diff --git a/ci/unit_tests/test_cxr_image_synthesis_latent_diffusion_model.py b/ci/unit_tests/test_cxr_image_synthesis_latent_diffusion_model.py
new file mode 100644
index 00000000..3e94e91d
--- /dev/null
+++ b/ci/unit_tests/test_cxr_image_synthesis_latent_diffusion_model.py
@@ -0,0 +1,45 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+# http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import os
+import unittest
+
+from monai.bundle import ConfigWorkflow
+from parameterized import parameterized
+from utils import check_workflow
+
+TEST_CASE_1 = [ # inference
+ {
+ "bundle_root": "models/cxr_image_synthesis_latent_diffusion_model",
+ "prompt": "Big right-sided pleural effusion. Normal left lung.",
+ "guidance_scale": 7.0,
+ }
+]
+
+
+class TestCXRLatentDiffusionInference(unittest.TestCase):
+ @parameterized.expand([TEST_CASE_1])
+ def test_inference(self, params):
+ bundle_root = params["bundle_root"]
+ inference_file = os.path.join(bundle_root, "configs/inference.json")
+ trainer = ConfigWorkflow(
+ workflow_type="inference",
+ config_file=inference_file,
+ logging_file=os.path.join(bundle_root, "configs/logging.conf"),
+ meta_file=os.path.join(bundle_root, "configs/metadata.json"),
+ **params,
+ )
+ check_workflow(trainer, check_properties=True)
+
+
+if __name__ == "__main__":
+ loader = unittest.TestLoader()
+ unittest.main(testLoader=loader)
diff --git a/ci/verify_bundle.py b/ci/verify_bundle.py
index 845611fa..7beca3e5 100644
--- a/ci/verify_bundle.py
+++ b/ci/verify_bundle.py
@@ -54,6 +54,8 @@ def _get_weights_names(bundle: str):
if bundle == "pediatric_abdominal_ct_segmentation":
# skip test for this bundle's ts file
return "dynunet_FT.pt", None
+ if bundle == "cxr_image_synthesis_latent_diffusion_model":
+ return "autoencoder.pt", None
return "model.pt", "model.ts"
diff --git a/models/cxr_image_synthesis_latent_diffusion_model/LICENSE b/models/cxr_image_synthesis_latent_diffusion_model/LICENSE
new file mode 100644
index 00000000..261eeb9e
--- /dev/null
+++ b/models/cxr_image_synthesis_latent_diffusion_model/LICENSE
@@ -0,0 +1,201 @@
+ Apache License
+ Version 2.0, January 2004
+ http://www.apache.org/licenses/
+
+ TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+ 1. Definitions.
+
+ "License" shall mean the terms and conditions for use, reproduction,
+ and distribution as defined by Sections 1 through 9 of this document.
+
+ "Licensor" shall mean the copyright owner or entity authorized by
+ the copyright owner that is granting the License.
+
+ "Legal Entity" shall mean the union of the acting entity and all
+ other entities that control, are controlled by, or are under common
+ control with that entity. For the purposes of this definition,
+ "control" means (i) the power, direct or indirect, to cause the
+ direction or management of such entity, whether by contract or
+ otherwise, or (ii) ownership of fifty percent (50%) or more of the
+ outstanding shares, or (iii) beneficial ownership of such entity.
+
+ "You" (or "Your") shall mean an individual or Legal Entity
+ exercising permissions granted by this License.
+
+ "Source" form shall mean the preferred form for making modifications,
+ including but not limited to software source code, documentation
+ source, and configuration files.
+
+ "Object" form shall mean any form resulting from mechanical
+ transformation or translation of a Source form, including but
+ not limited to compiled object code, generated documentation,
+ and conversions to other media types.
+
+ "Work" shall mean the work of authorship, whether in Source or
+ Object form, made available under the License, as indicated by a
+ copyright notice that is included in or attached to the work
+ (an example is provided in the Appendix below).
