small changes in references and hours used for training
05b4bce verified 25 days ago
24.3 kB
---
license: cc-by-nc-sa-4.0
language:
- en
pipeline_tag: image-segmentation
tags:
- nnU-Net
- AAA
- AbdominalAorta
- EVAR
---
# PrePostEVAR_CTA_Annotation

This repository contains trained 3D nnU-Net models for automated analysis of pre- and post-endovascular aneurysm repair (EVAR) computed tomography angiography (CTA).

# Table of Contents

- [Model Details](#model-details)
- [Uses](#uses)
- [Bias, Risks, and Limitations](#bias-risks-and-limitations)
- [Training Details](#training-details)
- [Evaluation](#evaluation)
- [Environmental Impact](#environmental-impact)
- [Citation](#citation)


# Model Details

## Model Description

The models weights provided are used in a 3D U-Net with the automatic dataset-based configuration provided by nnU-Net framework. Four independent tasks are provided, covering two main categories:

Anatomical segmentation
Segmentation-based landmark localization

Separate models were trained for pre-EVAR and post-EVAR CTA because these imaging scenarios differ substantially in anatomy, appearance, and target structures.

Each task was trained using five-fold cross-validation, resulting in five independently trained models per task and 20 trained models in total. The repository contains the checkpoint_best.pth weights corresponding to the best validation performance for each fold.
| Repository folder                     | Task                    | Input         | Output                                                     |
|---------------------------------------|-------------------------|---------------|------------------------------------------------------------|
| `Dataset100_SegmentationPreEVAR`      | Anatomical segmentation | Pre-EVAR CTA  | Aortic lumen and intraluminal thrombus                     |
| `Dataset101_LandmarksPreEVAR`         | Landmark localization   | Pre-EVAR CTA  | Abdominal aorta + 6 anatomical landmarks                   |
| `Dataset200_SegmentationPostEVAR`     | Anatomical segmentation | Post-EVAR CTA | Aortic lumen, excluded aorta and endoleak                  |
| `Dataset201_LandmarksPostEVAR`        | Landmark localization   | Post-EVAR CTA | Abdominal aorta + 9 anatomical/endograft-related landmarks |

Each model directory contains five fold-specific weights:

```text
DatasetXXX_TaskName/
└── nnUNetTrainer__nnUNetPlans__3d_fullres/
    ├── dataset.json
    ├── plans.json
    ├── fold_0/
    │   └── checkpoint_best.pth
    ├── fold_1/
    │   └── checkpoint_best.pth
    ├── fold_2/
    │   └── checkpoint_best.pth
    ├── fold_3/
    │   └── checkpoint_best.pth
    └── fold_4/
        └── checkpoint_best.pth
```

## Anatomical segmentation

### Pre-EVAR segmentation

**Model folder:** `Dataset100_SegmentationPreEVAR`

The pre-EVAR models were trained to segment:

| Label | Structure             |
|-------|-----------------------|
| 0     | Background            |
| 1     | Aortic lumen          |
| 2     | Intraluminal thrombus |

Voxel-wise predictions were converted into discrete segmentation masks by assigning each voxel to the class with the highest predicted probability.

#### Configuration

| Parameter         | Value                  |
|-------------------|------------------------|
| Target spacing    | 0.79 × 0.70 × 0.79 mm  |
| Median image size | 512 × 640 × 512 voxels |
| Patch size        | 128 × 160 × 112 voxels |

### Post-EVAR segmentation

**Model folder:** `Dataset200_SegmentationPostEVAR`

The post-EVAR models were trained to segment:

| Label | Structure      |
|-------|----------------|
| 0     | Background     |
| 1     | Aortic lumen   |
| 2     | Excluded aorta |
| 3     | Endoleak (EL)  |

The **excluded aorta** was defined as the aortic volume outside the stent-graft lumen, excluding the stent graft itself.

Voxel-wise predictions were converted into discrete segmentation masks by assigning each voxel to the class with the highest predicted probability.

