Fig 1.
Comparison of recognition results with conventional affinity and graph affinity method.
A: conventional affinity method. B: graph-based affinity methods.
Fig 2.
Overview of the proposed weakly supervised semantic segmentation method.
Table 1.
The annotated classification of the lung cavity size and quantity in a patient’s CT scan across three datasets. The “Class” column indicates the numerical index corresponding to each classification category.
Table 2.
Patient-level data distribution under 5-fold cross-validation.
For each dataset, a fixed test set is held out and excluded from cross-validation. Five-fold CV is performed only on the training pool at the patient level.
Table 3.
Statistical evaluation protocol for segmentation performance.
Fig 3.
Impact of parameter variations and module ablation on lung cavity segmentation under weak supervision.
Table 4.
Design of ablation study modules with different loss components used in our weakly supervised method.
Table 5.
The contribution of various ablation components to the lung cavity semantic segmentation task in the proposed weakly supervised method.
Fig 4.
Performance of conventional affinity and our graph affinity method.
A: conventional affinity method. B: graph-based affinity methods.
Fig 5.
Recognition results of class activation mapping for M-1 to M-5 ablation methods.
Fig 6.
Training and validation losses of methods M-1 to M-5 and ROC curve performance for lung cavity attribute classification.
Fig 7.
Confusion matrices for M-1 to M-5 ablation classification experiments.
Table 6.
Benchmark configuration and fairness comparison across weakly supervised semantic segmentation methods.
All methods are re-trained under identical settings with a patch size of and optimized using AdamW.
Table 7.
Quantitative results of various 3D medical weakly supervised (lung cavity) semantic segmentation methods.
Fig 8.
Semantic segmentation results of lung cavities using various methods.
C1 - C7 sequentially represent the methods from [36], [19], [37], [15], [24], [38], and [39].
Table 8.
Subgroup analysis by cavity size and quantity.
Average DSC (%) across different lesion subgroups.
Table 9.
Qualitative performance analysis by lesion location.
Table 10.
Subgroup analysis by cavity morphology.