×
  • Select the area you would like to search.
  • ACTIVE INVESTIGATIONS Search for current projects using the investigator's name, institution, or keywords.
  • EXPERTS KNOWLEDGE BASE Enter keywords to search a list of questions and answers received and processed by the ADNI team.
  • ADNI PDFS Search any ADNI publication pdf by author, keyword, or PMID. Use an asterisk only to view all pdfs.
Principal Investigator  
Principal Investigator's Name: Chengli Ni
Institution: Beijing Jiaotong University
Department: School of Economics and Management
Country:
Proposed Analysis: Completed undergraduate thesis Title:Research and implementation of disease diagnosis method based on multimodal information fusion This thesis mainly focuses on how to realize a widely adaptable disease intelligent diagnosis model based on multi-modal information fusion. On the basis of previous research, this paper studies the data preprocessing framework, including data imputation and characteristics Extraction, feature selection, etc. This paper focuses on the feature fusion method of multi-modal data and the disease diagnosis method based on multi-modal data fusion features. It fuses the laboratory test results, chief complaints, present medical history, past medical history and other multi-modal diagnostic treatment information in the electronic medical record, and realizes an intelligent diagnosis model that can adapt to any disease diagnosis purpose and only needs to be trained on special data sets. And in the end of this paper proposed a detailed example analysis and comparative analysis of the method. The main contents of this study are as follows: (1) Construct a word embedding model for Chinese electronic medical records combined with multi-modal information This paper reviews the research of domestic and foreign scholars on the word embedding model of Chinese electronic medical record and the data processing model of multi-modal information, and defines the existing common word embedding model system in the medical field. On this basis, through the research of literature and the consultation of experts, the construction method of the multi-modal information preprocessing framework suitable for most diseases is determined. So as to provide support for the construction of disease intelligent diagnosis model based on multi-modal information fusion. (2) An intelligent disease diagnosis model based on multi-modal information fusion is constructed In this study, feature fusion methods and classifiers are combined to construct an intelligent disease diagnosis model with versatility and universality. By improving the shortcomings of the existing disease diagnosis model, and proposing the improved feature fusion method and the multi-modal classifier model, the universal framework of disease intelligent diagnosis model based on multi-modal information fusion is established. It allows users to train models using new datasets within the framework to provide medical diagnosis services for specific diseases. (3) A detailed example analysis and comparative analysis are carried out to prove the superiority of the model This paper reviews the related research of domestic and foreign scholars and summarizes the common comparison methods of disease prediction research. A representative disease is selected for example analysis, and compared with other disease auxiliary diagnosis methods to analyze the superiority of the method in this paper.
Additional Investigators