#### Endoleak detection

Endoleak detection was derived directly from the post-EVAR anatomical segmentation models.

Unlike the other anatomical structures, the endoleak output was retained as a **voxel-wise probability map**. Binary endoleak masks were generated using the following threshold:

```text
P(EL) ≥ 0.5
```

After thresholding, three-dimensional connected components containing fewer than **10 voxels** were removed to reduce isolated false-positive predictions.

#### Configuration

| Parameter         | Value                  |
|-------------------|------------------------|
| Target spacing    | 0.77 × 0.70 × 0.77 mm  |
| Median image size | 512 × 591 × 512 voxels |
| Patch size        | 128 × 128 × 128 voxels |

---

## Landmark localization

Landmark localization was formulated as a **segmentation task** rather than direct coordinate regression.

Each manually annotated landmark coordinate was converted into a small cubic binary mask with an edge length of approximately **3 mm**.

Because the masks were defined in voxel space and CTA voxel spacing varied between scans, their exact physical dimensions could vary slightly after rounding to the nearest voxel.

This formulation enabled the use of the standard nnU-Net segmentation pipeline while reducing the extreme class imbalance that would result from representing each landmark as a single voxel.

After prediction:

1. Voxel-wise predictions were converted into discrete masks by assigning each voxel to the class with the highest predicted probability.
2. The final coordinate of each landmark was calculated as the **centroid of the predicted mask** corresponding to that landmark class.

### Pre-EVAR landmark localization

**Model folder:** `Dataset101_LandmarksPreEVAR`

The model predicts an abdominal aorta class together with six anatomical landmark classes:

| Label | Structure / landmark                         |
|-------|----------------------------------------------|
| 0     | Background                                   |
| 1     | Abdominal aorta                              |
| 2     | Celiac trunk                                 |
| 3     | Superior mesenteric artery                   |
| 4     | Renal arteries                               |
| 5     | Start of the abdominal aortic aneurysm (AAA) |
| 6     | Aortic bifurcation                           |
| 7     | End of the AAA                               |

#### Configuration

| Parameter         | Value                  |
|-------------------|------------------------|
| Target spacing    | 0.79 × 0.70 × 0.79 mm  |
| Median image size | 145 × 372 × 154 voxels |
| Patch size        | 96 × 256 × 96 voxels   |

### Post-EVAR landmark localization

**Model folder:** `Dataset201_LandmarksPostEVAR`

The model predicts an abdominal aorta class together with nine anatomical and endograft-related landmark classes:

| Label | Structure / landmark                         |
|-------|----------------------------------------------|
| 0     | Background                                   |
| 1     | Abdominal aorta                              |
| 2     | Celiac trunk                                 |
| 3     | Superior mesenteric artery                   |
| 4     | Renal arteries                               |
| 5     | Start of the abdominal aortic aneurysm (AAA) |
| 6     | Aortic bifurcation                           |
| 7     | End of the AAA                               |
| 8     | Proximal prosthesis landing                  |
| 9     | Proximal prosthesis fabric component         |
| 10    | Distal prosthesis landing                    |

#### Configuration

| Parameter         | Value                  |
|-------------------|------------------------|
| Target spacing    | 0.77 × 0.70 × 0.77 mm  |
| Median image size | 142 × 364 × 158 voxels |
| Patch size        | 96 × 256 × 96 voxels   |

---

## Dataset fingerprints

The main task-specific configurations automatically determined by nnU-Net are summarized below.

