SegVol: Universal and Interactive Volumetric Medical Image Segmentation

Precise image segmentation provides clinical study with instructive information. Despite the remarkable progress achieved in medical image segmentation, there is still an absence of a 3D foundation segmentation model that can segment a wide range of anatomical categories with easy user interaction. In this paper, we propose a 3D foundation segmentation model, named SegVol, supporting universal and interactive volumetric medical image segmentation. By scaling up training data to 90K unlabeled Computed Tomography (CT) volumes and 6K labeled CT volumes, this foundation model supports the segmentation of over 200 anatomical categories using semantic and spatial prompts. To facilitate efficient and precise inference on volumetric images, we design a zoom-out-zoom-in mechanism. Extensive experiments on 22 anatomical segmentation tasks verify that SegVol outperforms the competitors in 19 tasks, with improvements up to 37.24% compared to the runner-up methods. We demonstrate the effectiveness and importance of specific designs by ablation study. We expect this foundation model can promote the development of volumetric medical image analysis. The model and code are publicly available at: https://github.com/BAAI-DCAI/SegVol.

Paper

Similar papers

Peer review

Reviewer Ykum8/10 · confidence 4/52024-07-10

Summary

In this paper, the authors proposed SegVol, a 3D foundation segmentation model for medical images. This model supports universal and interactive volumetric medical image segmentation of more than 200 anatomical categories. This model is also well-designed with spatial and semantic prompts. In the inference stage, the authors designed a zoom-out-zoom-in mechanism to enable efficient and precise segmentation results Their extensive experiments show their method surpasses other SAM-like interactive segmentation methods.

Strengths

Interesting concepts. The idea of integrating spatial and semantic prompts is interesting and achieves prospective results. Paper clarity. The paper is overall well-written and structured. I also enjoyed the quality of the figures which help understanding the method. Good results. The method achieves SoTA results compared with the SAM-like interactive segmentation methods.

Weaknesses

I had a hard time finding weaknesses in the paper. Those I find are either nitpick or more directions for future work. I put the weaknesses in the questions section.

Questions

I would appreciate it if the authors could answer my questions. 1. Figure 2, the Violin plots seem confused and it is not cited in the paper. Can the authors explain more about this figure? 2. Will some of the 25 public volumetric medical segmentation datasets share the same data? Can the authors provide more detailed information about the datasets? 3. How do the authors process the collected data with inconsistent annotations? For example, different hospitals will have different annotations of the Aorta. Will this influence the results?

Rating

8

Confidence

4

Soundness

3

Presentation

4

Contribution

4

Limitations

The authors adequately addressed their limitations in the manuscript as well as the appendix.

Reviewer rczM7/10 · confidence 5/52024-07-11

Summary

This paper proposes a 3D foundation segmentation model, named SegVol, supporting universal and interactive volumetric medical image segmentation. By scaling up training data to 90K unlabeled Computed Tomography (CT) volumes and 6K labeled CT volumes, this foundation model supports the segmentation of over 200 anatomical categories using semantic and spatial prompts. Besides, a zoom-out-zoom-in mechanism is designed to facilitate efficient and precise inference on volumetric images.

Strengths

(1) Collect and process 25 public volumetric medical segmentation datasets, encompassing over 200 anatomical categories. The pseudo label is introduced to relieve the spurious correlation in the training data. (2) Implement massive 3D pre-training on 96K CT volumes and supervised fine-tuning on the 6k labeled datasets. (3) Support spatial-prompt, semantic-prompt, and combined-prompt segmentation, achieving high-precision segmentation and semantic disambiguation. (4) Design a zoom-out-zoom-in mechanism that significantly reduces the computational cost, meanwhile preserving precise segmentation.

Weaknesses

(1) As shown in Appendix Table 5, the numbers of trainset volumes is very dunbalanced, which will affect the performance of the proposed method. How the author address the above issue? They should give further analysis and validation. (2) For unseen anatomical categories, what is the segmentation effect of SegVol? Provide relevant experiments for further explanation. (3) What are the complexity、number of parameters and running speed of the SegVol, and provide a comparison with existing SOTA methods. (4) Minor: The reference format should be unified,for example,[8]、[43]