+
+ "Derivative Works" shall mean any work, whether in Source or Object
+ form, that is based on (or derived from) the Work and for which the
+ editorial revisions, annotations, elaborations, or other modifications
+ represent, as a whole, an original work of authorship. For the purposes
+ of this License, Derivative Works shall not include works that remain
+ separable from, or merely link (or bind by name) to the interfaces of,
+ the Work and Derivative Works thereof.
+
+ "Contribution" shall mean any work of authorship, including
+ the original version of the Work and any modifications or additions
+ to that Work or Derivative Works thereof, that is intentionally
+ submitted to Licensor for inclusion in the Work by the copyright owner
+ or by an individual or Legal Entity authorized to submit on behalf of
+ the copyright owner. For the purposes of this definition, "submitted"
+ means any form of electronic, verbal, or written communication sent
+ to the Licensor or its representatives, including but not limited to
+ communication on electronic mailing lists, source code control systems,
+ and issue tracking systems that are managed by, or on behalf of, the
+ Licensor for the purpose of discussing and improving the Work, but
+ excluding communication that is conspicuously marked or otherwise
+ designated in writing by the copyright owner as "Not a Contribution."
+
+ "Contributor" shall mean Licensor and any individual or Legal Entity
+ on behalf of whom a Contribution has been received by Licensor and
+ subsequently incorporated within the Work.
+
+ 2. Grant of Copyright License. Subject to the terms and conditions of
+ this License, each Contributor hereby grants to You a perpetual,
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+ copyright license to reproduce, prepare Derivative Works of,
+ publicly display, publicly perform, sublicense, and distribute the
+ Work and such Derivative Works in Source or Object form.
+
+ 3. Grant of Patent License. Subject to the terms and conditions of
+ this License, each Contributor hereby grants to You a perpetual,
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+ (except as stated in this section) patent license to make, have made,
+ use, offer to sell, sell, import, and otherwise transfer the Work,
+ where such license applies only to those patent claims licensable
+ by such Contributor that are necessarily infringed by their
+ Contribution(s) alone or by combination of their Contribution(s)
+ with the Work to which such Contribution(s) was submitted. If You
+ institute patent litigation against any entity (including a
+ cross-claim or counterclaim in a lawsuit) alleging that the Work
+ or a Contribution incorporated within the Work constitutes direct
+ or contributory patent infringement, then any patent licenses
+ granted to You under this License for that Work shall terminate
+ as of the date such litigation is filed.
+
+ 4. Redistribution. You may reproduce and distribute copies of the
+ Work or Derivative Works thereof in any medium, with or without
+ modifications, and in Source or Object form, provided that You
+ meet the following conditions:
+
+ (a) You must give any other recipients of the Work or
+ Derivative Works a copy of this License; and
+
+ (b) You must cause any modified files to carry prominent notices
+ stating that You changed the files; and
+
+ (c) You must retain, in the Source form of any Derivative Works
+ that You distribute, all copyright, patent, trademark, and
+ attribution notices from the Source form of the Work,
+ excluding those notices that do not pertain to any part of
+ the Derivative Works; and
+
+ (d) If the Work includes a "NOTICE" text file as part of its
+ distribution, then any Derivative Works that You distribute must
+ include a readable copy of the attribution notices contained
+ within such NOTICE file, excluding those notices that do not
+ pertain to any part of the Derivative Works, in at least one
+ of the following places: within a NOTICE text file distributed
+ as part of the Derivative Works; within the Source form or
+ documentation, if provided along with the Derivative Works; or,
+ within a display generated by the Derivative Works, if and
+ wherever such third-party notices normally appear. The contents
+ of the NOTICE file are for informational purposes only and
+ do not modify the License. You may add Your own attribution
+ notices within Derivative Works that You distribute, alongside
+ or as an addendum to the NOTICE text from the Work, provided
+ that such additional attribution notices cannot be construed
+ as modifying the License.
+
+ You may add Your own copyright statement to Your modifications and
+ may provide additional or different license terms and conditions
+ for use, reproduction, or distribution of Your modifications, or
+ for any such Derivative Works as a whole, provided Your use,
+ reproduction, and distribution of the Work otherwise complies with
+ the conditions stated in this License.