| Parameter                  | Pre-EVAR Segmentation          | Pre-EVAR Landmarks          | Post-EVAR Segmentation          | Post-EVAR Landmarks          |
|----------------------------|--------------------------------|-----------------------------|---------------------------------|------------------------------|
| Dataset                    | `Dataset100_SegmentationPreEVAR` | `Dataset101_LandmarksPreEVAR` | `Dataset200_SegmentationPostEVAR` | `Dataset201_LandmarksPostEVAR` |
| Target spacing (mm)        | 0.79 × 0.70 × 0.79             | 0.79 × 0.70 × 0.79          | 0.77 × 0.70 × 0.77              | 0.77 × 0.70 × 0.77           |
| Median image size (voxels) | 512 × 640 × 512                | 145 × 372 × 154             | 512 × 591 × 512                 | 142 × 364 × 158              |
| Patch size (voxels)        | 128 × 160 × 112                | 96 × 256 × 96               | 128 × 128 × 128                 | 96 × 256 × 96                |
| Foreground classes         | 2                              | 7                           | 3                               | 10                           |


* **Developed by:**  
Cardiovascular Image Unit | Cardiovascular diseases research team at Vall d'Hebron Institut de Recerca (VHIR)

* **Funded by:**  
This study was funded by the "Red de Investigación Vascular" of the "Sociedad Española de Angiología y Cirugía Vascular", the Instituto de Salud Carlos III (CP24/00121, co-funded by the European Union), the Spanish Ministry of Science, Innovation and Universities (FORT23/00034) and the Biomedical Research Networking Center on Cardiovascular Diseases (CIBERCV). Juan Garrido-Oliver has received funding from Secretaria d'Universitats i Recerca del Departament de Recerca i Universitats de la Generalitat de Catalunya i del Fons Europeu Social Plus (AGAUR-FI 2023 FI-1 00322 Joan Oró). 
* **Model type:**  
Deep Learning model (CNN)

* **License:**  
cc-by-nc-sa-4.0

# Uses

## Direct Use

Follow the official instructions to install the [**nnU-Net framework**](https://github.com/MIC-DKFZ/nnUNet)  

+
Once nnU-Net is installed, make sure you have created the `nnUNet_results` directory and assigned it to the corresponding environment variable as required by nnU-Net.

Then, copy the folder **DatasetXXX_YYYYY** into the `nnUNet_results` directory.
This folder contains the **trained model weights** for **abdominal aorta segmentation** and **landmarks annotation** in **pre- and post-EVAR contrast-enhanced computed tomography angiography (CTA)** images.

For inference, you need to specify a folder with the images in the proper nnU-Net format (see in their guide). 

```
CUDA_VISIBLE_DEVICES=0 nnUNetv2_predict -i INPUT_FOLDER -o OUTPUT_FOLDER -d XXX -c 3d_fullres -chk checkpoint_best.pth
```
Remember, XXX defines the models you are using (i.e. 100 for pre-EVAR segmentation, 101 for pre-EVAR landmark annotation, 200 for post-EVAR segmentation, and 201 for post-EVAR landmark annotation)

## Out-of-Scope Use

Use it with CTA images that include the abdominal aorta. It is not recommended to use it with whole-body images (unless your GPU and RAM are large) or other modalities. 

# Bias, Risks, and Limitations

This model was trained using images from patients with abdominal aortic aneurysms indicated for surgery or post-EVAR CTAs. No healthy subjects were included in training, therefore there might be a bias towards better segmentation in patients with aneurysms.


# Training Details

One hundred and thirty-five pre-EVAR CTAs and 160 post-EVAR CTAs from 135 patients from one center were used in training. 

## Network architecture

All models follow the general **3D U-Net architecture automatically configured by nnU-Net**, with task-specific parameters adapted to the corresponding dataset fingerprint.

Each resolution stage contains two convolutional blocks composed of:

- 3 × 3 × 3 convolutions
- Instance normalization
- Leaky ReLU activation

The encoder uses strided convolutions for downsampling across **six resolution stages**, while the decoder uses **five transposed-convolution upsampling stages** to progressively recover the spatial resolution.

