Methods PCA - T2 PCA - SPE PCA - KD ICA - I ICA - I 2 d 2 e ICA - SPE ICA - KD ADR 99.00 98.00 100.00 99.00 84.00 78.00 100.00 sitting posture due to high pressure being applied on the right side of the human body. The detection scores (i .e. , ADR, FAR and F1-score) in identifying this posture are listed in Table 1. We observe that both PCA and ICAbased schemes can detect this wrong posture with a good detection rate. In this scenario, the proposed ICA-KD approach may arguably be the best detector by correctly detecting this abnormal posture sitting without false alarm. The ICAI scheme followed it 2 d by achieving a high ADR of 99 and FAR of 1. Results for the first scenario demonstrate the superiority of ICA-KD compared to the other methods. Scenario 2: Sitting Posture with Higher Pressure on the Left Side: Here, the aim is to evaluate the capacity of the considered schemes to sense improper sitting posture with high pressure on the left side of the human body. The detection results of the seven procedures are reported in Table 1. In this scenario, the PCA-SPE, PCA-KD, ICA - I , ICA - I , and ICA - SPE2 d 2 e based monitoring schemes detect this improper sitting June 2023 Fig. 4. Detection results under position 3 sitting posture: (a) PCA - T2; (b) PCA-SPE;(c) PCA-KD; (d) ICA- 2 ICA - SPE; and (g) ICA-KD. d IEEE Instrumentation & Measurement Magazine I ; (e) ICA- 2 e I ; (f) 41 Table 1 - Detection results of the investigated schemes for the three scenarios Scenario 1 FAR 1.43 5.71 0.00 1.00 F1-score 99.50 97.50 98.90 99.50 7.00 1.12 0.00 87.96 88.15 100.00 ADR 95.00 60.00 70.00 70.00 90.00 85.00 100.00 Scenario 2 FAR 1.43 5.71 0.00 0.00 0.00 0.00 0.00 F1-score 95.00 66.67 82.35 82.35 94.74 91.89 100.00 ADR 74.00 8.25 24.24 80.00 71.12 91.00 98.25 Scenario 3 FAR 1.43 7.14 0.00 0.00 3.00 0.00 0.00 F1-score 84.57 15.86 38.71 88.80 80.61 95.29 98.98