A Novel Multi-view Bi-clustering method for identifying abnormal Co-occurrence medical visit behaviors.

Abnormal co-occurrence medical visit behavior is a form of medical insurance fraud. Specifically, an organized gang of fraudsters hold multiple medical insurance cards and purchase similar drugs frequently at the same time and the same location in order to siphon off medical insurance funds. Conventional identification methods to identify such behaviors rely mainly on manual auditing, making it difficult to satisfy the needs of identifying the small number of fraudulent behaviors in the large-scale medical data. On the other hand, the existing single-view bi-clustering algorithms only consider the features of the time-location dimension while neglecting the similarities in prescriptions and neglecting the fact that fraudsters may belong to multiple gangs. Therefore, in this paper, we present a multi-view bi-clustering method for identifying abnormal co-occurrence medical visit behavioral patterns, which performs cluster analysis simultaneously on the large-scale, complex and diverse visiting record dimension and prescription dimension to identify bi-clusters with similar time-location features. The proposed method constructs a matrix view of patients and visit records as well as a matrix view of patients and prescriptions, while decomposing multiple data matrices into sparse row and column vectors to obtain a consistent patient population across views. Subsequently the proposed method identifies the corresponding abnormal co-occurrence medical visit behavior and may greatly facilitate the safe operations and the sustainability of medical insurance funds. The experimental results show that our proposed method leads to more efficient and more accurate identifications of abnormal co-occurrence medical visit behavior, demonstrating its high efficiency and effectiveness.

Heterogeneous Information Network-Based Patient Similarity Search.

Patient similarity search is a fundamental and important task in artificial intelligence-assisted medicine service, which is beneficial to medical diagnosis, such as making accurate predictions for similar diseases and recommending personalized treatment plans. Existing patient similarity search methods retrieve medical events associated with patients from Electronic Health Record (EHR) data and map them to vectors. The similarity between patients is expressed by calculating the similarity or dissimilarity between the corresponding vectors of medical events, thereby completing the patient similarity measurement. However, the obtained vectors tend to be high dimensional and sparse, which makes it hard to calculate patient similarity accurately. In addition, most of existing methods cannot capture the time information in the EHR, which is not conducive to analyzing the influence of time factors on patient similarity search. To solve these problems, we propose a patient similarity search method based on a heterogeneous information network. On the one hand, the proposed method uses a heterogeneous information network to connect patients, diseases, and drugs, which solves the problem of vector representation of mixed information related to patients, diseases, and drugs. Meanwhile, our method measures the similarity between patients by calculating the similarity between nodes in the heterogeneous information network. In this way, the challenges caused by high-dimensional and sparse vectors can be addressed. On the other hand, the proposed method solves the problem of inaccurate patient similarity search caused by the lack of use of time information in the patient similarity measurement process by encoding time information into an annotated heterogeneous information network. Experiments show that our method is better than the compared baseline methods.