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 AI methods used by recent studies for the classification of LGDs using different medical imaging.

More »

Table 1 Expand

Fig 1.

Study flow diagram for the classification of LGDs using CT scan.

More »

Fig 1 Expand

Table 2.

A comprehensive summary of LGDs datasets.

More »

Table 2 Expand

Fig 2.

Sample CT scan images of LGDs used for training and testing the proposed LGD_Net model and baseline models.

More »

Fig 2 Expand

Table 3.

A comprehensive summary of LGDs datasets after applying BL-SMT.

More »

Table 3 Expand

Fig 3.

Synthetic LGD image generated after applying BL-SMT.

More »

Fig 3 Expand

Table 4.

Comparison of the proposed LGD_Net with ANN.

More »

Table 4 Expand

Fig 4.

Structure of LGD_Net model for the classification of LGDs using CT scans.

More »

Fig 4 Expand

Table 5.

Hyperparameters of LGD_Net.

More »

Table 5 Expand

Table 6.

Extensive comparison of LGD_Net with D1, D2, D3, and D4 in terms Z1, Z2, Z3, Z4, and Z6.

More »

Table 6 Expand

Fig 5.

Accuracy (Z1) graph executed up to 30 epochs; (a) D1, (b) D2, (c) D3, (d) D4, (e) LGD_Net without ATPs & BL-SMT, and (f) Proposed LGD_Net.

More »

Fig 5 Expand

Fig 6.

Recall (Z2) graph executed up to 30 epochs; (a) D1, (b) D2, (c) D3, (d) D4, (e) LGD_Net without ATPs & BL-SMT, and (f) Proposed LGD_Net.

More »

Fig 6 Expand

Fig 7.

Precision (Z3) graph executed up to 30 epochs; (a) D1, (b) D2, (c) D3, (d) D4, (e) LGD_Net without ATPs & BL-SMT, and (f) Proposed LGD_Net.

More »

Fig 7 Expand

Fig 8.

F1-score (Z4) graph executed up to 30 epochs; (a) D1, (b) D2, (c) D3, (d) D4, (e) LGD_Net without ATPs & BL-SMT, and (f) Proposed LGD_Net.

More »

Fig 8 Expand

Fig 9.

AUC (Z6) graph executed up to 30 epochs; (a) D1, (b) D2, (c) D3, (d) D4, (e) LGD_Net without ATPs & BL-SMT, and (f) Proposed LGD_Net.

More »

Fig 9 Expand

Fig 10.

Loss curve executed up to 30 epochs; (a) D1, (b) D2, (c) D3, (d) D4, (e) LGD_Net without ATPs & BL-SMT, and (f) Proposed LGD_Net.

More »

Fig 10 Expand

Fig 11.

ROC graphs show the performance of the models in classifying different lung diseases; (a) D1, (b) D2, (c) D3, (d) D4, (e) LGD_Net without ATPs & BL-SMT, and (f) Proposed LGD_Net.

More »

Fig 11 Expand

Fig 12.

AU (ROC) graphs for classwise evaluation; (a) D1, (b) D2, (c) D3, (d) D4, (e) LGD_Net without ATPs & BL-SMT, and (f) Proposed LGD_Net.

More »

Fig 12 Expand

Table 7.

Segmenting the LGDs by using proposed LGD_Net with SegCaps, UNET, & UNET++.

More »

Table 7 Expand

Fig 13.

Segmentation of several LGDs such as COO, TB, LC, and TB using CT scan images.

The original image of LGD CT scans is illustrated in the first row. The second row presents the original mask of the original image LGDs. From third to fifth, predicted masks are shown that were produced by the UNET, UNET++, and LGD_Net with SegCaps, respectively.

More »

Fig 13 Expand

Fig 14.

Performance evaluation by using C_Mat; (a) D1, (b) D2, (c) D3, (d) D4, (e) LGD_Net without ATPs & BL-SMT, and (f) Proposed LGD_Net.

More »

Fig 14 Expand

Fig 15.

GRAD-CAM of LGD_Net for highlighting the infected area of LGDs using CT scan images.

More »

Fig 15 Expand

Table 8.

Outcomes produced by LGD_Net with and without ATPs and BL-SMT.

More »

Table 8 Expand

Table 9.

Summary of computational cost analysis. Training time represents the time taken to train the model in minutes (min), inference time represents the average time to classify the single image in milliseconds (ms), and model complexity is measured in terms of parameters (Millions = M) and model size (Megabytes = MB).

More »

Table 9 Expand

Table 10.

Performance analysis of LGD_Net with SOTA in classifying LGDs.

More »

Table 10 Expand