---
license: apache-2.0
---

# Chest X-Ray (CXR) Report Generation: Single-CXR

More details to come...

## Example:
```python
import os, torch, transformers
from PIL import Image
from torchvision import transforms

ckpt_name = 'aehrc/mimic-cxr-report-gen-single'

encoder_decoder = transformers.AutoModel.from_pretrained(ckpt_name, trust_remote_code=True)
tokenizer = transformers.PreTrainedTokenizerFast.from_pretrained(ckpt_name)
image_processor = transformers.AutoFeatureExtractor.from_pretrained(ckpt_name)

test_transforms = transforms.Compose(
    [
        transforms.Resize(size=image_processor.size['shortest_edge']),
        transforms.CenterCrop(size=[
            image_processor.size['shortest_edge'],
            image_processor.size['shortest_edge'],
        ]
        ),
        transforms.ToTensor(),
        transforms.Normalize(
            mean=image_processor.image_mean,
            std=image_processor.image_std,
        ),
    ]
)

url = 'https://www.stritch.luc.edu/lumen/meded/radio/curriculum/IPM/PCM/86a_labelled.jpg'
response = requests.get(url)
image_a = Image.open(BytesIO(response.content))
image_a = image_a.convert('RGB')
image_a = test_transforms(image_a)

url = 'https://prod-images-static.radiopaedia.org/images/566180/d527ff6fc1482161c9225345c4ab42_big_gallery.jpg'
response = requests.get(url)
image_b = Image.open(BytesIO(response.content))
image_b = image_b.convert('RGB')
image_b = test_transforms(image_b)

images = torch.stack([image_a, image_b], dim=0)
images.shape

outputs = encoder_decoder.generate(
    pixel_values=images,
    special_token_ids=[tokenizer.sep_token_id],
    bos_token_id=tokenizer.bos_token_id,
    eos_token_id=tokenizer.eos_token_id,
    pad_token_id=tokenizer.pad_token_id,
    return_dict_in_generate=True,
    use_cache=True,
    max_length=256,
    num_beams=4,
)

findings, impression = encoder_decoder.split_and_decode_sections(
    outputs.sequences,
    [tokenizer.sep_token_id, tokenizer.eos_token_id],
    tokenizer,
)

for i, j in zip(findings, impression):
    print(f'Findings: {i}\nImpression: {j}\n')
```