## Training configuration

All four tasks were trained independently using the nnU-Net training framework. 

| Parameter              | Value                             |
|------------------------|-----------------------------------|
| Architecture           | 3D U-Net                          |
| Optimizer              | Stochastic gradient descent (SGD) |
| Momentum               | Nesterov momentum = 0.99          |
| Loss function          | Dice + Cross-entropy              |
| Batch size             | 2 patches                         |
| Initial learning rate  | 0.01                              |
| Learning-rate schedule | Polynomial decay                  |
| Training length        | 1000 epochs                       |
| Iterations per epoch   | 250                               |
| Cross-validation       | 5 folds                           |


# Evaluation

Evaluation is described in the research article Garrido-Oliver et al., 2026, Eur J Vasc Endovasc Surg (see references).

## Testing Data, Factors & Metrics
Aortic segmentations and landmarks were evaluated using the cross-validation data from one center and an external validation dataset from another center that included data from 30 patients including 30 pre-EVAR CTAs and 40 post-EVAR CTAs. 

### Testing Data

Thirty CTA images were annotated by two different endovascular surgeons using 3DSlicer.

### Metrics

Segmentation performance was evaluated using Dice score (DS), an overlap volume-based metric, and Hausdorff distance 95th percentile (HD), a surface distance-based metric. Automatic segmentations were compared with each observer annotation and a further test was designed to identify whether the availability of an DL segmentation as a starting point improved the inter-observer agreement of manual segmentation.

Landmark annotation was evaluated using the Euclidean distance between manual annotations and model's predictions. Also, the absolute axial distance and the signed axial distance were computed to identify bias toward craneo-caudal direction. 

## Results
**Table 1.** Results of manual interobserver variability and DL accuracy in segmentation for the center 1 and center 2 CTAs.  
\* Significant difference (p<0.05) between interobserver variability and DL errors.  
† Significant difference (p<0.05) between DL error in C1 and DL error in C2 data.  
DS: Dice score. HD95: 95th percentile of the Hausdorff distance in mm. C1: Training set (cross-validation). C2: External validation set. 

| Structure | Interobserver variability DS | Interobserver variability HD95 | DL vs. manual C1 DS | DL vs. manual C1 HD95 | DL vs. manual C2 DS | DL vs. manual C2 HD95 |
|-----------|------------------------------|--------------------------------|----------------------|------------------------|----------------------|------------------------|
| **Pre-EVAR CTA (N)** | 30 |  | 135 |  | 30 |  |
| **Aortic lumen** | 0.955<br>[0.935, 0.970] | 1.89<br>[1.76, 2.07] | 0.969<br>[0.952, 0.978]\* | 1.73<br>[1.58, 1.98] | 0.955<br>[0.938, 0.967]† | 1.93<br>[1.74, 2.19]† |
| **Intraluminal thrombus** | 0.902<br>[0.861, 0.931] | 3.10<br>[2.63, 8.32] | 0.922<br>[0.888, 0.947] | 2.62<br>[1.99, 3.97]\* | 0.907<br>[0.879, 0.934] | 3.02<br>[2.42, 4.23] |
| **Whole aorta** | 0.950<br>[0.944, 0.967] | 2.47<br>[2.21, 2.97] | 0.962<br>[0.948, 0.974] | 2.34<br>[2.02, 2.83] | 0.953<br>[0.939, 0.962]† | 2.40<br>[2.10, 2.93] |
| **Post-EVAR CTA (N)** | 30 |  | 160 |  | 40 |  |
| **Aortic lumen** | 0.879<br>[0.868, 0.933] | 2.12<br>[1.82, 3.02] | 0.935<br>[0.908, 0.954]\* | 1.73<br>[1.58, 2.25]\* | 0.914<br>[0.875, 0.930]† | 2.14<br>[1.79, 3.06] |
| **Excluded aorta** | 0.926<br>[0.913, 0.934] | 3.36<br>[2.40, 4.43] | 0.940<br>[0.916, 0.953] | 2.58<br>[2.17, 3.70]\* | 0.934<br>[0.911, 0.943] | 2.64<br>[2.39, 3.66] |
| **Whole aorta** | 0.918<br>[0.890, 0.932] | 4.11<br>[3.21, 5.51] | 0.938<br>[0.912, 0.950]\* | 3.09<br>[2.48, 15.13] | 0.923<br>[0.905, 0.940]† | 3.38<br>[2.88, 4.99] |

