Table 3 Models hyperparameters results including Precision, Sensitivity, F1-score and mAP.

From: Improving diagnosis accuracy with an intelligent image retrieval system for lung pathologies detection: a features extractor approach

Pathologies

Precision

Sensitivity

F1-score

mAP 0.5

mAP [0.5:0.95]

Aortic enlargement

0.985

0.521

0.681

0.71

0.559

Atelectasis

0.833

0.778

0.804

0.824

0.593

Calcification

0.815

0.769

0.791

0.809

0.587

Cardiomegaly

0.991

0.44

0.609

0.63

0.487

Consolidation

0.875

0.74

0.801

0.807

0.607

ILD

0.949

0.716

0.816

0.828

0.628

Infiltration

0.912

0.739

0.816

0.824

0.619

Lung Opacity

0.894

0.781

0.833

0.854

0.643

Nodule/Mass

0.904

0.701

0.789

0.77

0.555

Other lesions

0.897

0.831

0.862

0.884

0.652

Pleural effusion

0.887

0.72

0.794

0.826

0.567

Pleural thickening

0.941

0.804

0.867

0.891

0.631

Pneumothorax

0.86

0.567

0.683

0.67

0.521

Pulmonary fibrosis

0.91

0.728

0.808

0.823

0.599

avg

0.909

0.702

0.792

0.796

0.589

Back to article page