Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

< Back to Article

Table 1.

Comparative Analysis of Baseline Data.

More »

Table 1 Expand

Fig 1.

Evaluation and Comparative Analysis of Various Machine Learning Models.

(a: The performance of six machine learning models in predicting clinical remission of CD patients treated with adalimumab for 12 weeks was evaluated. b: The performance of six machine learning models in predicting clinical remission of CD patients treated with adalimumab for 48 weeks was evaluated.).

More »

Fig 1 Expand

Fig 2.

SHAP Analysis of the 12 Weeks Prediction Results.

(In the Summary Plot and Feature Importance Donut Chart, features such as Fc and CRP exhibit significant influence. The average Shapley value for Fc is notably high, indicating a strong negative contribution to the prediction results. Specifically, lower Fc values are associated with a higher likelihood of achieving 12 weeks clinical remission, underscoring its critical role in immune regulation or drug action mechanisms. The Force Plot analysis provides an intuitive visualization of each feature’s impact on individual sample predictions. The positive or negative sign and magnitude of the Shapley values for each feature reflect their respective promoting or inhibiting effects on clinical remission predictions. Multiple features interact synergistically to collectively determine the final prediction outcome.).

More »

Fig 2 Expand

Fig 3.

SHAP Analysis of the 48 Weeks Prediction Results.

(In the Summary Plot and Feature Importance Donut Chart, Fc remains a significant factor. Hemoglobin (Hb) has become increasingly important, with its average Shapley value rising, likely due to anemia’s growing impact on treatment efficacy as the disease progresses. Erythrocyte sedimentation rate (ESR), platelets (plt), and C-reactive protein (CRP) also play crucial roles in predicting 48 weeks clinical remission. ESR reflects inflammation levels, while plt influences immune and coagulation processes. The Force Plot analysis visually demonstrates how different feature values combine to affect predictions. Changes in key features like Hb and ESR can shift prediction results toward or away from clinical remission.).

More »

Fig 3 Expand