computed tomography and magnetic resonance imaging, and surgical video. Major trends in intelligent medical image recognition include intelligent image diagnosis, three-dimensional reconstruction and registration, intelligent surgery video parsing and so on. Intelligent image diagnosis and three-dimensional reconstruction and registration can improve the efficiency and quality of image recognition, and provide a helpful method for clinical diagnosis and treatment; intelligent surgery video parsing can help surgeons learn and understand surgical procedures, and further guide the operation process. Now the research of intelligent medical image recognition has gained some theoretical and technological achievement and gradually been applied in clinic. In this paper, we summarized the progress of intelligent medical image recognition and put forward the development prospect in this field.

Abstract:Presently morphological evaluation and special staining scoring system are important components of basic and clinical research, and are very important for judging the efficacy of drugs and gene intervention. However, the current visual scoring system has some disadvantages such as strong subjectivity, poor repeatability and low accuracy, and is prone to missed diagnosis and misdiagnosis. Artificial intelligence technology based on deep learning is expected to overcome these problems. In our study, we found that the convolutional neural network can be used to accurately extract internal features related to the treatment and prognosis of tumors, such as tumor-stroma ratio, nerve invasion and spatial distribution of lymphatic cells in tumor specimens, visualizing and digitalizing the curative effect of drug intervention on disease progression, and can quantify and automatic evaluate the expression of molecular biomarkers related to clinical treatment, classification and prognosis. The application of artificial intelligence technology in tissue and cell morphology assessment will promote the consistency, repeatability and accuracy of clinical drug evaluation and basic scientific research evaluation, and is expected to further promote the development of medical research.

Abstract:Artificial intelligence technology has made breakthroughs in the field of clinical medicine, including diagnosis, imaging, and disease classification. Electronic medical record contains a large number of clinical data such as disease description, diagnosis, examination and treatment. With the participation of medical experts and information scientists, the studies of data mining of electronic medical record using artificial intelligence technology have greatly increased. Although now the method has some limitations, it is more rapid, economic and convenient compared with the traditional method, and is expected to promote the development of human health. In this paper, we reviewed the current status of data mining of electronic medical record using artificial intelligence technology, regarding related technologies, specific examples, and limitations.

Abstract:In recent years, artificial intelligence technology in medical field has become a research focus of modern science and technology. The application of artificial intelligence technology in the diagnosis of dizziness can not only save medical resources, but also treat dizziness in time. In this paper, we analyzed the application of artificial intelligence technology in the field of vertigo diagnosis by illuminating the expert systems for vertigo disease such as "Vertigo" and "ONE", and other methods, summarized the advantages and disadvantages of various artificial intelligence methods applied in vertigo disease, and prospected the development prospect of artificial intelligence technology in vertigo diagnosis system.

Abstract:Artificial intelligence technology is widely used in medical fields such as disease diagnosis, pathological analysis, and development of new drugs.Combining artificial intelligence technology with nursing profession, and formulating corresponding expert system, decision support system and intelligent assistant design will improve the comprehensive benefit of economy and society. In this paper, we summarized the research and application status of artificial intelligence technology in various aspects of nursing profession, such as clinical nursing, nursing education and extended nursing, hoping to provide reference for further research in the future.