Towards Ground-truth-free Evaluation of Any Segmentation in Medical Images
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866916408348114944 |
|---|---|
| author | Senbi, Ahjol Huang, Tianyu Lyu, Fei Li, Qing Tao, Yuhui Shao, Wei Chen, Qiang Wang, Chengyan Wang, Shuo Zhou, Tao Zhang, Yizhe |
| author_facet | Senbi, Ahjol Huang, Tianyu Lyu, Fei Li, Qing Tao, Yuhui Shao, Wei Chen, Qiang Wang, Chengyan Wang, Shuo Zhou, Tao Zhang, Yizhe |
| contents | We explore the feasibility and potential of building a ground-truth-free evaluation model to assess the quality of segmentations generated by the Segment Anything Model (SAM) and its variants in medical imaging. This evaluation model estimates segmentation quality scores by analyzing the coherence and consistency between the input images and their corresponding segmentation predictions. Based on prior research, we frame the task of training this model as a regression problem within a supervised learning framework, using Dice scores (and optionally other metrics) along with mean squared error to compute the training loss. The model is trained utilizing a large collection of public datasets of medical images with segmentation predictions from SAM and its variants. We name this model EvanySeg (Evaluation of Any Segmentation in Medical Images). Our exploration of convolution-based models (e.g., ResNet) and transformer-based models (e.g., ViT) suggested that ViT yields better performance for this task. EvanySeg can be employed for various tasks, including: (1) identifying poorly segmented samples by detecting low-percentile segmentation quality scores; (2) benchmarking segmentation models without ground truth by averaging quality scores across test samples; (3) alerting human experts to poor-quality segmentation predictions during human-AI collaboration by applying a threshold within the score space; and (4) selecting the best segmentation prediction for each test sample at test time when multiple segmentation models are available, by choosing the prediction with the highest quality score. Models and code will be made available at https://github.com/ahjolsenbics/EvanySeg. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_14874 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Towards Ground-truth-free Evaluation of Any Segmentation in Medical Images Senbi, Ahjol Huang, Tianyu Lyu, Fei Li, Qing Tao, Yuhui Shao, Wei Chen, Qiang Wang, Chengyan Wang, Shuo Zhou, Tao Zhang, Yizhe Image and Video Processing Artificial Intelligence Computer Vision and Pattern Recognition Machine Learning We explore the feasibility and potential of building a ground-truth-free evaluation model to assess the quality of segmentations generated by the Segment Anything Model (SAM) and its variants in medical imaging. This evaluation model estimates segmentation quality scores by analyzing the coherence and consistency between the input images and their corresponding segmentation predictions. Based on prior research, we frame the task of training this model as a regression problem within a supervised learning framework, using Dice scores (and optionally other metrics) along with mean squared error to compute the training loss. The model is trained utilizing a large collection of public datasets of medical images with segmentation predictions from SAM and its variants. We name this model EvanySeg (Evaluation of Any Segmentation in Medical Images). Our exploration of convolution-based models (e.g., ResNet) and transformer-based models (e.g., ViT) suggested that ViT yields better performance for this task. EvanySeg can be employed for various tasks, including: (1) identifying poorly segmented samples by detecting low-percentile segmentation quality scores; (2) benchmarking segmentation models without ground truth by averaging quality scores across test samples; (3) alerting human experts to poor-quality segmentation predictions during human-AI collaboration by applying a threshold within the score space; and (4) selecting the best segmentation prediction for each test sample at test time when multiple segmentation models are available, by choosing the prediction with the highest quality score. Models and code will be made available at https://github.com/ahjolsenbics/EvanySeg. |
| title | Towards Ground-truth-free Evaluation of Any Segmentation in Medical Images |
| topic | Image and Video Processing Artificial Intelligence Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2409.14874 |