+
+ 5. Submission of Contributions. Unless You explicitly state otherwise,
+ any Contribution intentionally submitted for inclusion in the Work
+ by You to the Licensor shall be under the terms and conditions of
+ this License, without any additional terms or conditions.
+ Notwithstanding the above, nothing herein shall supersede or modify
+ the terms of any separate license agreement you may have executed
+ with Licensor regarding such Contributions.
+
+ 6. Trademarks. This License does not grant permission to use the trade
+ names, trademarks, service marks, or product names of the Licensor,
+ except as required for reasonable and customary use in describing the
+ origin of the Work and reproducing the content of the NOTICE file.
+
+ 7. Disclaimer of Warranty. Unless required by applicable law or
+ agreed to in writing, Licensor provides the Work (and each
+ Contributor provides its Contributions) on an "AS IS" BASIS,
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+ implied, including, without limitation, any warranties or conditions
+ of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+ PARTICULAR PURPOSE. You are solely responsible for determining the
+ appropriateness of using or redistributing the Work and assume any
+ risks associated with Your exercise of permissions under this License.
+
+ 8. Limitation of Liability. In no event and under no legal theory,
+ whether in tort (including negligence), contract, or otherwise,
+ unless required by applicable law (such as deliberate and grossly
+ negligent acts) or agreed to in writing, shall any Contributor be
+ liable to You for damages, including any direct, indirect, special,
+ incidental, or consequential damages of any character arising as a
+ result of this License or out of the use or inability to use the
+ Work (including but not limited to damages for loss of goodwill,
+ work stoppage, computer failure or malfunction, or any and all
+ other commercial damages or losses), even if such Contributor
+ has been advised of the possibility of such damages.
+
+ 9. Accepting Warranty or Additional Liability. While redistributing
+ the Work or Derivative Works thereof, You may choose to offer,
+ and charge a fee for, acceptance of support, warranty, indemnity,
+ or other liability obligations and/or rights consistent with this
+ License. However, in accepting such obligations, You may act only
+ on Your own behalf and on Your sole responsibility, not on behalf
+ of any other Contributor, and only if You agree to indemnify,
+ defend, and hold each Contributor harmless for any liability
+ incurred by, or claims asserted against, such Contributor by reason
+ of your accepting any such warranty or additional liability.
+
+ END OF TERMS AND CONDITIONS
+
+ APPENDIX: How to apply the Apache License to your work.
+
+ To apply the Apache License to your work, attach the following
+ boilerplate notice, with the fields enclosed by brackets "[]"
+ replaced with your own identifying information. (Don't include
+ the brackets!) The text should be enclosed in the appropriate
+ comment syntax for the file format. We also recommend that a
+ file or class name and description of purpose be included on the
+ same "printed page" as the copyright notice for easier
+ identification within third-party archives.
+
+ Copyright [yyyy] [name of copyright owner]
+
+ Licensed under the Apache License, Version 2.0 (the "License");
+ you may not use this file except in compliance with the License.
+ You may obtain a copy of the License at
+
+ http://www.apache.org/licenses/LICENSE-2.0
+
+ Unless required by applicable law or agreed to in writing, software
+ distributed under the License is distributed on an "AS IS" BASIS,
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ See the License for the specific language governing permissions and
+ limitations under the License.
