Converter-based data privacy protection system

• Main Authors: LONG HAIXIA, CHEN ZHEN, LIU SHENGQUAN, HUANG JIA
• Format: Patent
• Language: Chinese; English
• Published: 22.12.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATIONTECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING ORPROCESSING OF MEDICAL OR HEALTHCARE DATA; INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTEDFOR SPECIFIC APPLICATION FIELDS; PHYSICS

The invention provides a data privacy protection system based on a converter, the system comprises a model training architecture and a data processing architecture, the model training architecture is used for training an effective data desensitization model and an anomaly detection model, and the data processing architecture is used for desensitizing electrocardiogram data based on the data desensitization model. The desensitized electrocardiogram data is subjected to anomaly detection based on the anomaly detection model, the credibility of the anomaly detection result is explained through the interpretable artificial intelligence module, the system can help a user better understand and trustworthy the anomaly detection result output by the model, and the accuracy and error of the system are superior to those of other existing models. 本发明提供一种基于转换器的数据隐私保护系统，所述系统包括模型训练架构和数据处理架构，所述模型训练架构用于训练出有效的数据脱敏模型和异常检测模型，所述数据处理架构用于基于数据脱敏模型对心电图数据进行脱敏处理，并基于异常检测模型对脱敏后的心电图数据进行异常检测，通过可解释人工智能模块对异常检测结果的可信性进行解释，所述系统能够帮助用户更好地理解和信赖模型