{
  "id": 252833,
  "title": "Papers on Radiogenomics and Machine Learning",
  "url": "/competitions/rsna-miccai-brain-tumor-radiogenomic-classification/discussion/252833",
  "author_name": "Charlie Craine",
  "post_date": "2021-07-13T23:52:44.968000",
  "votes": 112,
  "comment_count": 35,
  "views": 0,
  "content": "<p>Hey everyone!</p>\n<p>I wanted to start a papers thread and build on it, and hope others share as well, papers to gain domain knowledge. I have no domain knowledge in this area so I have downloaded papers that appeared to be relevant after reading their abstracts. I'll be reading these over the coming days and commenting more as I go through all of them. Hope this helps!</p>\n<p><strong>Research Papers:</strong></p>\n<ul>\n<li><p><a href=\"https://arxiv.org/abs/2104.08072\" target=\"_blank\">Machine Learning and Glioblastoma: Treatment Response Monitoring Biomarkers in 2021</a> - The aim of the systematic review was to assess recently published studies on diagnostic test accuracy of glioblastoma treatment response monitoring biomarkers in adults, developed through machine learning (ML). Articles were searched for using MEDLINE, EMBASE, and the Cochrane Register. Included study participants were adult patients with high grade glioma who had undergone standard treatment (maximal resection, radiotherapy with concomitant and adjuvant temozolomide) and subsequently underwent follow-up imaging to determine treatment response status.</p></li>\n<li><p><a href=\"https://arxiv.org/abs/2104.01149\" target=\"_blank\">Glioblastoma Multiforme Prognosis: MRI Missing Modality Generation, Segmentation and Radiogenomic Survival Prediction</a> -  The proposed approaches are evaluated by comparative experiments with state-of-the-art models in synthesis, segmentation, and overall survival (OS) prediction. We observe that adding missing MRI modality improves the segmentation prediction, and expression levels of gene markers have a high contribution in the GBM prognosis prediction, and fused radiogenomic features boost the OS estimation.</p></li>\n<li><p><a href=\"https://arxiv.org/abs/2103.11678\" target=\"_blank\">Feature Selection for Imbalanced Data with Deep Sparse Autoencoders Ensemble</a> - Class imbalance is a common issue in many domain applications of learning algorithms. Oftentimes, in the same domains it is much more relevant to correctly classify and profile minority class observations. This need can be addressed by Feature Selection (FS), that offers several further advantages, s.a. decreasing computational costs, aiding inference and interpretability. However, traditional FS techniques may become sub-optimal in the presence of strongly imbalanced data. To achieve FS advantages in this setting, we propose a filtering FS algorithm ranking feature importance on the basis of the Reconstruction Error of a Deep Sparse AutoEncoders Ensemble (DSAEE).</p></li>\n<li><p><a href=\"https://arxiv.org/abs/2006.09878\" target=\"_blank\">Spatial-And-Context aware (SpACe) \"virtual biopsy\" radiogenomic maps to target tumor mutational status on structural MRI</a> - With growing emphasis on personalized cancer-therapies,radiogenomics has shown promise in identifying target tumor mutational status on routine imaging (i.e. MRI) scans. These approaches fall into 2 categories: (1) deep-learning/radiomics (context-based), using image features from the entire tumor to identify the gene mutation status, or (2) atlas (spatial)-based to obtain likelihood of gene mutation status based on population statistics. While many genes (i.e. EGFR, MGMT) are spatially variant, a significant challenge in reliable assessment of gene mutation status on imaging has been the lack of available co-localized ground truth for training the models.