人工神经网络和支持向量机性能比较及其在DMD疾病识别中的应用

对人工神经网络(ANN)和支持向量机(SVM)这两种机器学习方法进行了分析与比较,并分别利用两种算法对神经肌肉罕见病DMD的磁共振图像(MRI)数据进行建模、分类预测.经对比后得出结论:两种算法结果均表明,DMD的两类MRI(T1和T2)中,T1更具特征性,故此类患者的MRI检查可仅进行T1扫描;若能选择合适的模型参数,则两种算法模型均具有极好的分类预测效果,其灵敏度、特异度和准确率分别高达98.5%,97.3%,97.9%和96.9%,97.3%,97.1%;利用机器学习方法对DMD患者的MRI进行分析处理可作为该病无创检测的技术探索,有望为临床提供客观有效的辅助诊断手段.

Comparative Study on the Performances of ANN and SVM and Their Application in the Identification of DMD Disease

Based on analyzing and comparatively studying the two algorithms of machine learning,namely,the artificial neural network (ANN) and support vector machine (SVM),the modeling,data classification and prediction for magnetic resonance images (MRIs) of the neuromuscular disease,(duchenne musular dystrophy,DMD),were carried out.The conclusions of the study are as follows.The outcomes of the two algorithms indicate that between the two kinds of DMD MRI images(T1 and T2),the T1 image has clearer texture feature.Therefore,patients could just need T1 scanning for MRI examination.If the model parameters are selected appropriately,the two algorithm models could produce very excellent classification prediction outcomes respectively.The sensitivity,specificity and accuracy rate might reach as high as 98.5%,97.3%,97.9% and 96.9%,97.3% and 97.1% respectively.The machine learning method could be used as a non-invasive detection technology in the treatment of DMD MRI images and it is expected to provide an objective and effective diagnostic method for clinical.