**Table 2.** Euclidean distance (mm), represented as median [1st-3rd] quartiles, between landmarks annotated by two observers, and DL vs manual reference in both centers.  
\* Significant difference (p<0.05) between interobserver difference and DL errors. No significant differences were found in DL error between C1 and C2 data.  
SMA: superior mesenteric artery; AAA: abdominal aortic aneurysm. C1: Training set (cross-validation). C2: External validation set. 

| Landmark | Pre-EVAR Interobserver variability | Pre-EVAR DL vs manual C1 | Pre-EVAR DL vs manual C2 | Post-EVAR Interobserver variability | Post-EVAR DL vs manual C1 | Post-EVAR DL vs manual C2 |
|----------|------------------------------------|--------------------------|--------------------------|-------------------------------------|---------------------------|---------------------------|
| **Celiac trunk** | 3.35<br>[2.03, 4.14] | 2.38<br>[1.88, 3.29]\* | 2.01<br>[1.63, 3.90] | 2.43<br>[2.02, 3.97] | 2.52<br>[1.76, 3.54] | 3.57<br>[1.73, 11.18] |
| **SMA** | 2.87<br>[2.23, 3.96] | 2.94<br>[2.13, 4.10] | 2.80<br>[2.06, 4.38] | 3.05<br>[2.48, 4.00] | 2.89<br>[2.00, 4.14] | 2.88<br>[1.72, 4.07] |
| **Renal arteries** | 2.03<br>[1.35, 3.01] | 2.44<br>[1.88, 3.33] | 2.55<br>[2.26, 3.45] | 1.93<br>[1.45, 2.97] | 2.49<br>[1.97, 4.22]\* | 2.84<br>[1.84, 5.52]\* |
| **Proximal origin AAA** | 7.13<br>[4.76, 10.90] | 5.82<br>[3.98, 9.15] | 4.53<br>[3.55, 6.15]\* | 5.37<br>[3.78, 10.82] | 4.90<br>[2.84, 8.22] | 4.54<br>[3.19, 5.73] |
| **Aortic bifurcation** | 3.89<br>[2.25, 6.66] | 5.90<br>[4.11, 7.60]\* | 6.74<br>[5.37, 8.27]\* | 3.79<br>[2.27, 5.75] | 6.98<br>[5.38, 8.85]\* | 7.33<br>[5.97, 8.94]\* |
| **Distal end AAA** | 9.13<br>[4.56, 13.03] | 9.64<br>[4.17, 24.31] | 12.78<br>[7.27, 30.32] | 8.33<br>[5.77, 14.14] | 10.20<br>[5.11, 23.89] | 8.33<br>[3.74, 12.86] |
| **Prox. land. EVAR** | - | - | - | 6.18<br>[4.33, 8.84] | 4.25<br>[2.83, 6.08]\* | 3.70<br>[2.55, 7.11]\* |
| **Fabric EVAR** | - | - | - | 4.01<br>[2.57, 6.06] | 4.27<br>[2.81, 6.98] | 4.31<br>[2.32, 14.87] |
| **Distal land. EVAR** | - | - | - | 4.04<br>[2.49, 6.64] | 11.21<br>[3.68, 28.37]\* | 14.27<br>[3.76, 24.84]\* |
| **All landmarks** | 3.61<br>[2.19, 6.74] | 3.69<br>[2.31, 6.87] | 4.21<br>[2.33, 6.70] | 3.87<br>[2.34, 7.11] | 4.26<br>[2.48, 7.80] | 4.34<br>[2.31, 9.41] |

**Table 3.** Absolute distance in the axial direction (mm), represented as median [1st-3rd] quartiles, between landmarks annotated by two observers, and DL vs manual reference in both centers.  
\* Significant difference (p<0.05) between interobserver difference and DL errors.  
† Significant difference (p<0.05) between DL error in C1 and DL error in C2 data.  
SMA: superior mesenteric artery; AAA: abdominal aortic aneurysm. C1: Training set (cross-validation). C2: External validation set.