</p></li>\n<li><p><a href=\"https://arxiv.org/abs/1907.03728\" target=\"_blank\">Correlation via synthesis: end-to-end nodule image generation and radiogenomic map learning based on generative adversarial network</a> - Radiogenomic map linking image features and gene expression profiles is useful for noninvasively identifying molecular properties of a particular type of disease. Conventionally, such map is produced in three separate steps: 1) gene-clustering to \"metagenes\", 2) image feature extraction, and 3) statistical correlation between metagenes and image features. Each step is independently performed and relies on arbitrary measurements. In this work, we investigate the potential of an end-to-end method fusing gene data with image features to generate synthetic image and learn radiogenomic map simultaneously. To achieve this goal, we develop a generative adversarial network (GAN) conditioned on both background images and gene expression profiles, synthesizing the corresponding image. Image and gene features are fused at different scales to ensure the realism and quality of the synthesized image. We tested our method on non-small cell lung cancer (NSCLC) dataset. Results demonstrate that the proposed method produces realistic synthetic images, and provides a promising way to find gene-image relationship in a holistic end-to-end manner.</p></li>\n<li><p><a href=\"https://arxiv.org/abs/2107.02314\" target=\"_blank\">The RSNA-ASNR-MICCAI BraTS 2021 Benchmark on Brain Tumor Segmentation and Radiogenomic Classification</a> - The BraTS 2021 challenge celebrates its 10th anniversary and is jointly organized by the Radiological Society of North America (RSNA), the American Society of Neuroradiology (ASNR), and the Medical Image Computing and Computer Assisted Interventions (MICCAI) society. Since its inception, BraTS has been focusing on being a common benchmarking venue for brain glioma segmentation algorithms, with well-curated multi-institutional multi-parametric magnetic resonance imaging (mpMRI) data.</p></li>\n<li><p><a href=\"https://arxiv.org/abs/2106.10941\" target=\"_blank\">Tumor Radiogenomics with Bayesian Layered Variable Selection</a> - We propose a statistical framework to integrate radiological magnetic resonance imaging (MRI) and genomic data to identify the underlying radiogenomic associations in lower grade gliomas (LGG). We devise a novel imaging phenotype by dividing the tumor region into concentric spherical layers that mimics the tumor evolution process. MRI data within each layer is represented by voxel--intensity-based probability density functions which capture the complete information about tumor heterogeneity.</p></li>\n</ul>\n<p><strong>Added 07/14/2021:</strong></p>\n<ul>\n<li><p><a href=\"https://onlinelibrary.wiley.com/doi/epdf/10.1002/cnr2.1226\" target=\"_blank\">Machine learning approaches to study glioblastoma: A review of the last decade of applications</a></p></li>\n<li><p><a href=\"https://pubmed.ncbi.nlm.nih.gov/32573435/\" target=\"_blank\">Unsupervised machine learning reveals risk stratifying glioblastoma tumor cells</a></p></li>\n</ul>",
  "messages": [
    {
      "id": 1387070,
      "postDate": "2021-07-13T23:52:44.970Z",
      "content": "<p>Hey everyone!</p>\n<p>I wanted to start a papers thread and build on it, and hope others share as well, papers to gain domain knowledge. I have no domain knowledge in this area so I have downloaded papers that appeared to be relevant after reading their abstracts. I'll be reading these over the coming days and commenting more as I go through all of them. Hope this helps!</p>\n<p><strong>Research Papers:</strong></p>\n<ul>\n<li><p><a href=\"https://arxiv.org/abs/2104.08072\" target=\"_blank\">Machine Learning and Glioblastoma: Treatment Response Monitoring Biomarkers in 2021</a> - The aim of the systematic review was to assess recently published studies on diagnostic test accuracy of glioblastoma treatment response monitoring biomarkers in adults, developed through machine learning (ML). Articles were searched for using MEDLINE, EMBASE, and the Cochrane Register. Included study participants were adult patients with high grade glioma who had undergone standard treatment (maximal resection, radiotherapy with concomitant and adjuvant temozolomide) and subsequently underwent follow-up imaging to determine treatment response status.</p></li>\n<li><p><a href=\"https://arxiv.org/abs/2104.01149\" target=\"_blank\">Glioblastoma Multiforme Prognosis: MRI Missing Modality Generation, Segmentation and Radiogenomic Survival Prediction</a> -  The proposed approaches are evaluated by comparative experiments with state-of-the-art models in synthesis, segmentation, and overall survival (OS) prediction. We observe that adding missing MRI modality improves the segmentation prediction, and expression levels of gene markers have a high contribution in the GBM prognosis prediction, and fused radiogenomic features boost the OS estimation.