{ "id": 273671, "title": "Research papers about classification of MGMT status from MRI", "url": "/competitions/rsna-miccai-brain-tumor-radiogenomic-classification/discussion/273671", "author_name": "", "post_date": "2021-09-22T04:21:17.480949500Z", "votes": 7, "comment_count": 2, "views": 0, "content": "

Just sharing a list of research papers that have tackled this issue in the past:

\n
    \n
  1. https://aapm.onlinelibrary.wiley.com/doi/epdf/10.1118/1.4948668 (from the abstract:
    \n\"The best classification system (an SVM-based classifier) had a maximum area under the receiver-operating characteristic (ROC) curve of 0.85 (95% CI: 0.78–0.91) using four texture features(correlation, energy, entropy, and local intensity) originating from the T2-weighted images, yielding at the optimal threshold of the ROC curve, a sensitivity of 0.803 and a specificity of 0.813\"
    2. \n
  2. http://www.ajnr.org/content/early/2021/03/04/ajnr.A7029 (Deep learning based)
    \n\"We demonstrate high classification accuracy in predicting MGMT promoter methylation status using only T2WI\"
    3. \n
  3. https://pubmed.ncbi.nlm.nih.gov/29218894/ (CRNN; Deep Learning).
    4. \n
  4. https://www.nature.com/articles/s41598-019-50849-y
    5. \n
  5. Both papers 1 and 2 got best results on T2w images. this was suprising to me as I expected better results on T1wCE images.
    6. \n
", "messages": [ { "id": "1519959", "postDate": "09/22/2021 04:21:17", "content": "

Just sharing a list of research papers that have tackled this issue in the past:

\n
    \n
  1. https://aapm.onlinelibrary.wiley.com/doi/epdf/10.1118/1.4948668 (from the abstract:
    \n\"The best classification system (an SVM-based classifier) had a maximum area under the receiver-operating characteristic (ROC) curve of 0.85 (95% CI: 0.78–0.91) using four texture features(correlation, energy, entropy, and local intensity) originating from the T2-weighted images, yielding at the optimal threshold of the ROC curve, a sensitivity of 0.803 and a specificity of 0.813\"
    2. \n
  2. http://www.ajnr.org/content/early/2021/03/04/ajnr.A7029 (Deep learning based)
    \n\"We demonstrate high classification accuracy in predicting MGMT promoter methylation status using only T2WI\"
    3. \n
  3. https://pubmed.ncbi.nlm.nih.gov/29218894/ (CRNN; Deep Learning).
    4. \n
  4. https://www.nature.com/articles/s41598-019-50849-y
    5. \n
  5. Both papers 1 and 2 got best results on T2w images. this was suprising to me as I expected better results on T1wCE images.
    6. \n
", "rawMarkdown": "Just sharing a list of research papers that have tackled this issue in the past:\n1. https://aapm.onlinelibrary.wiley.com/doi/epdf/10.1118/1.4948668 (from the abstract:\n\"The best classification system (an SVM-based classifier) had a maximum **area under the receiver-operating characteristic (ROC) curve of 0.85 (95% CI: 0.78–0.91)** using four texture features(correlation, energy, entropy, and local intensity) originating from the T2-weighted images, yielding at the optimal threshold of the ROC curve, a sensitivity of 0.803 and a specificity of 0.813\"\n2. http://www.ajnr.org/content/early/2021/03/04/ajnr.A7029 (Deep learning based)\n\"We demonstrate high classification accuracy in predicting MGMT promoter methylation status using only T2WI\"\n3. https://pubmed.ncbi.nlm.nih.gov/29218894/ (CRNN; Deep Learning). \n4. https://www.nature.com/articles/s41598-019-50849-y \n5. Both papers 1 and 2 got best results on T2w images. this was suprising to me as I expected better results on T1wCE images.", "votes": null }, { "id": "1522726", "postDate": "09/24/2021 13:58:41", "content": "

Thanks for sharing. I am also getting better results with the T2W type only.

", "rawMarkdown": "Thanks for sharing. I am also getting better results with the T2W type only.", "votes": null }, { "id": "1523752", "postDate": "09/25/2021 17:13:11", "content": "

Thanks for your sharing

", "rawMarkdown": "Thanks for your sharing", "votes": null } ], "comments": [ { "id": 1522726, "author_name": "trinayanbharadwaj", "author_url": "", "post_date": "09/24/2021 13:58:41", "content": "

Thanks for sharing. I am also getting better results with the T2W type only.

", "votes": null, "replies": [] }, { "id": 1523752, "author_name": "minhvnnguyn", "author_url": "", "post_date": "09/25/2021 17:13:11", "content": "

Thanks for your sharing

", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "1519959": "Just sharing a list of research papers that have tackled this issue in the past:\n1. https://aapm.onlinelibrary.wiley.com/doi/epdf/10.1118/1.4948668 (from the abstract:\n\"The best classification system (an SVM-based classifier) had a maximum **area under the receiver-operating characteristic (ROC) curve of 0.85 (95% CI: 0.78–0.91)** using four texture features(correlation, energy, entropy, and local intensity) originating from the T2-weighted images, yielding at the optimal threshold of the ROC curve, a sensitivity of 0.803 and a specificity of 0.813\"\n2. http://www.ajnr.org/content/early/2021/03/04/ajnr.A7029 (Deep learning based)\n\"We demonstrate high classification accuracy in predicting MGMT promoter methylation status using only T2WI\"\n3. https://pubmed.ncbi.nlm.nih.gov/29218894/ (CRNN; Deep Learning). \n4. https://www.nature.com/articles/s41598-019-50849-y \n5. Both papers 1 and 2 got best results on T2w images. this was suprising to me as I expected better results on T1wCE images.", "1522726": "Thanks for sharing. I am also getting better results with the T2W type only.", "1523752": "Thanks for your sharing" }, "source": "meta" }