基于AI的HAE罕见病风险预测方法及装置

• Format: Patent
• Language: Chinese
• Published: 25.02.2025
• 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

本发明涉及基于AI的HAE罕见病风险预测方法及装置，包括：通过预处理后的医疗数据对开源Llama3基座模型进行训练微调，提取HAE疾病医疗知识，作为训练数据输入微调后的开源Llama3基座模型，以输出第一风险概率分布。将训练数据作为TinyBERT模型的输入，以输出HAE疾病医疗知识对应的第二风险概率分布，调用交叉熵损失函数计算第一风险概率分布与第二风险概率分布的差异。通过反向传播算法计算差异的梯度，使用优化器更新模型参数，对TinyBERT模型进行迭代训练和微调，得到HAE罕见病风险预测模型。将当前患者预处理后的病历数据作为HAE罕见病风险预测模型的输入，输出病历数据对应的风险预测评分。 The invention relates to an AI-based HAE rare disease risk prediction method and device, and the method comprises the steps: carrying out the training and fine adjustment of an open-source Llama3 pedestal model through preprocessed medical data, extracting the medical knowledge of HAE diseases, and inputting the medical knowledge of HAE diseases as training data into the finely-adjusted open-source Llama3 pedestal model, so as to output first risk probability distribution. And taking the training data as the input of a TinyBERT model to output second risk probability distribution corresponding to the HAE disease medical knowledge, and calling a cross entropy loss function to calculate the difference between the firs