Questions

Answer the questions in Weakness Section

Rating

7

Confidence

5

Soundness

3

Presentation

3

Contribution

3

Limitations

Not Applicable

Reviewer 8irr8/10 · confidence 4/52024-07-24

Summary

The paper describes the utilization of SegVol, a deep learning model, to segment any organ/tumor/lesion on 3D CT data. **objectives** Create a universal segmentation model that can segment with : - any type of labeling - good performances on complex tasks - low computational cost (i.e., no sliding window) **contributions** The authors contribute to the state of the art in 3 ways : - first model for segmenting over 200 anatomical categories - support spatial (points/bbox) and semantic prompt (text describing each category) - introduction of a zoom-out-zoom-in mechanism to improve the model's efficiency and accuracy **datasets** - 25 open-source segmentation CT datasets + background filtering - generation of pseudo-labels with Felzenswalb-Huttenlocher to remove spurious correlation between the datasets **architecture** The whole architectures is divided into four blocks: - image encoder: 3D ViT pretrained using SimMIM on 96K unlabeled CTs + fine-tuning on 6K labeled CTs - text encoder: the CLIP model is used to generate a text embedding from the text prompt. The text prompt only consists in the name of the anatomical region to be segmented. - prompt encoder: text embedding and spatial prompt (bbox+points) are given to generate the prompt embedding. - mask decoder: from the prompt embedding, the segmentation is generated. **training algorithm** - the encoder is pretrained using SimMIM - the model is fine-tuned using an algorithm described in appendix B. For each epoch, an iteration of training is made using the ground truth, and a second one using the pseudo masks. **results** Experimental setup is given, as well as a description of the three external datasets used for the evaluation. - comparison with SAM-like methods : Table 2 shows the superiority of the method with Dice score > 0.7 on each task. Figure 2 helps to visualize the dice score distribution for each model, on each task. An additional comparison to standard fully supervised segmentation models such as nnUNet, SwinUNetR is provided. Superiority of SegVol is also shown compared to this model on Figure 8 and Table 6. - ablation studies - Zoom-out-zoom-in mechanism : Table 3 shows the interest of the method, improving performances and reducing duration compared to sliding window - Scaling up training data : Fig 3.(a) shows that the model performances improve as the number of training data increases. - Prompt impact : Fig 3 (a and b) shows that point prompt is inferior to text prompt, and best prompt combination is bbox+text - Case studies - Disambiguation via semantic-prompt: Figure 4 shows that when the spatial prompt is not sufficient to describe the segmenting task, the text prompt adds the complementary information to locate the good anatomical region. - spatial-prompt impact: Figure 5 shows how the points prompt are related to the results. When choosing two points, the category becomes clear, whereas with one point can identify two categories. **discussion** - Scalability : the authors believe that the model have a strong model scalability. - Generalizability : Generalization of SegVol to MRI is discussed, with an example given in appendix C showing a strong dice score of 80%. - Limitations : only limitation is that text prompt only consider a category, and not a full sentence with logical reasoning. - Broader impact : the authors believe that their method is universal and doesn't have any negative impact.

Strengths

- **originality** : the concept of SegVol is not original itself. However, the addition of the zoom-in-zoom-out principle and the combination of bbox + text + points prompts is new in this kind of models. - **quality** : the work is done rigorously, respecting a scientific reasoning all along the article. - **clarity** : the paper lacks clarity as there are two parts difficult to understand. - 1) The description of the training algorithm is not made in the text, only the steps are written in appendix B. - 2) The hyperparameters regarding the model's dimension (number of layer, ViT dimension) is not explicited. - **significance** : The results are impactful for the community of medical image segmentation, and could have plenty of applications. The property of segmenting multiple organs, from multiple entirely different datasets is remarkable.

Weaknesses

The only weakness regards the clarity of the training algorithm, which is hard to understand even with appendix B and training steps. This point could be improved by describing the steps in the appendix with a paragraph, or maybe with another diagram that follows the exact same steps.

Questions

**datasets** l. 83 : Why is Felzenswalb-Huttenlocher (FH) algorithm to generate pseudo masks ? l. 81 : How is it supposed to remove spurious correlation between datasets ? **architecture** l. 96 : The architecture principle is described, but some details are lacking such as the number of layers of each network, the number of heads in the ViT, etc. l. 133 : why is the computational cost reduced using the zoom-in-zoom-out principle compared to sliding window? clarify this point. l. 140 : how are prompts generated from the coarse segmentation masks? clarify if the prompts are generated thanks to the prompt encoder as a bbox, or points. **results** l. 169 : Could you indicate the training time (SimMIM and finetuning) of the model? l. 169 : Why did you choose these 3 specific datasets as external test sets? **discussion** l. 257 : Why is data salability a good thing ? Why is this part in the discussion? The data scalability of the model is not discussed or compared with other models. If not discussed, move to the conclusion.

Rating

8

Confidence

4

Soundness

4

Presentation

2

Contribution

4

Limitations

Only one limitation is addressed, considering the text prompt which cannot be a logical sentence at the moment. However, the amount of data required to train the model is not discussed. Would a few-shot finetuning method work on small datasets (10 samples)?

Reviewer 8irr2024-08-13

Thank you for your detailed answers to each of my interrogations. All the questions have been answered by the authors. Q1 : the algorithm is clearer with this small text, it should be included in the article. Q2 : the choice of FH is then arbitrary, it should be specified in the article. Q3 : For the network details, maybe a results of a torchinfo summary, or something in this idea could be added in appendix so that every hyperparameters could be seen visually (it would be be a bonus, but optional). Q9 : The results of joint finetuning compared to few-shot finetuning is relevent, and could be at least mentionned. I will maintain my grading, considering the clarity (presentation) will be improved, but is not yet done.

Authorsrebuttal2024-08-13

Thank you for the further feedback. We will polish the final version based on your suggestions.

Reviewer Ykum2024-08-12

I appreciate the author's detailed response. Most of my concerns have been addressed. I will maintain my original score.

Authorsrebuttal2024-08-13

Thank you for your feedback. We will polish the final version based on your suggestions.

Program Chairsdecision2024-09-25

Decision

Accept (spotlight)

© 2026 NYSGPT2525 LLC