{
  "id": 253451,
  "title": "Automatic Prediction of MGMT Status in Glioblastoma",
  "url": "/competitions/rsna-miccai-brain-tumor-radiogenomic-classification/discussion/253451",
  "author_name": "",
  "post_date": "2021-07-16T16:55:16.242996500Z",
  "votes": 15,
  "comment_count": 3,
  "views": 0,
  "content": "<p>Below is the abstract from a paper discusses a DL pipeline for predicting Methylation of the O6-methylguanine methyltransferase (MGMT) - I found this paper very informative.  The link to the paper is below that…</p>\n<p><strong>Abstract</strong><br>\nMethylation of the O6-methylguanine methyltransferase (MGMT) gene promoter is correlated with the effectiveness of the current standard of care in glioblastoma patients. In this study, a deep learning pipeline is designed for automatic prediction of MGMT status in 87 glioblastoma patients with contrast-enhanced T1W images and 66 with fluid-attenuated inversion recovery(FLAIR) images. The end-to-end pipeline completes both tumor segmentation and status classification. The better tumor segmentation performance comes from FLAIR images (Dice score, ) compared to contrast-enhanced T1WI (Dice score, ), and the better status prediction is also from the FLAIR images (accuracy, ; recall, ; precision, ; and  score, ). This proposed pipeline not only saves the time in tumor annotation and avoids interrater variability in glioma segmentation but also achieves good prediction of MGMT methylation status. It would help find molecular biomarkers from routine medical images and further facilitate treatment planning.</p>\n<p><a href=\"https://www.hindawi.com/journals/bmri/2020/9258649/\" target=\"_blank\">https://www.hindawi.com/journals/bmri/2020/9258649/</a></p>",
  "messages": [
    {
      "id": "1390454",
      "postDate": "07/16/2021 16:55:16",
      "content": "<p>Below is the abstract from a paper discusses a DL pipeline for predicting Methylation of the O6-methylguanine methyltransferase (MGMT) - I found this paper very informative.  The link to the paper is below that…</p>\n<p><strong>Abstract</strong><br>\nMethylation of the O6-methylguanine methyltransferase (MGMT) gene promoter is correlated with the effectiveness of the current standard of care in glioblastoma patients. In this study, a deep learning pipeline is designed for automatic prediction of MGMT status in 87 glioblastoma patients with contrast-enhanced T1W images and 66 with fluid-attenuated inversion recovery(FLAIR) images. The end-to-end pipeline completes both tumor segmentation and status classification. The better tumor segmentation performance comes from FLAIR images (Dice score, ) compared to contrast-enhanced T1WI (Dice score, ), and the better status prediction is also from the FLAIR images (accuracy, ; recall, ; precision, ; and  score, ). This proposed pipeline not only saves the time in tumor annotation and avoids interrater variability in glioma segmentation but also achieves good prediction of MGMT methylation status. It would help find molecular biomarkers from routine medical images and further facilitate treatment planning.</p>\n<p><a href=\"https://www.hindawi.com/journals/bmri/2020/9258649/\" target=\"_blank\">https://www.hindawi.com/journals/bmri/2020/9258649/</a></p>",
      "rawMarkdown": "Below is the abstract from a paper discusses a DL pipeline for predicting Methylation of the O6-methylguanine methyltransferase (MGMT) - I found this paper very informative.  The link to the paper is below that...\n\n**Abstract**\nMethylation of the O6-methylguanine methyltransferase (MGMT) gene promoter is correlated with the effectiveness of the current standard of care in glioblastoma patients. In this study, a deep learning pipeline is designed for automatic prediction of MGMT status in 87 glioblastoma patients with contrast-enhanced T1W images and 66 with fluid-attenuated inversion recovery(FLAIR) images. The end-to-end pipeline completes both tumor segmentation and status classification. The better tumor segmentation performance comes from FLAIR images (Dice score, ) compared to contrast-enhanced T1WI (Dice score, ), and the better status prediction is also from the FLAIR images (accuracy, ; recall, ; precision, ; and  score, ). This proposed pipeline not only saves the time in tumor annotation and avoids interrater variability in glioma segmentation but also achieves good prediction of MGMT methylation status. It would help find molecular biomarkers from routine medical images and further facilitate treatment planning.\n\nhttps://www.hindawi.com/journals/bmri/2020/9258649/",
      "votes": null
    },
    {
      "id": "1390534",
      "postDate": "07/16/2021 18:27:34",
      "content": "<p>Awesome content, <a href=\"https://www.kaggle.com/mlconsult\" target=\"_blank\">@mlconsult</a>! Thanks a lot for sharing this.</p>",
