Updated the metadata for all models, focused on name, description, and task for Model Zoo website improvements.

hf_models/ct_chat/metadata.json

@@ -1,7 +1,8 @@
 {
     "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_hf_20250321.json",
-    "version": "1.0.0",
+    "version": "1.1.0",
     "changelog": {
+        "1.1.0": "enhanced metadata with improved descriptions, task specification, and intended use documentation",
         "1.0.0": "initial release of CT_CHAT model"
     },
     "monai_version": "1.4.0",
@@ -18,9 +19,9 @@
         "ct_clip": "",
         "ct_chat": ""
     },
-    "name": "CT_CHAT",
-    "task": "Vision-language foundational chat model for 3D chest CT volumes",
-    "description": "CT-CHAT is a multimodal AI assistant designed to enhance the interpretation and diagnostic capabilities of 3D chest CT imaging. Building on the strong foundation of CT-CLIP, it integrates both visual and language processing to handle diverse tasks like visual question answering, report generation, and multiple-choice questions. Trained on over 2.7 million question-answer pairs from CT-RATE, it leverages 3D spatial information, making it superior to 2D-based models.",
+    "name": "CT-CHAT",
+    "task": "Vision-Language Chat Model for 3D Chest CT Analysis",
+    "description": "CT-CHAT is a multimodal AI assistant specifically designed for 3D chest CT imaging interpretation and analysis. The model excels at tasks including visual question answering, report generation, and multiple-choice questions, leveraging full 3D spatial information for superior performance compared to 2D-based approaches.",
     "authors": "Ibrahim Ethem Hamamci, Sezgin Er, Furkan Almas, et al.",
     "copyright": "Ibrahim Ethem Hamamci and collaborators",
     "data_source": "CT-RATE dataset",

hf_models/exaonepath/metadata.json

@@ -1,7 +1,8 @@
 {
     "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_hf_20250321.json",
-    "version": "1.0.0",
+    "version": "1.1.0",
     "changelog": {
+        "1.1.0": "Enhanced metadata with detailed model architecture, performance metrics on downstream tasks, and preprocessing requirements.",
         "1.0.0": "initial release of EXAONEPath 1.0"
     },
     "monai_version": "1.4.0",
@@ -19,11 +20,11 @@
         "exaonepath": ""
     },
     "name": "EXAONEPath",
-    "task": "Pathology foundation model",
-    "description": "EXAONEPath is a patch-level pathology pretrained model with 86 million parameters, pretrained on 285,153,903 patches extracted from 34,795 WSIs.",
-    "authors": "LG AI Research",
+    "task": "Pathology Foundation Model and Feature Extraction",
+    "description": "EXAONEPath is a patch-level pathology foundation model that achieves state-of-the-art performance across multiple pathology tasks while maintaining computational efficiency. It excels in tissue classification, tumor detection, and microsatellite instability assessment.",
+    "authors": "LG AI Research Team",
     "copyright": "LG AI Research",
-    "data_source": "LG AI Research",
+    "data_source": "Large-scale collection of pathology WSIs processed into patches",
     "data_type": "WSI patches",
     "image_classes": "RGB pathology image patches",
     "huggingface_model_id": "LGAI-EXAONE/EXAONEPath",