diff --git a/models/cxr_image_synthesis_latent_diffusion_model/configs/inference.json b/models/cxr_image_synthesis_latent_diffusion_model/configs/inference.json
new file mode 100644
index 00000000..e186b319
--- /dev/null
+++ b/models/cxr_image_synthesis_latent_diffusion_model/configs/inference.json
@@ -0,0 +1,115 @@
+{
+ "imports": [
+ "$import torch",
+ "$from datetime import datetime",
+ "$from pathlib import Path",
+ "$from transformers import CLIPTextModel",
+ "$from transformers import CLIPTokenizer"
+ ],
+ "bundle_root": ".",
+ "dataset_dir": "",
+ "dataset": "",
+ "evaluator": "",
+ "inferer": "",
+ "load_old": 1,
+ "model_dir": "$@bundle_root + '/models'",
+ "output_dir": "$@bundle_root + '/output'",
+ "create_output_dir": "$Path(@output_dir).mkdir(exist_ok=True)",
+ "prompt": "Big right-sided pleural effusion",
+ "prompt_list": "$['', @prompt]",
+ "guidance_scale": 7.0,
+ "device": "$torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')",
+ "tokenizer": "$CLIPTokenizer.from_pretrained(\"stabilityai/stable-diffusion-2-1-base\", subfolder=\"tokenizer\")",
+ "text_encoder": "$CLIPTextModel.from_pretrained(\"stabilityai/stable-diffusion-2-1-base\", subfolder=\"text_encoder\")",
+ "tokenized_prompt": "$@tokenizer(@prompt_list, padding=\"max_length\", max_length=@tokenizer.model_max_length, truncation=True,return_tensors=\"pt\")",
+ "prompt_embeds": "$@text_encoder(@tokenized_prompt.input_ids.squeeze(1))[0].to(@device)",
+ "out_file": "$datetime.now().strftime('sample_%H%M%S_%d%m%Y')",
+ "autoencoder_def": {
+ "_target_": "monai.networks.nets.AutoencoderKL",
+ "spatial_dims": 2,
+ "in_channels": 1,
+ "out_channels": 1,
+ "latent_channels": 3,
+ "channels": [
+ 64,
+ 128,
+ 128,
+ 128
+ ],
+ "num_res_blocks": 2,
+ "norm_num_groups": 32,
+ "norm_eps": 1e-06,
+ "attention_levels": [
+ false,
+ false,
+ false,
+ false
+ ],
+ "with_encoder_nonlocal_attn": false,
+ "with_decoder_nonlocal_attn": false
+ },
+ "network_def": "@autoencoder_def",
+ "load_autoencoder_path": "$@model_dir + '/autoencoder.pt'",
+ "load_autoencoder_func": "$@autoencoder_def.load_old_state_dict if bool(@load_old) else @autoencoder_def.load_state_dict",
+ "load_autoencoder": "$@load_autoencoder_func(torch.load(@load_autoencoder_path))",
+ "autoencoder": "$@autoencoder_def.to(@device)",
+ "diffusion_def": {
+ "_target_": "monai.networks.nets.DiffusionModelUNet",
+ "spatial_dims": 2,
+ "in_channels": 3,
+ "out_channels": 3,
+ "channels": [
+ 256,
+ 512,
+ 768
+ ],
+ "num_res_blocks": 2,
+ "attention_levels": [
+ false,
+ true,
+ true
+ ],
+ "norm_num_groups": 32,
+ "norm_eps": 1e-06,
+ "resblock_updown": false,
+ "num_head_channels": [
+ 0,
+ 512,
+ 768
+ ],
+ "with_conditioning": true,
+ "transformer_num_layers": 1,
+ "cross_attention_dim": 1024
+ },
+ "load_diffusion_path": "$@model_dir + '/model.pt'",
+ "load_diffusion_func": "$@diffusion_def.load_old_state_dict if bool(@load_old) else @diffusion_def.load_state_dict",
+ "load_diffusion": "$@load_diffusion_func(torch.load(@load_diffusion_path))",
+ "diffusion": "$@diffusion_def.to(@device)",
+ "scheduler": {
+ "_target_": "monai.networks.schedulers.DDIMScheduler",
+ "_requires_": [
+ "@load_diffusion",
+ "@load_autoencoder"
+ ],
+ "beta_start": 0.0015,
+ "beta_end": 0.0205,
+ "num_train_timesteps": 1000,
+ "schedule": "scaled_linear_beta",
+ "prediction_type": "v_prediction",
+ "clip_sample": false
+ },
+ "noise": "$torch.randn((1, 3, 64, 64)).to(@device)",
+ "set_timesteps": "$@scheduler.set_timesteps(num_inference_steps=50)",
+ "sampler": {
