Yu Zhao - research works

The five traditional ML models showed varying performance, with RF demonstrating the highest average AUC in the training set (AUC = 0.795, 95% CI 0.709-0.881).

GGT_3M, ALB_3M, DBIL_3M, DBIL_1M and GLO_1W were identified as the most critical predictive features.

A TL-based clinical decision support system was deployed, providing dynamic visualization of individualized 24-month NLS trajectories.

During the 12-month follow-up, 42 children (14.5%) progressed to TA.

Hosmer-Lemeshow tests indicated good fit (training:χ 2 =6.7277, p = 0.5663; validation:χ 2 =7.7376, p = 0.4595).

DCA demonstrated favorable clinical net benefit across a wide range of threshold probabilities.

There were no significant differences in baseline characteristics (age, gender, BMI, surgical parameters, scoliosis type, preoperative/postoperative Cobb angles) between the two groups (all p > 0.05).

Preoperatively, SRS-22 and SF-36 scores showed no inter-group differences (all p > 0.05).

Both groups showed significant improvements in physical function-related dimensions postoperatively.

The Non-Negative Emotion Group also exhibited significant improvements in SRS-22 self-image/pain and SF-36 bodily pain (all p < 0.05), while the Negative Emotion Group showed no significant improvements in these dimensions.