hf_models/llama3_vila_m3_13b/metadata.json

@@ -1,8 +1,9 @@
 {
     "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_hf_20250321.json",
-    "version": "1.0.0",
+    "version": "1.1.0",
     "changelog": {
-        "1.0.0": "initial release of VILA_M3_13B model"
+        "1.1.0": "enhanced metadata with improved descriptions, task specification, and intended use documentation",
+        "1.0.0": "initial release of VILA_M3_3B model"
     },
     "monai_version": "1.4.0",
     "pytorch_version": "2.4.0",
@@ -12,12 +13,12 @@
         "huggingface_hub": "0.24.2",
         "transformers": "4.43.3"
     },
-    "name": "VILA_M3_13B",
-    "task": "Medical vision-language model",
-    "description": "VILA_M3 is a medical vision language model that enhances VLMs with medical expert knowledge, utilizing domain-expert models to improve precision in medical imaging tasks.",
-    "authors": "Vishwesh Nath, Wenqi Li, Dong Yang, Andriy Myronenko, et al. from NVIDIA, SingHealth, and NIH",
+    "name": "Llama3-VILA-M3-13B",
+    "task": "Medical Visual Language Understanding and Generation",
+    "description": "VILA-M3 is a medical visual language model built on Llama 3 and VILA architecture. This 13B parameter model performs medical image analysis including segmentation, classification, visual question answering, and report generation across multiple imaging modalities.",
+    "authors": "Vishwesh Nath, Wenqi Li, Dong Yang, Andriy Myronenko, et al. from NVIDIA, SingHealth and NIH",
     "copyright": "NVIDIA",
-    "data_source": "NVIDIA",
+    "data_source": "MONAI and specialized medical datasets",
     "data_type": "Medical images and text",
     "image_classes": "Various medical imaging modalities",
     "huggingface_model_id": "MONAI/Llama3-VILA-M3-13B",

hf_models/llama3_vila_m3_3b/metadata.json

@@ -1,7 +1,8 @@
 {
     "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_hf_20250321.json",
-    "version": "1.0.0",
+    "version": "1.1.0",
     "changelog": {
+        "1.1.0": "enhanced metadata with improved descriptions, task specification, and intended use documentation",
         "1.0.0": "initial release of VILA_M3_3B model"
     },
     "monai_version": "1.4.0",
@@ -12,12 +13,12 @@
         "huggingface_hub": "0.24.2",
         "transformers": "4.43.3"
     },
-    "name": "VILA_M3_3B",
-    "task": "Medical vision-language model",
-    "description": "VILA_M3 is a medical vision language model that enhances VLMs with medical expert knowledge, utilizing domain-expert models to improve precision in medical imaging tasks.",
-    "authors": "Vishwesh Nath, Wenqi Li, Dong Yang, Andriy Myronenko, et al. from NVIDIA, SingHealth, and NIH",
+    "name": "Llama3-VILA-M3-3B",
+    "task": "Medical Visual Language Understanding and Generation",
+    "description": "VILA-M3 is a medical visual language model built on Llama 3 and VILA architecture. This 3B parameter model performs medical image analysis including segmentation, classification, visual question answering, and report generation across multiple imaging modalities.",
+    "authors": "Vishwesh Nath, Wenqi Li, Dong Yang, Andriy Myronenko, et al. from NVIDIA, SingHealth and NIH",
     "copyright": "NVIDIA",
-    "data_source": "NVIDIA",
+    "data_source": "MONAI and specialized medical datasets",
     "data_type": "Medical images and text",
     "image_classes": "Various medical imaging modalities",
     "huggingface_model_id": "MONAI/Llama3-VILA-M3-3B",

hf_models/llama3_vila_m3_8b/metadata.json

@@ -1,8 +1,9 @@
 {
     "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_hf_20250321.json",
-    "version": "1.0.0",
+    "version": "1.1.0",
     "changelog": {
-        "1.0.0": "initial release of VILA_M3_8B model"
+        "1.1.0": "enhanced metadata with improved descriptions, task specification, and intended use documentation",
+        "1.0.0": "initial release of VILA_M3_3B model"
     },
     "monai_version": "1.4.0",
     "pytorch_version": "2.4.0",
@@ -12,12 +13,12 @@
         "huggingface_hub": "0.24.2",
         "transformers": "4.43.3"
     },
-    "name": "VILA_M3_8B",
-    "task": "Medical vision-language model",
-    "description": "VILA_M3 is a medical vision language model that enhances VLMs with medical expert knowledge, utilizing domain-expert models to improve precision in medical imaging tasks.",
-    "authors": "Vishwesh Nath, Wenqi Li, Dong Yang, Andriy Myronenko, et al. from NVIDIA, SingHealth, and NIH",
+    "name": "Llama3-VILA-M3-8B",
+    "task": "Medical Visual Language Understanding and Generation",
+    "description": "VILA-M3 is a medical visual language model built on Llama 3 and VILA architecture. This 8B parameter model performs medical image analysis including segmentation, classification, visual question answering, and report generation across multiple imaging modalities.",
+    "authors": "Vishwesh Nath, Wenqi Li, Dong Yang, Andriy Myronenko, et al. from NVIDIA, SingHealth and NIH",
     "copyright": "NVIDIA",
-    "data_source": "NVIDIA",
+    "data_source": "MONAI and specialized medical datasets",
     "data_type": "Medical images and text",
     "image_classes": "Various medical imaging modalities",
     "huggingface_model_id": "MONAI/Llama3-VILA-M3-8B",

