Table 10.
Overview of LMs for report generation. The asterisks (*) indicate terms that are either not present in the original paper or do not apply in this context.
| References | ROI | Modality | Dataset | Model name | Vision model | Language model |
|---|---|---|---|---|---|---|
| Zhong et al (2023) | Chest, Abdomen, musculoskeletal system, head, maxillofacial and neck | CT, MRI | Institutional (Six Chinese Hospital) | ChatRadio-Valuer | * | Llama2 |
| Yan and Pei (2022) | Chest | x-ray | MIMIC-CXR, IU x-ray, COV-CTR, NIH ChestX-ray14 | Clinical-BERT | DenseNet121 | BERT-base |
| Leonardi et al (2023) | Chest | x-ray | MIMIC-CXR | * | ViT, CheXNet | Transformer |
| Li (2023) | Chest | x-ray | MIMIC-CXR, RSNA Pneumonia, COVID, IU Chest x-ray | * | CLIP | GPT-2, OPT-1.3B, OPT-2.7B |
| Tanida et al (2023) | Chest | x-ray | Chest ImaGenome v1.0.0 | RGRG Method | ResNet-50, Faster R-CNN | Transformer |
| Huang et al (2023) | Chest | x-ray | IU-Xray, MIMIC-CXR | KiUT | ResNet101, U Transformer | BERT |
| Cao et al (2023) | Gastrointestinal tract, chest | Endoscope, x-ray | Gastrointestinal endoscope image dataset (GE), IU-CX, MIMIC-CXR | MMTN | DenseNet-121 | BERT |
| Moon et al (2022) | Chest | x-ray | MIMIC-CXR, Open-I | MedViLL | CNN | BERT |
| Kong et al (2022) | Chest | x-ray | IU x-ray, MIMIC-CXR | TranSQ | ViT-B/32 | MPNet |