</p></li>\n<li><p><a href=\"https://arxiv.org/abs/2103.11678\" target=\"_blank\">Feature Selection for Imbalanced Data with Deep Sparse Autoencoders Ensemble</a> - Class imbalance is a common issue in many domain applications of learning algorithms. Oftentimes, in the same domains it is much more relevant to correctly classify and profile minority class observations. This need can be addressed by Feature Selection (FS), that offers several further advantages, s.a. decreasing computational costs, aiding inference and interpretability. However, traditional FS techniques may become sub-optimal in the presence of strongly imbalanced data. To achieve FS advantages in this setting, we propose a filtering FS algorithm ranking feature importance on the basis of the Reconstruction Error of a Deep Sparse AutoEncoders Ensemble (DSAEE).</p></li>\n<li><p><a href=\"https://arxiv.org/abs/2006.09878\" target=\"_blank\">Spatial-And-Context aware (SpACe) \"virtual biopsy\" radiogenomic maps to target tumor mutational status on structural MRI</a> - With growing emphasis on personalized cancer-therapies,radiogenomics has shown promise in identifying target tumor mutational status on routine imaging (i.e. MRI) scans. These approaches fall into 2 categories: (1) deep-learning/radiomics (context-based), using image features from the entire tumor to identify the gene mutation status, or (2) atlas (spatial)-based to obtain likelihood of gene mutation status based on population statistics. While many genes (i.e. EGFR, MGMT) are spatially variant, a significant challenge in reliable assessment of gene mutation status on imaging has been the lack of available co-localized ground truth for training the models.</p></li>\n<li><p><a href=\"https://arxiv.org/abs/1907.03728\" target=\"_blank\">Correlation via synthesis: end-to-end nodule image generation and radiogenomic map learning based on generative adversarial network</a> - Radiogenomic map linking image features and gene expression profiles is useful for noninvasively identifying molecular properties of a particular type of disease. Conventionally, such map is produced in three separate steps: 1) gene-clustering to \"metagenes\", 2) image feature extraction, and 3) statistical correlation between metagenes and image features. Each step is independently performed and relies on arbitrary measurements. In this work, we investigate the potential of an end-to-end method fusing gene data with image features to generate synthetic image and learn radiogenomic map simultaneously. To achieve this goal, we develop a generative adversarial network (GAN) conditioned on both background images and gene expression profiles, synthesizing the corresponding image. Image and gene features are fused at different scales to ensure the realism and quality of the synthesized image. We tested our method on non-small cell lung cancer (NSCLC) dataset. Results demonstrate that the proposed method produces realistic synthetic images, and provides a promising way to find gene-image relationship in a holistic end-to-end manner.</p></li>\n<li><p><a href=\"https://arxiv.org/abs/2107.02314\" target=\"_blank\">The RSNA-ASNR-MICCAI BraTS 2021 Benchmark on Brain Tumor Segmentation and Radiogenomic Classification</a> - The BraTS 2021 challenge celebrates its 10th anniversary and is jointly organized by the Radiological Society of North America (RSNA), the American Society of Neuroradiology (ASNR), and the Medical Image Computing and Computer Assisted Interventions (MICCAI) society. Since its inception, BraTS has been focusing on being a common benchmarking venue for brain glioma segmentation algorithms, with well-curated multi-institutional multi-parametric magnetic resonance imaging (mpMRI) data.</p></li>\n<li><p><a href=\"https://arxiv.org/abs/2106.10941\" target=\"_blank\">Tumor Radiogenomics with Bayesian Layered Variable Selection</a> - We propose a statistical framework to integrate radiological magnetic resonance imaging (MRI) and genomic data to identify the underlying radiogenomic associations in lower grade gliomas (LGG). We devise a novel imaging phenotype by dividing the tumor region into concentric spherical layers that mimics the tumor evolution process. MRI data within each layer is represented by voxel--intensity-based probability density functions which capture the complete information about tumor heterogeneity.</p></li>\n</ul>\n<p><strong>Added 07/14/2021:</strong></p>\n<ul>\n<li><p><a href=\"https://onlinelibrary.wiley.com/doi/epdf/10.1002/cnr2.1226\" target=\"_blank\">Machine learning approaches to study glioblastoma: A review of the last decade of applications</a></p></li>\n<li><p><a href=\"https://pubmed.ncbi.nlm.nih.gov/32573435/\" target=\"_blank\">Unsupervised machine learning reveals risk stratifying glioblastoma tumor cells</a></p></li>\n</ul>",
