Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Inquiry about MedSegDiff(v1) for Brain Tumor Segmentation on BraTS2021 Dataset #75

Open
MaggieLSY opened this issue Apr 6, 2023 · 6 comments

Comments

@MaggieLSY
Copy link

Hi, I am currently working on applying diffusion models for brain tumor segmentation, and I have some questions regarding your experiments on the BraTS2021 dataset.

  1. Your paper mentioned that you performed binary segmentation for brain tumors. Have you tried multi-class segmentation for different tumor sub-regions, such as enhancing tumor core, whole tumor, and tumor core?
  2. How did you calculate the Dice score for the final evaluation? Did you average it over all 2D slices, or did you concatenate them into patient-level 3D volumes?
  3. Did you use any post-processing steps, such as morphological operations or connected component analysis, to refine the segmentation results?

I would appreciate it if you could answer these questions or point me to some relevant resources. Thank you very much for your time and attention.

@smallboy-code
Copy link

can you solve the problem of the multi-class segmentation ?

@princerice
Copy link

嗨,我目前正在研究将扩散模型应用于脑肿瘤分割,我对您在 BraTS2021 数据集上的实验有一些疑问。

  1. 您的论文提到您对脑肿瘤进行了二元分割。您是否尝试过针对不同肿瘤亚区域的多类分割,例如增强肿瘤核心、整个肿瘤和肿瘤核心?
  2. 你是如何计算最终评估的骰子分数的?您是在所有 2D 切片上取平均值,还是将它们连接成患者级别的 3D 体积?
  3. 您是否使用任何后处理步骤(例如形态学操作或连接成分分析)来优化分割结果?

如果您能回答这些问题或向我指出一些相关资源,我将不胜感激。非常感谢您抽出宝贵时间关注。
Hi, I want to ask you when you run the V1 version, what is the dice you get? It would be much appreciated if you could tell me your batch number and step number

@Issues-translate-bot
Copy link

Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑‍🤝‍🧑👫🧑🏿‍🤝‍🧑🏻👩🏾‍🤝‍👨🏿👬🏿


Hi, I'm currently working on applying diffusion models to brain tumor segmentation and I have some questions about your experiments on the BraTS2021 dataset.

  1. Your paper mentions that you performed binary segmentation of brain tumors. Have you tried multi-class segmentation for different tumor subregions, such as enhanced tumor core, whole tumor, and tumor core?
  2. How do you calculate the final evaluated dice score? Do you average over all 2D slices, or concatenate them into a patient-level 3D volume?
  3. Do you use any post-processing steps (such as morphological operations or connected component analysis) to optimize the segmentation results?

I would be very grateful if you could answer these questions or point me to some relevant resources. Thank you very much for taking the time to follow.
Hi, I want to ask you when you run the V1 version, what is the dice you get? It would be much appreciated if you could tell me your batch number and step number

@princerice
Copy link

Hi, I want to ask you when you run the V1 version, what is the dice you get? It would be much appreciated if you could tell me your batch number and step number

Hi, I want to ask you when you run the V1 version, what is the dice you get? It would be much appreciated if you could tell me your batch number and step number

@smallboy-code
Copy link

@princerice Hi, Have you tried multi-class segmentation for different tumor sub-regions, such as enhanced tumor core, whole tumor, and tumor core? I'm working for it, can we discuss about it?

@princerice
Copy link

@princerice Hi, Have you tried multi-class segmentation for different tumor sub-regions, such as enhanced tumor core, whole tumor, and tumor core? I'm working for it, can we discuss about it?

Hello, I would like to ask you what is the effect index of V2 recurrence? Is the effect good? I'm trying to split multiple classes but binary classes are not working well

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants