Design And Implementation Of An Electronic Medical Record Assistant Editing System Based On Voice Recognition

With the continuous application of information science and technology in the field of medicine and the support of related policies,The technical level and application scope of electronic medical records have been developed by leaps and bounds;However,in recent years,there has been no breakthrough in the editing mode of electronic medical records,However,in recent years,there has been no breakthrough in the editing mode of electronic medical records,which is still in the mode of manual input.As an important legal basis for the settlement of medical disputes,the legal effect of electronic medical records has also been questioned.At the same time,artificial intelligence technology is developing rapidly,voice recognition technology is becoming matures,the current electronic medical record system does not make full use of these achievements.Based on these problems at above,this paper investigates the current situation and problems of electronic medical record system both at home and abroad,According to the actual demand,an electronic medical record assistant editing system based on identification and recognition is designed and implemented.The main contents are as follows:(1)Based on mature application of speech recognition technology,this paper puts forward to collect voice of the whole process of diagnosis,treatment and surgery and convert it into text,and then transfer the voice and text to the electronic medical record system,The medical staff members will be able to write the medical records by pasting the corresponding text and making some changes when they use the electronic medical records assisted editing system designed by this paper,which realized the semi-automation of medical record editing.This will greatly reduce the medical staff's workload of editing medical records and the medical records editing errors caused by the medical staff because of forgetting,to a certain extent.(2)For medical words that may be identified as errors,smart error correction is performed.A method of positioning medical vocabulary based on full Pinyin using BLSTM model is proposed.(3)Add a trusted time stamp,which is based on the ANSI X9.95 protocol,to the recording files in time.The ANSI X9.95 standard protocol is extended on the RFC3161 time stamp protocol which is widely used in China,which have higher security.It is shown that the recording file is present truly before the time indicated by the timestamp,in this way.This make it capable of un-Forged,Non-repudiation,restoring truth and verifying authenticity,which will preserve credible evidence for possible medical disputes.In this paper,the problems of applicability,legality and standardization of evidence collection in electronic medical record system and the low efficiency in editing electronic medical records are solved.It has high research significance and practical application value.