Page 99 - 南京医科大学自然版
P. 99

第44卷第12期           丁晨琛,刘卫国. 应用多点可穿戴设备定量评估早期帕金森病运动障碍特征研究[J].
                 2024年12月                  南京医科大学学报(自然科学版),2024,44(12):1708-1714,1722                   ·1713 ·


                    表3 上半身、下半身和全身运动模型的诊断性能                                Chance                    ±1 std.dev.
                Table 3  Diagnostic performance of upper body,lower       PD mean ROC(AUC=0.916±0.134)  ±1 std.dev.
                        bodys and whole body motion models                Chance                    ±1 std.dev.
                                                                          PD mean ROC(AUC=0.939±0.107)
                                 Upper body  Lower body  Whole body       Chance
                   Performance                                            PD mean ROC(AUC=0.966±0.067)
                                 movements  movements  movements
                                                                          1.0
                Kappa            0.64 ± 0.29  0.89 ± 0.18  0.92 ± 0.18
                Area under the curve  0.94 ± 0.11  0.92 ± 0.14  0.97 ± 0.07
                                                                          0.8
                Specificity      0.85 ± 0.20  0.97 ± 0.11  0.95 ± 0.16
                                                                         1-Sensitivity
                Sensitivity      0.80 ± 0.28  0.93 ± 0.14  0.97 ± 0.11    0.6
                Yoden index      0.65 ± 0.28  0.90 ± 0.16  0.92 ± 0.18
                Accuracy         0.81 ± 0.16  0.94 ± 0.09  0.96 ± 0.08    0.4

                                                                          0.2
                3  讨 论
                                                                           0
                    本研究采用多点可穿戴设备采集早期 PD 患者                                  00   0.2   0.4  0.6  0.8   1.0
                                                                                      1-Specificity
                与正常对照组完成 UPDRSⅢ量表指定动作时的运
                                                                      The solid part represented of the mean,while the shaded area indi⁃
                动参数显著差异的基础上,进一步采用交叉验证的                            cated the range of ± one standard deviation. The standard deviation of
                方法评估上述运动参数在辅助识别早期PD 患者中                           the upper body movement model and the lower body movement model
                的诊断价值。结果显示,言语动作、行走与站立运                            were greater than that of the whole body movement model.
                                                                  图 3  上半身动作、下半身动作与全身动作最佳模型的
                动信息的下半身动作模型识别早期PD 患者的准确
                                                                       ROC曲线
                性较高,有望在受试人群中更准确地识别早期 PD                           Figure 3  ROC curves for the best models for upper body
                患者。多点可穿戴设备可客观有效地评估早期 PD                                    movements,lower body movements,and whole
                患者运动障碍特征,克服传统量表评估中主观偏倚                                     body movements

                                              表4 指定动作触发的接受信号的传感器数量
                                Table 4 The number of sensors that receive signals triggered by specified actions

                                                                     1    2    3   4   5   6    7   8    9   10
                Experiment Specified
                                           Full name speech         Waist Left Right Left Right Left Right Left Right Chest
                number    actions
                                                                         thigh thigh calf calf foot foot hand hand
                01         SP              Left finger tapping       √    √   √   √   √    √   √    √   √    √
                02         FT              Right finger tapping                                     √
                02         FT             Left hand movements                                           √
                03         HM             Right hand movements                                      √
                03         HM           Postural tremor of the hands                                    √
                04        PTOH   Left pronation⁃supination movements of hands                       √   √
                05        PSOH   Right pronation⁃supination movements of hands                      √
                05        PSOH         Left kinetic tremor of the hands                                 √
                06        KTOH        Right kinetic tremor of the hands                             √
                06        KTOH              Left toe⁃tapping                                            √
                07         TT               Right toe⁃tapping             √       √
                07         TT                Left leg agility                 √       √
                08         LA               Right leg agility             √       √
                08         LA              Arising rrom chair                 √       √
                09         AFC                 Posture               √    √   √   √   √    √   √    √   √    √
                10         POS                 Posture               √    √   √   √   √    √   √    √   √    √
                11         PS                  Stability             √    √   √   √   √    √   √    √   √    √
                12         Gait                 Gait                 √    √   √   √   √    √   √    √   √    √
   94   95   96   97   98   99   100   101   102   103   104