+ "_target_": "scripts.sampler.Sampler",
+ "_requires_": "@set_timesteps"
+ },
+ "sample": "$@sampler.sampling_fn(@noise, @autoencoder, @diffusion, @scheduler, @prompt_embeds)",
+ "saver": {
+ "_target_": "scripts.saver.JPGSaver",
+ "_requires_": "@create_output_dir",
+ "output_dir": "@output_dir"
+ },
+ "run": "$@saver.save(@sample, @out_file)",
+ "save": "$torch.save(@sample, @output_dir + '/' + @out_file + '.pt')"
+}
diff --git a/models/cxr_image_synthesis_latent_diffusion_model/configs/logging.conf b/models/cxr_image_synthesis_latent_diffusion_model/configs/logging.conf
new file mode 100644
index 00000000..91c1a21c
--- /dev/null
+++ b/models/cxr_image_synthesis_latent_diffusion_model/configs/logging.conf
@@ -0,0 +1,21 @@
+[loggers]
+keys=root
+
+[handlers]
+keys=consoleHandler
+
+[formatters]
+keys=fullFormatter
+
+[logger_root]
+level=INFO
+handlers=consoleHandler
+
+[handler_consoleHandler]
+class=StreamHandler
+level=INFO
+formatter=fullFormatter
+args=(sys.stdout,)
+
+[formatter_fullFormatter]
+format=%(asctime)s - %(name)s - %(levelname)s - %(message)s
diff --git a/models/cxr_image_synthesis_latent_diffusion_model/configs/metadata.json b/models/cxr_image_synthesis_latent_diffusion_model/configs/metadata.json
new file mode 100644
index 00000000..1e898ed7
--- /dev/null
+++ b/models/cxr_image_synthesis_latent_diffusion_model/configs/metadata.json
@@ -0,0 +1,83 @@
+{
+ "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_20220324.json",
+ "version": "1.0.0",
+ "changelog": {
+ "1.0.0": "Initial release"
+ },
+ "monai_version": "1.4.0",
+ "pytorch_version": "2.5.1",
+ "numpy_version": "1.26.4",
+ "optional_packages_version": {
+ "transformers": "4.46.3"
+ },
+ "task": "Chest X-ray image synthesis",
+ "description": "A generative model for creating high-resolution chest X-ray based on MIMIC dataset",
+ "copyright": "Copyright (c) MONAI Consortium",
+ "authors": "Walter Hugo Lopez Pinaya, Mark Graham, Eric Kerfoot, Virginia Fernandez",
+ "data_source": "https://physionet.org/content/mimic-cxr-jpg/2.0.0/",
+ "data_type": "image",
+ "image_classes": "Radiography (X-ray) with 512 x 512 pixels",
+ "intended_use": "This is a research tool/prototype and not to be used clinically",
+ "network_data_format": {
+ "inputs": {
+ "image": {
+ "type": "N/A",
+ "format": "N/A",
+ "modality": "N/A",
+ "num_channels": 0,
+ "spatial_shape": [],
+ "dtype": "N/A",
+ "value_range": [],
+ "is_patch_data": false
+ },
+ "latent_representation": {
+ "type": "image",
+ "format": "magnitude",
+ "modality": "CXR",
+ "num_channels": 3,
+ "spatial_shape": [
+ 77,
+ 64,
+ 64
+ ],
+ "dtype": "float32",
+ "value_range": [],
+ "is_patch_data": false
+ },
+ "timesteps": {
+ "type": "vector",
+ "value_range": [
+ 0,
+ 1000
+ ],
+ "dtype": "long"
+ },
+ "context": {
+ "type": "vector",
+ "value_range": [],
+ "dtype": "float32"
+ }
+ },
+ "outputs": {
+ "pred": {
+ "type": "image",
+ "format": "magnitude",
+ "modality": "CXR",
+ "num_channels": 1,
+ "spatial_shape": [
+ 512,
+ 512
+ ],
+ "dtype": "float32",
+ "value_range": [
+ 0,
+ 1
+ ],
+ "is_patch_data": false,
+ "channel_def": {
+ "0": "X-ray"
+ }
+ }
+ }
+ }
+}
diff --git a/models/cxr_image_synthesis_latent_diffusion_model/docs/README.md b/models/cxr_image_synthesis_latent_diffusion_model/docs/README.md
new file mode 100644
index 00000000..32a8ee23
--- /dev/null
+++ b/models/cxr_image_synthesis_latent_diffusion_model/docs/README.md
@@ -0,0 +1,71 @@
+# Description
+
+A diffusion model to synthetise X-Ray images based on radiological report impressions.
