摘要
为了提高高强度聚焦超声(HIFU)治疗过程中生物组织变性识别率,提出了一种基于改进鲸鱼优化算法优化概率神经网络(IWOA-PNN)模型的生物组织变性识别方法。首先通过改进收敛因子和加入自适应权重因子提高WOA优化算法的寻优速度和精度,然后利用IWOA算法优化PNN的平滑因子,以提高变性识别精度,最后以超声回波信号多尺度熵为特征参数输入IWOA-PNN模型,得出生物组织变性识别率。实验结果表明,与普通PNN和WOA-PNN模型相比,基于IWOA-PNN模型的生物组织变性识别率更高,更能精确地识别HIFU治疗过程中生物组织是否变性,指导临床医生进行准确的HIFU疗效评价。
To improve the identification rate of denatured biological tissue during high intensity focused ultrasound(HIFU),based on the improved whale optimization algorithm(IWOA),a probabilistic neural network(PNN)identification model of denatured biological tissue was established.By improving convergence factor and adding adaptive weighted factor,the speed and accuracy of WOA optimization algorithm are improved,and the smoothing factor of PNN is optimized by IWOA algorithm to improve the denatured identification accuracy.The identification rate of denatured biological tissue is then obtained by putting multi-scale entropy of ultrasonic echo signal as the characteristic parameter into IWOA-PNN model.Compared with common PNN and WOA-PNN models,IWOA-PNN model has higher accuracy,and can more accurately identify whether biological tissues are denatured during HIFU treatment,which will guide clinicians to accurately evaluate the efficacy of HIFU.
作者
曹菁
贺绍相
陈光强
杨江河
刘备
彭梓齐
CAO Jing;HE Shaoxiang;CHEN Guangqiang;YANG Jianghe;LIU Bei;PENG Ziqi(College of Mathematics and Physics,Hunan University of Arts and Science,Changde 415000,China)
出处
《湖南文理学院学报（自然科学版）》
CAS
2024年第3期24-29,共6页
Journal of Hunan University of Arts and Science(Science and Technology)
基金
国家自然科学基金(U2031112)
湖南省自然科学基金青年项目(2020JJ5396)
湖南省教育厅科学研究项目优秀青年项目(20B405)
湖南文理学院博士科研启动项目(20BSQD06)。
关键词
高强度聚焦超声
生物组织
变性识别
改进鲸鱼优化算法
概率神经网络
HIFU
biological tissues
denatured identification
improved whale optimization algorithm
probabilistic neural network