models/brain_image_synthesis_latent_diffusion_model/configs/metadata.json

@@ -1,7 +1,8 @@
 {
     "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_20240725.json",
-    "version": "1.0.2",
+    "version": "1.0.3",
     "changelog": {
+        "1.0.3": "enhanced metadata with improved descriptions, task specification",
         "1.0.2": "fix missing dependencies",
         "1.0.1": "update to huggingface hosting",
         "1.0.0": "Initial release"
@@ -13,8 +14,9 @@
         "nibabel": "5.3.2",
         "einops": "0.7.0"
     },
-    "task": "Brain image synthesis",
-    "description": "A generative model for creating high-resolution 3D brain MRI based on UK Biobank",
+    "name": "Brain MRI Latent Diffusion Synthesis",
+    "task": "Conditional Synthesis of 3D Brain MRI with Demographic and Morphological Control",
+    "description": "A latent diffusion model that generates 160x224x160 voxel T1-weighted brain MRI volumes with 1mm isotropic resolution. The model accepts conditional inputs for age, gender, ventricular volume, and brain volume, enabling controlled generation of brain images with specific demographic and morphological characteristics.",
     "authors": "Walter H. L. Pinaya, Petru-Daniel Tudosiu, Jessica Dafflon, Pedro F Da Costa, Virginia Fernandez, Parashkev Nachev, Sebastien Ourselin, and M. Jorge Cardoso",
     "copyright": "Copyright (c) MONAI Consortium",
     "data_source": "https://www.ukbiobank.ac.uk/",

models/brats_mri_axial_slices_generative_diffusion/configs/metadata.json

@@ -1,7 +1,8 @@
 {
     "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_20240725.json",
-    "version": "1.1.3",
+    "version": "1.1.4",
     "changelog": {
+        "1.1.4": "enhance metadata with improved descriptions and task specification",
         "1.1.3": "update to huggingface hosting and fix missing dependencies",
         "1.1.2": "update issue for IgniteInfo",
         "1.1.1": "enable tensorrt",
@@ -28,9 +29,9 @@
         "pytorch-ignite": "0.4.11"
     },
     "supported_apps": {},
-    "name": "BraTS MRI axial slices latent diffusion generation",
-    "task": "BraTS MRI axial slices synthesis",
-    "description": "A generative model for creating 2D brain MRI axial slices from Gaussian noise based on BraTS dataset",
+    "name": "BraTS MRI Axial Slices Latent Diffusion Generation",
+    "task": "Conditional Synthesis of Brain MRI Axial Slices",
+    "description": "Latent diffusion model that synthesizes 2D brain MRI axial slices (240x240 pixels) from Gaussian noise, trained on the BraTS dataset. The model processes 1-channel latent space features (64x64) and generates FLAIR sequences with 1mm in-plane resolution, capturing diverse tumor and brain tissue appearances.",
     "authors": "MONAI team",
     "copyright": "Copyright (c) MONAI Consortium",
     "data_source": "http://medicaldecathlon.com/",

models/brats_mri_generative_diffusion/configs/metadata.json

@@ -1,7 +1,8 @@
 {
     "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_20240725.json",
-    "version": "1.1.3",
+    "version": "1.1.4",
     "changelog": {
+        "1.1.4": "enhanced metadata with improved descriptions and task specification",
         "1.1.3": "update to huggingface hosting and fix missing dependencies",
         "1.1.2": "update issue for IgniteInfo",
         "1.1.1": "enable tensorrt",
@@ -28,9 +29,9 @@
         "tensorboard": "2.17.0"
     },
     "supported_apps": {},
-    "name": "BraTS MRI image latent diffusion generation",
-    "task": "BraTS MRI image synthesis",
-    "description": "A generative model for creating 3D brain MRI from Gaussian noise based on BraTS dataset",
+    "name": "BraTS MRI Latent Diffusion Generation",
+    "task": "Conditional Synthesis of Brain MRI with Tumor Features",
+    "description": "Volumetric latent diffusion model that generates 3D brain MRI volumes (112x128x80 voxels) with tumor features from Gaussian noise, trained on the BraTS multimodal MRI dataset.",
     "authors": "MONAI team",
     "copyright": "Copyright (c) MONAI Consortium",
     "data_source": "http://medicaldecathlon.com/",