+
+# Model Overview
+This model is trained from scratch using the Latent Diffusion Model architecture [1] and is used for the synthesis of
+2D Chest X-ray conditioned on Radiological reports. The model is divided into two parts: an autoencoder with a
+KL-regularisation model that compresses data into a latent space and a diffusion model that learns to generate
+conditioned synthetic latent representations. This model is conditioned on Findings and Impressions from radiological
+reports. The original repository can be found [here](https://github.com/Warvito/generative_chestxray)
+
+
+
+Figure 1 - Synthetic images from the model.
+
+# Data
+The model was trained on brain data from 90,000 participants from the MIMIC dataset [2] [3]. We downsampled the
+original images to have a format of 512 x 512 pixels.
+
+## Preprocessing
+We resized the original images to make the smallest sides have 512 pixels. When inputting it to the network, we center
+cropped the images to 512 x 512. The pixel intensity was normalised to be between [0, 1]. The text data was obtained
+from associated radiological reports. We randoomly extracted sentences from the findings and impressions sections of the
+reports, having a maximum of 5 sentences and 77 tokens. The text was tokenised using the CLIPTokenizer from
+transformers package (https://github.com/huggingface/transformers) (pretrained model
+"stabilityai/stable-diffusion-2-1-base") and then encoded using CLIPTextModel from the same package and pretrained
+model.
+
+# Examples of inference
+
+Here we included a few examples of commands to sample images from the model and save them as .jpg files. The available
+arguments for this task are: "--prompt" (str) text prompt to condition the model on; "--guidance_scale" (float), the
+parameter that controls how much the image generation process follows the text prompt. The higher the value, the more
+the image sticks to a given text input (the common range is between 1-21).
+
+Examples:
+
+```shell
+$ python -m monai.bundle run --config_file configs/inference.json --prompt "Big right-sided pleural effusion" --guidance_scale 7.0
+```
+
+```shell
+$ python -m monai.bundle run --config_file configs/inference.json --prompt "Small right-sided pleural effusion" --guidance_scale 7.0
+```
+
+```shell
+$ python -m monai.bundle run --config_file configs/inference.json --prompt "Bilateral pleural effusion" --guidance_scale 7.0
+```
+
+```shell
+$ python -m monai.bundle run --config_file configs/inference.json --prompt "Cardiomegaly" --guidance_scale 7.0
+```
+
+## Using a new version of the model
+
+If you want to use the checkpoints from a newly fine-tuned model, you need to set parameter load_old to 0 when you run inference,
+to avoid the function load_old_state_dict being called instead of load_state_dict to be called, currently default, as it is
+required to load the checkpoint from the original GenerativeModels repository.
+
+```shell
+$ python -m monai.bundle run --config_file configs/inference.json --prompt "Pleural effusion." --guidance_scale 7.0 --load_old 0
+```
+
+## References
+
+
+[1] Pinaya, Walter HL, et al. "Brain imaging generation with latent diffusion models." MICCAI Workshop on Deep Generative Models. Springer, Cham, 2022.
+
+[2] Johnson, A., Lungren, M., Peng, Y., Lu, Z., Mark, R., Berkowitz, S., & Horng, S. (2019). MIMIC-CXR-JPG - chest radiographs with structured labels (version 2.0.0). PhysioNet. https://doi.org/10.13026/8360-t248.
