In the medical field,it is very important for doctors to make effective and correct decision-making.In order to improve the accuracy of doctors' diagnosis and avoid the misdiagnosis of doctors' intuition,subconscious and incomplete knowledge.ABC-NB algorithm is used in the field of chronic disease diagnosis to improve the diagnostic efficiency and reduce the chance of misjudgment.The artificial bee colony algorithm based on improved scale factor is applied to the selection of chronic disease characteristics,and the data are dimensioned,the redundant and irrelevant features are removed,the convergence speed is improved,and the algorithm is applied to search the global optimal solution.Then,the eigenvalues of the pre-processed data are trained and learned to generate the Bayesian classifier to construct the prediction model.The prediction module displays the diagnostic results for medical staff to assist in the diagnosis and decision making.Experiments show that the model has good flexibility and robustness,can have a stable calculation of the probability of diagnosis of chronic diseases,and it is effective for the diagnosis of medical staff.%在医疗领域,医生做出有效正确的决策非常重要,为了提高医生诊断的准确性,避免诊断结果受到医生的直觉、潜意识和自身知识不全面等因素的干扰而造成误判;提出了将改进的ABC-NB算法应用于慢性病诊断领域,以提高诊断效率,减少误判几率;将基于改进尺度因子的人工蜂群算法应用于慢性病特征的选择,对数据进行降维,剔除冗余、无关的特征,提高收敛速度,增强算法搜索全局最优解的能力;接着将预处理后的数据各特征值进行训练和学习生成贝叶斯分类器,构建预测模型;预测模块将诊断结果显示出来供医护人员参考,辅助进行诊断和决策;实验表明该模型具有很好的柔性和鲁棒性,能够稳定的计算出慢性病的概率,有效的辅助医护人员进行诊断.
展开▼
