các textbox nên làm mờ đi

các textbox nên làm mờ đi

Kết quả (Anh) 1:

the textbox should be dimmed

Kết quả (Anh) 2:

The textbox should be blurred.

Kết quả (Anh) 3:

Results for different sensors number configurations and for the additional reference accelerometer are shown in Fig. 3. We first report results when no reference accelerometer on the back was used. Mean sensitivity and PPV for the case of one single sensor were 0.51 and 0.51, with the exception of sensor 6 (placed on the back) that resulted in sensitivity 0.0, PVV 0.0, highlighting how this placement is optimal to detect maternal movement instead of fetal movement. Sensitivity and PPV for the case of two sensors were 0.63 and 0.54 respectively, while for three sensors sensitivity was 0.69 and PPV was 0.57. When using four sensors mean sensitivity was 0.70 and PPV was 0.58. Finally, using all five sensors together provided the same performance (sensitivity 0.70, PPV 0.58). Including a reference accelerometer consistently improved detection performance, as shown in Fig. 3. In particular, when adding the reference accelerometer to a single sensor, we obtained sensitivity 0.57 and PPV 0.56. When adding the reference accelerometer to a two sensors system, we obtained sensitivity 0.68 and PPV 0.61. Results improved marginally when moving to three sensors, with sensitivity of 0.70 and specificity of 0.63 and more consistently when moving to four sensors, with sensitivity of 0.75 and PPV 0.65. All sensors did not improve results compared to the four sensors case. 0.8Single sensor Reference accelerometerAbsent Present 0.70.60.50.40.3 1 1_6 2 2_6 3 3_6 4 4_6 5 5_6Sensors IDs 0.80.70.60.50.40.3Combinations of 2 sensors 1_3 1_3_61_4 1_4_63_4 3_4_6 4_5Sensors IDs4_5_62_3 2_3_6 0.80.70.60.50.40.3Combinations of 3 sensors 1_2_31_2_3_61_3_41_3_4_61_3_51_3_5_63_4_53_4_5_62_3_42_3_4_6Sensors IDs 0.80.70.60.50.40.3Combinations of 4 sensors1_2_3_4 1_2_3_4_6 1_3_4_5 1_3_4_5_6 2_3_4_5 2_3_4_5_6 1_2_3_5 1_2_3_5_6 1_2_3_4 1_2_3_4_6Sensors IDs Fig. 3.when including a reference accelerometers on the back are shown in darker gray. Only the five best combinations are listed for clarity.accelerometer on the same dataset allows us to make mean- ingful comparisons and determine performance differences in detecting fetal kicks under different conditions. Future work will investigate the possibility to explore different machine learning tools to add temporal dependency to consecutive times as well as include different types of fetal movements and input signals.Ref