+
+[3] Johnson AE, Pollard TJ, Berkowitz S, Greenbaum NR, Lungren MP, Deng CY, Mark RG, Horng S. MIMIC-CXR: A large publicly available database of labeled chest radiographs. arXiv preprint arXiv:1901.07042. 2019 Jan 21.
diff --git a/models/cxr_image_synthesis_latent_diffusion_model/docs/figure_1.png b/models/cxr_image_synthesis_latent_diffusion_model/docs/figure_1.png
new file mode 100644
index 00000000..1baa0adb
Binary files /dev/null and b/models/cxr_image_synthesis_latent_diffusion_model/docs/figure_1.png differ
diff --git a/models/cxr_image_synthesis_latent_diffusion_model/large_files.yml b/models/cxr_image_synthesis_latent_diffusion_model/large_files.yml
new file mode 100644
index 00000000..716abf71
--- /dev/null
+++ b/models/cxr_image_synthesis_latent_diffusion_model/large_files.yml
@@ -0,0 +1,9 @@
+large_files:
+ - path: "models/autoencoder.pt"
+ url: "https://drive.google.com/uc?export=download&id=1paDN1m-Q_Oy8d_BanPkRTi3RlNB_Sv_h"
+ hash_val: "7f579cb789597db7bb5de1488f54bc6c"
+ hash_type: "md5"
+ - path: "models/model.pt"
+ url: "https://drive.google.com/uc?export=download&id=1CjcmiPu5_QWr-f7wDJsXrCCcVeczneGT"
+ hash_val: "c3fd4c8e38cd1d7250a8903cca935823"
+ hash_type: "md5"
diff --git a/models/cxr_image_synthesis_latent_diffusion_model/scripts/__init__.py b/models/cxr_image_synthesis_latent_diffusion_model/scripts/__init__.py
new file mode 100644
index 00000000..e69de29b
diff --git a/models/cxr_image_synthesis_latent_diffusion_model/scripts/sampler.py b/models/cxr_image_synthesis_latent_diffusion_model/scripts/sampler.py
new file mode 100644
index 00000000..c0e602e3
--- /dev/null
+++ b/models/cxr_image_synthesis_latent_diffusion_model/scripts/sampler.py
@@ -0,0 +1,43 @@
+from __future__ import annotations
+
+import torch
+import torch.nn as nn
+from monai.utils import optional_import
+from torch.cuda.amp import autocast
+
+tqdm, has_tqdm = optional_import("tqdm", name="tqdm")
+
+
+class Sampler:
+ def __init__(self) -> None:
+ super().__init__()
+
+ @torch.no_grad()
+ def sampling_fn(
+ self,
+ noise: torch.Tensor,
+ autoencoder_model: nn.Module,
+ diffusion_model: nn.Module,
+ scheduler: nn.Module,
+ prompt_embeds: torch.Tensor,
+ guidance_scale: float = 7.0,
+ scale_factor: float = 0.3,
+ ) -> torch.Tensor:
+ if has_tqdm:
+ progress_bar = tqdm(scheduler.timesteps)
+ else:
+ progress_bar = iter(scheduler.timesteps)
+
+ for t in progress_bar:
+ noise_input = torch.cat([noise] * 2)
+ model_output = diffusion_model(
+ noise_input, timesteps=torch.Tensor((t,)).to(noise.device).long(), context=prompt_embeds
+ )
+ noise_pred_uncond, noise_pred_text = model_output.chunk(2)
+ noise_pred = noise_pred_uncond + guidance_scale * (noise_pred_text - noise_pred_uncond)
+ noise, _ = scheduler.step(noise_pred, t, noise)
+
+ with autocast():
+ sample = autoencoder_model.decode_stage_2_outputs(noise / scale_factor)
+
+ return sample
diff --git a/models/cxr_image_synthesis_latent_diffusion_model/scripts/saver.py b/models/cxr_image_synthesis_latent_diffusion_model/scripts/saver.py
new file mode 100644
index 00000000..05e88722
--- /dev/null
+++ b/models/cxr_image_synthesis_latent_diffusion_model/scripts/saver.py
@@ -0,0 +1,17 @@
+from __future__ import annotations
+
+import numpy as np
+import torch
+from PIL import Image
+
+
+class JPGSaver:
+ def __init__(self, output_dir: str) -> None:
+ super().__init__()
+ self.output_dir = output_dir
+
+ def save(self, image_data: torch.Tensor, file_name: str) -> None:
+ image_data = np.clip(image_data.cpu().numpy(), 0, 1)
+ image_data = (image_data * 255).astype(np.uint8)
+ im = Image.fromarray(image_data[0, 0])
+ im.save(self.output_dir + "/" + file_name + ".jpg")