Fig 1.
The procedure of the study.
Table 1.
Baseline characteristics.
Table 2.
Analysis of the demographic characteristics.
Table 3.
Baseline characteristics and laboratory examinations of two groups.
Table 4.
Nonnormal data of two groups.
Table 5.
The multiple logistic regression model.
Fig 2.
The optimized multiple logistic regression model (backward stepwise regression).
(The odds ratio (OR) > 1.00, indicating that the variable is associated with an increased risk of poorly differentiated LUAD. AISI:Aggregate Index of Systemic Inflammation;CEA: Carcinoembryonic Antigen;ProGRP: Progastrin-Releasing Peptide).
Fig 3.
The nomogram model to predict low differentiation in lung cancer.
(Using the nomogram, clinicians can derive a score based on individual patient characteristics. This total score is then translated into a probability of risk of poorly differentiated LUAD, facilitating stratified management and personalized interventions for higher risk patients. AISI: Aggregate Index of Systemic Inflammation; CEA: Carcinoembryonic Antigen; ProGRP: Progastrin-Releasing Peptide).
Fig 4.
The AUC of the nomogram model.
( (AUC is 0.795 (95% CI: 0.726 - 0.864), indicateing that the model possesses a relatively good discriminatory ability and accuracy).
Fig 5.
The calibration curve of the nomogram model.
(The calibration curve was generated using 1,000 bootstrap repetitions. The diagonal dashed line represents the ideal case of perfect prediction, while the solid line indicates the actual performance of our model. Closer agreement between the two lines signifies better predictive accuracy).
Fig 6.
The decision curve of the nomogram model.
(The nomogram offered a greater net benefit compared to both the “treat-all” and “treat-none” approaches over a threshold probability range from 30% to 95%, confirming its clinical utility in routine practice).