| Landmark | Pre-EVAR Interobserver variability | Pre-EVAR DL vs. manual C1 | Pre-EVAR DL vs. manual C2 | Post-EVAR Interobserver variability | Post-EVAR DL vs. manual C1 | Post-EVAR DL vs. manual C2 |
|----------|------------------------------------|---------------------------|---------------------------|-------------------------------------|----------------------------|----------------------------|
| **Celiac trunk** | 1.41<br>[0.70, 2.50] | 1.41<br>[0.69, 2.31] | 1.75<br>[1.26, 3.57] | 0.75<br>[0.26, 1.73] | 1.28<br>[0.43, 2.74]† | 2.62<br>[0.83, 8.65]\*† |
| **SMA** | 1.45<br>[0.79, 2.76] | 1.76<br>[0.68, 2.85] | 1.81<br>[1.12, 2.86] | 1.25<br>[0.54, 2.20] | 1.64<br>[0.70, 2.99] | 1.94<br>[0.92, 3.69]\* |
| **Renal arteries** | 0.89<br>[0.36, 1.58] | 1.68<br>[0.85, 2.46]\* | 1.80<br>[0.82, 2.69]\* | 0.87<br>[0.54, 1.78] | 1.88<br>[0.85, 3.20]\* | 2.31<br>[0.96, 4.19]\* |
| **Proximal origin AAA** | 2.80<br>[1.86, 7.08] | 3.63<br>[1.91, 7.04] | 3.79<br>[2.24, 5.18] | 2.95<br>[1.76, 5.77] | 3.10<br>[1.38, 6.65] | 2.60<br>[1.44, 5.08] |
| **Aortic bifurcation** | 1.91<br>[1.36, 2.91] | 2.83<br>[1.10, 4.15] | 4.05<br>[2.01, 6.67]\*† | 2.59<br>[1.30, 4.33] | 2.74<br>[1.21, 4.69] | 3.41<br>[1.61, 5.36] |
| **Distal end AAA** | 4.90<br>[2.28, 9.49] | 6.17<br>[2.21, 13.51] | 7.11<br>[4.56, 13.36] | 5.27<br>[3.99, 8.33] | 6.52<br>[3.00, 15.81] | 4.87<br>[1.92, 8.56] |
| **Prox. land. EVAR** | — | — | — | 4.92<br>[3.42, 6.24] | 3.09<br>[1.39, 4.95]\* | 2.89<br>[2.13, 5.76] |
| **Fabric EVAR** | — | — | — | 2.70<br>[1.10, 3.94] | 2.89<br>[1.09, 5.26] | 3.08<br>[1.19, 10.49] |
| **Distal land. EVAR** | — | — | — | 1.83<br>[0.72, 3.10] | 2.43<br>[0.99, 5.21] | 2.48<br>[0.97, 6.30] |
| **All landmarks** | 1.87<br>[0.90, 3.51] | 2.21<br>[1.01, 4.08] | 2.70<br>[1.39, 4.96]\* | 2.20<br>[0.82, 4.49] | 2.46<br>[1.01, 4.77] | 2.74<br>[1.16, 5.96]\* |

**Table 4.** Distance (mm) in axial view, represented as median [1st-3rd] quartiles, between landmarks annotated by two observers, and DL vs manual reference in both centers. Positive differences are a shift of DL predictions toward foot direction; negative differences are a shift of the DL predictions toward cranial direction.  
\* Significant difference (p<0.05) between interobserver difference and DL errors.  
† Significant difference (p<0.05) between DL error in C1 and DL error in C2 data.  
SMA: superior mesenteric artery; AAA: abdominal aortic aneurysm. C1: Training set (cross-validation). C2: External validation set.