models/brats_mri_segmentation/configs/metadata.json

@@ -1,7 +1,8 @@
 {
     "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_20240725.json",
-    "version": "0.5.3",
+    "version": "0.5.4",
     "changelog": {
+        "0.5.4": "enhanced metadata with improved descriptions and task specification",
         "0.5.3": "update to huggingface hosting",
         "0.5.2": "use monai 1.4 and update large files",
         "0.5.1": "update to use monai 1.3.1",
@@ -42,11 +43,11 @@
     },
     "supported_apps": {},
     "name": "BraTS MRI segmentation",
-    "task": "Multimodal Brain Tumor segmentation",
-    "description": "A pre-trained model for volumetric (3D) segmentation of brain tumor subregions from multimodal MRIs based on BraTS 2018 data",
+    "task": "Multimodal Brain Tumor Subregion Segmentation",
+    "description": "3D segmentation model for delineating brain tumor subregions from multimodal MRI scans (T1, T1c, T2, FLAIR). The model processes 4-channel input volumes with 1mm isotropic resolution and outputs 3-channel segmentation masks for tumor core (TC), whole tumor (WT), and enhancing tumor (ET).",
     "authors": "MONAI team",
     "copyright": "Copyright (c) MONAI Consortium",
-    "data_source": "https://www.med.upenn.edu/sbia/brats2018/data.html",
+    "data_source": "BraTS 2018 Challenge Dataset (https://www.med.upenn.edu/sbia/brats2018/data.html)",
     "data_type": "nibabel",
     "image_classes": "4 channel data, T1c, T1, T2, FLAIR at 1x1x1 mm",
     "label_classes": "3 channel data, channel 0 for Tumor core, channel 1 for Whole tumor, channel 2 for Enhancing tumor",

models/breast_density_classification/configs/metadata.json

@@ -1,7 +1,8 @@
 {
     "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_20240725.json",
-    "version": "0.1.7",
+    "version": "0.1.8",
     "changelog": {
+        "0.1.8": "enhance metadata with improved descriptions and task specification",
         "0.1.7": "update to huggingface hosting",
         "0.1.6": "Remove meta dict usage",
         "0.1.5": "Fixed duplication of input output format section",
@@ -19,12 +20,12 @@
     },
     "supported_apps": {},
     "name": "Breast density classification",
-    "task": "Breast Density Classification",
-    "description": "A pre-trained model for classifying breast images (mammograms)  ",
+    "task": "Mammographic Breast Density Classification (BI-RADS)",
+    "description": "A deep learning model for automated classification of breast tissue density in mammograms according to the BI-RADS density categories (A through D). The model processes 299x299 pixel images and classifies breast tissue into four categories: fatty, scattered fibroglandular, heterogeneously dense, and extremely dense.",
     "authors": "Center for Augmented Intelligence in Imaging, Mayo Clinic Florida",
     "copyright": "Copyright (c) Mayo Clinic",
-    "data_source": "Mayo Clinic ",
-    "data_type": "Jpeg",
+    "data_source": "Mayo Clinic",
+    "data_type": "jpeg",
     "image_classes": "three channel data, intensity scaled to [0, 1]. A single grayscale is copied to 3 channels",
     "label_classes": "four classes marked as [1, 0, 0, 0], [0, 1, 0, 0], [0, 0, 1, 0] and [0, 0, 0, 1] for the classes A, B, C and D respectively.",
     "pred_classes": "One hot data",

