Table 6 The experimental results of the proposed BPI-MVQA on the ImageCLEF2019 VQA-Med dataset
From: BPI-MVQA: a bi-branch model for medical visual question answering
Type | Accuracy | BLEU | Recall | Precision |
|---|---|---|---|---|
Modality | 0.848 | 0.882 | 0.645 | 0.619 |
Plane | 0.848 | 0.85 | 0.464 | 0.407 |
Organ | 0.728 | 0.759 | 0.646 | 0.565 |
Abnormality | 0.192 | 0.256 | 0.076 | 0.079 |
Open-ended | 0.626 | 0.662 | 0.187 | 0.178 |
Close-ended | 0.844 | 0.851 | 0.807 | 0.807 |
All | 0.654 | 0.687 | 0.189 | 0.181 |