Spaces:
Running
Running
Update README.md
Browse files
README.md
CHANGED
@@ -7,4 +7,31 @@ sdk: static
|
|
7 |
pinned: false
|
8 |
---
|
9 |
|
10 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
7 |
pinned: false
|
8 |
---
|
9 |
|
10 |
+
# Lighter zoo x CT-FM: Through lighter zoo we provide several models pre-trained using the CT-FM vision foundation model for Computed Tomography (CT) scans.
|
11 |
+
CT-FM is a large-scale 3D image-based pre-trained model designed for diverse radiological tasks. The model was pre-trained on 148,000 CT scans from the Imaging Data Commons using label-agnostic contrastive learning.
|
12 |
+
## Model Details
|
13 |
+
The model demonstrates strong capabilities across multiple tasks:
|
14 |
+
- Whole-body
|
15 |
+
- Heterogenous umor segmentation
|
16 |
+
- Head CT triage
|
17 |
+
- Medical image retrieval
|
18 |
+
- Semantic understanding of anatomical structures
|
19 |
+
Key features:
|
20 |
+
- Learns anatomical clustering without explicit labels
|
21 |
+
- Identifies similar anatomical structures across different scans
|
22 |
+
- Shows robustness in test-retest scenarios
|
23 |
+
- Provides interpretable salient regions in its embeddings
|
24 |
+
|
25 |
+
## Models Available
|
26 |
+
- Feature extractor `ct_fm_feature_extractor` which can be used for several feature-based tasks such as image retrieval, semantic search and outlier detection
|
27 |
+
- Fine-tuned whole body segmentation model `whole_body_segmentation` that segments 117 labels from the TotalSegmentator dataset
|
28 |
+
-
|
29 |
+
## Installation
|
30 |
+
We provide pre-trained as well as fine-tuned models in the `lighter-zoo` package that interfaces with HF to provide easy to use APIs
|
31 |
+
|
32 |
+
To install the `lighter-zoo` package, use pip:
|
33 |
+
```bash
|
34 |
+
pip install lighter-zoo
|
35 |
+
```
|
36 |
+
|
37 |
+
Inspect specific models to see how you can interact with these
|