| Landmark | Pre-EVAR Interobserver variability | Pre-EVAR DL vs. manual C1 | Pre-EVAR DL vs. manual C2 | Post-EVAR Interobserver variability | Post-EVAR DL vs. manual C1 | Post-EVAR DL vs. manual C2 |
|----------|------------------------------------|---------------------------|---------------------------|-------------------------------------|----------------------------|----------------------------|
| **Celiac trunk** | -0.20<br>[-1.99, 0.79] | 0.24<br>[-1.34, 1.56]† | 1.61<br>[0.21, 3.18]\*† | -0.06<br>[-1.60, 0.27] | 0.10<br>[-1.04, 1.47] | 0.78<br>[-0.70, 2.95]\* |
| **SMA** | -0.71<br>[-2.35, 1.09] | 0.21<br>[-1.45, 2.07] | 1.61<br>[0.67, 2.83]\*† | -0.15<br>[-1.70, 0.86] | 0.06<br>[-1.37, 1.66] | 1.30<br>[0.12, 2.39]\*† |
| **Renal arteries** | -0.47<br>[-1.23, 0.34] | 0.58<br>[-1.32, 1.83]\* | 1.55<br>[0.36, 2.54]\*† | -0.53<br>[-1.30, 0.53] | 0.56<br>[-1.08, 2.15]\* | 1.62<br>[0.56, 3.07]\*† |
| **Proximal origin AAA** | -1.63<br>[-5.27, 1.76] | 1.78<br>[-2.03, 5.02]\*† | 3.79<br>[2.18, 5.18]\*† | -2.17<br>[-3.80, 0.36] | 1.21<br>[-1.61, 4.30]\* | 1.50<br>[0.24, 4.48]\* |
| **Aortic bifurcation** | 1.91<br>[1.36, 2.91] | -2.47<br>[-3.90, -0.37]\*† | -3.69<br>[-6.61, 1.11]\*† | 1.90<br>[0.07, 4.33] | -2.10<br>[-4.23, 0.00]\* | -3.07<br>[-5.19, 0.15]\* |
| **Distal end AAA** | -4.55<br>[-9.49, -1.65] | 0.87<br>[-3.97, 6.47]\* | 2.42<br>[-3.93, 12.85]\* | -5.27<br>[-8.33, -3.69] | 3.05<br>[-2.42, 10.31]\* | 3.52<br>[-1.14, 7.73]\* |
| **Prox. land. EVAR** | — | — | — | -4.92<br>[-6.24, -2.19] | 0.96<br>[-1.80, 3.28]\* | 2.81<br>[1.44, 4.95]\*† |
| **Fabric EVAR** | — | — | — | -2.49<br>[-3.75, -0.36] | 1.06<br>[-1.16, 4.26]\* | 2.56<br>[0.07, 10.12]\* |
| **Distal land. EVAR** | — | — | — | 0.37<br>[-1.14, 2.00] | -0.14<br>[-2.27, 2.60] | -0.70<br>[-3.18, 1.65] |
| **All landmarks** | -0.30<br>[-2.72, 1.36] | 0.09<br>[-2.15, 2.22]\* | 1.56<br>[-0.82, 3.68]\*† | -0.91<br>[-3.72, 0.69] | 0.35<br>[-2.08, 2.80]\* | 1.21<br>[-1.07, 3.81]\*† |
# Environmental Impact
* **Hardware Type:**  
Training and validation of these models was done in a HPC node with two RTX 3090, 128 GB of VRAM

* **Hours used:**  
10 days for all models. 

---

# References 

Garrido-Oliver, J., Tenezaca-Sari, X., Ribeiro, T. et al, Comprehensive Artificial Intelligence Based Analysis of Pre- and Postoperative Computed Tomography Scans in Patients Undergoing Endovascular Abdominal Aortic Repair. Eur J Vasc Endovasc Surg (2026). https://doi.org/10.1016/j.ejvs.2026.09.035.

Isensee, F., Jaeger, P.F., Kohl, S.A.A. et al. nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation. Nat Methods 18, 203–211 (2021). https://doi.org/10.1038/s41592-020-01008-z