      "rawMarkdown": "Awesome content, @mlconsult! Thanks a lot for sharing this.",
      "votes": null
    },
    {
      "id": "1391583",
      "postDate": "07/17/2021 18:25:53",
      "content": "<p>Great to have this resource! I particularly appreciated the section on Parameter Settings and Software that lead me to the open-source platform <a href=\"https://keras.io/\" target=\"_blank\">https://keras.io/</a>.</p>\n<p>Thank you!</p>",
      "rawMarkdown": "Great to have this resource! I particularly appreciated the section on Parameter Settings and Software that lead me to the open-source platform https://keras.io/.\n\nThank you!",
      "votes": null
    },
    {
      "id": "1392880",
      "postDate": "07/19/2021 07:16:02",
      "content": "<p>Thanks for sharing - seems like they did some detailed thinking about normalization.<br>\nMy understanding is the approach in the abstract is pinned on segmentation followed by classification. Is that everyone's understanding? I mean, they did not go directly for classification, but rather classified the tumor that was segmented by the segmentation network.</p>",
      "rawMarkdown": "Thanks for sharing - seems like they did some detailed thinking about normalization.\nMy understanding is the approach in the abstract is pinned on segmentation followed by classification. Is that everyone's understanding? I mean, they did not go directly for classification, but rather classified the tumor that was segmented by the segmentation network.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1390534,
      "author_name": "jonaslneri",
      "author_url": "",
      "post_date": "07/16/2021 18:27:34",
      "content": "<p>Awesome content, <a href=\"https://www.kaggle.com/mlconsult\" target=\"_blank\">@mlconsult</a>! Thanks a lot for sharing this.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1391583,
      "author_name": "elenaeb",
      "author_url": "",
      "post_date": "07/17/2021 18:25:53",
      "content": "<p>Great to have this resource! I particularly appreciated the section on Parameter Settings and Software that lead me to the open-source platform <a href=\"https://keras.io/\" target=\"_blank\">https://keras.io/</a>.</p>\n<p>Thank you!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1392880,
      "author_name": "pixelpassion",
      "author_url": "",
      "post_date": "07/19/2021 07:16:02",
      "content": "<p>Thanks for sharing - seems like they did some detailed thinking about normalization.<br>\nMy understanding is the approach in the abstract is pinned on segmentation followed by classification. Is that everyone's understanding? I mean, they did not go directly for classification, but rather classified the tumor that was segmented by the segmentation network.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1390454": "Below is the abstract from a paper discusses a DL pipeline for predicting Methylation of the O6-methylguanine methyltransferase (MGMT) - I found this paper very informative.  The link to the paper is below that...\n\n**Abstract**\nMethylation of the O6-methylguanine methyltransferase (MGMT) gene promoter is correlated with the effectiveness of the current standard of care in glioblastoma patients. In this study, a deep learning pipeline is designed for automatic prediction of MGMT status in 87 glioblastoma patients with contrast-enhanced T1W images and 66 with fluid-attenuated inversion recovery(FLAIR) images. The end-to-end pipeline completes both tumor segmentation and status classification. The better tumor segmentation performance comes from FLAIR images (Dice score, ) compared to contrast-enhanced T1WI (Dice score, ), and the better status prediction is also from the FLAIR images (accuracy, ; recall, ; precision, ; and  score, ). This proposed pipeline not only saves the time in tumor annotation and avoids interrater variability in glioma segmentation but also achieves good prediction of MGMT methylation status. It would help find molecular biomarkers from routine medical images and further facilitate treatment planning.\n\nhttps://www.hindawi.com/journals/bmri/2020/9258649/",
    "1390534": "Awesome content, @mlconsult! Thanks a lot for sharing this.",
    "1391583": "Great to have this resource! I particularly appreciated the section on Parameter Settings and Software that lead me to the open-source platform https://keras.io/.\n\nThank you!",
    "1392880": "Thanks for sharing - seems like they did some detailed thinking about normalization.\nMy understanding is the approach in the abstract is pinned on segmentation followed by classification. Is that everyone's understanding? I mean, they did not go directly for classification, but rather classified the tumor that was segmented by the segmentation network."
  },
  "source": "meta"
}