models/classification_template/configs/metadata.json

@@ -1,7 +1,8 @@
 {
     "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_20240725.json",
-    "version": "0.0.3",
+    "version": "0.0.4",
     "changelog": {
+        "0.0.4": "enhanced metadata with improved descriptions and task specification",
         "0.0.3": "update to huggingface hosting",
         "0.0.2": "update large file yml",
         "0.0.1": "Initial version"
@@ -14,9 +15,9 @@
         "pyyaml": "6.0.2"
     },
     "supported_apps": {},
-    "name": "Classification Template",
-    "task": "Classification Template in 2D images",
-    "description": "This is a template bundle for classifying in 2D, take this as a basis for your own bundles.",
+    "name": "Medical Image Classification Template",
+    "task": "Template for 2D Medical Image Classification",
+    "description": "A comprehensive template for developing 2D medical image classification models, featuring a modular architecture and standardized training pipeline. The template supports single-channel 128x128 pixel input images and outputs 4-class probability distributions, serving as a foundation for custom medical image classification tasks.",
     "authors": "Yun Liu",
     "copyright": "Copyright (c) 2023 MONAI Consortium",
     "network_data_format": {

models/cxr_image_synthesis_latent_diffusion_model/configs/metadata.json

@@ -1,18 +1,22 @@
 {
     "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_20240725.json",
-    "version": "1.0.1",
+    "version": "1.0.2",
     "changelog": {
+        "1.0.2": "enhanced metadata with improved descriptions and task specification",
         "1.0.1": "update to huggingface hosting",
         "1.0.0": "Initial release"
     },
     "monai_version": "1.4.0",
     "pytorch_version": "2.5.1",
     "numpy_version": "1.26.4",
     "required_packages_version": {
-        "transformers": "4.46.3"
+        "transformers": "4.46.3",
+        "einops": "0.8.1",
+        "pillow": "10.4.0"
     },
-    "task": "Chest X-ray image synthesis",
-    "description": "A generative model for creating high-resolution chest X-ray based on MIMIC dataset",
+    "name": "Chest X-ray Latent Diffusion Synthesis",
+    "task": "Conditional Synthesis of Chest X-ray Images with Pathology Control",
+    "description": "A latent diffusion model that generates 512x512 pixel chest X-ray images from a 64x64x77 dimensional latent space. The model processes text-based condition inputs through a 1024-dimensional context vector, enabling controlled generation of X-rays with specific pathological features.",
     "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/",

models/endoscopic_inbody_classification/configs/metadata.json

@@ -1,7 +1,8 @@
 {
     "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_20240725.json",
-    "version": "0.5.0",
+    "version": "0.5.1",
     "changelog": {
+        "0.5.1": "enhance metadata with improved descriptions and task specification",
         "0.5.0": "update to huggingface hosting and fix missing dependencies",
         "0.4.9": "use monai 1.4 and update large files",
         "0.4.8": "update to use monai 1.3.1",
@@ -39,11 +40,11 @@
         "tensorboard": "2.17.0"
     },
     "supported_apps": {},
-    "name": "Endoscopic inbody classification",
-    "task": "Endoscopic inbody classification",
-    "description": "A pre-trained binary classification model for endoscopic inbody classification task",
-    "authors": "NVIDIA DLMED team",
-    "copyright": "Copyright (c) 2021-2022, NVIDIA CORPORATION",
+    "name": "Endoscopic In-Body Classification",
+    "task": "Endoscopic Frame Classification for In-Body vs Out-Body Detection",
+    "description": "A binary classification model based on SENet that distinguishes between inside-body and outside-body frames in endoscopic videos. The model processes 256x256 pixel RGB images and filters irrelevant frames, enabling automated procedure analysis.",
+    "authors": "MONAI team",
+    "copyright": "Copyright (c) MONAI Consortium",
     "data_source": "private dataset",
     "data_type": "RGB",
     "image_classes": "three channel data, intensity [0-255]",
@@ -52,6 +53,7 @@
     "eval_metrics": {
         "accuracy": 0.99
     },
+    "intended_use": "This is a research tool/prototype and not to be used clinically",
     "references": [
         "J. Hu, L. Shen and G. Sun, Squeeze-and-Excitation Networks, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018, pp. 7132-7141. https://arxiv.org/pdf/1709.01507.pdf"
     ],