      "rawMarkdown": "Hey everyone!\n\nI wanted to start a papers thread and build on it, and hope others share as well, papers to gain domain knowledge. I have no domain knowledge in this area so I have downloaded papers that appeared to be relevant after reading their abstracts. I'll be reading these over the coming days and commenting more as I go through all of them. Hope this helps!\n\n**Research Papers:**\n\n- [Machine Learning and Glioblastoma: Treatment Response Monitoring Biomarkers in 2021](https://arxiv.org/abs/2104.08072) - The aim of the systematic review was to assess recently published studies on diagnostic test accuracy of glioblastoma treatment response monitoring biomarkers in adults, developed through machine learning (ML). Articles were searched for using MEDLINE, EMBASE, and the Cochrane Register. Included study participants were adult patients with high grade glioma who had undergone standard treatment (maximal resection, radiotherapy with concomitant and adjuvant temozolomide) and subsequently underwent follow-up imaging to determine treatment response status.\n\n- [Glioblastoma Multiforme Prognosis: MRI Missing Modality Generation, Segmentation and Radiogenomic Survival Prediction](https://arxiv.org/abs/2104.01149) -  The proposed approaches are evaluated by comparative experiments with state-of-the-art models in synthesis, segmentation, and overall survival (OS) prediction. We observe that adding missing MRI modality improves the segmentation prediction, and expression levels of gene markers have a high contribution in the GBM prognosis prediction, and fused radiogenomic features boost the OS estimation.\n\n- [Feature Selection for Imbalanced Data with Deep Sparse Autoencoders Ensemble](https://arxiv.org/abs/2103.11678) - Class imbalance is a common issue in many domain applications of learning algorithms. Oftentimes, in the same domains it is much more relevant to correctly classify and profile minority class observations. This need can be addressed by Feature Selection (FS), that offers several further advantages, s.a. decreasing computational costs, aiding inference and interpretability. However, traditional FS techniques may become sub-optimal in the presence of strongly imbalanced data. To achieve FS advantages in this setting, we propose a filtering FS algorithm ranking feature importance on the basis of the Reconstruction Error of a Deep Sparse AutoEncoders Ensemble (DSAEE).\n\n- [Spatial-And-Context aware (SpACe) \"virtual biopsy\" radiogenomic maps to target tumor mutational status on structural MRI](https://arxiv.org/abs/2006.09878) - With growing emphasis on personalized cancer-therapies,radiogenomics has shown promise in identifying target tumor mutational status on routine imaging (i.e. MRI) scans. These approaches fall into 2 categories: (1) deep-learning/radiomics (context-based), using image features from the entire tumor to identify the gene mutation status, or (2) atlas (spatial)-based to obtain likelihood of gene mutation status based on population statistics. While many genes (i.e. EGFR, MGMT) are spatially variant, a significant challenge in reliable assessment of gene mutation status on imaging has been the lack of available co-localized ground truth for training the models.\n\n- [Correlation via synthesis: end-to-end nodule image generation and radiogenomic map learning based on generative adversarial network](https://arxiv.org/abs/1907.03728) - Radiogenomic map linking image features and gene expression profiles is useful for noninvasively identifying molecular properties of a particular type of disease. Conventionally, such map is produced in three separate steps: 1) gene-clustering to \"metagenes\", 2) image feature extraction, and 3) statistical correlation between metagenes and image features. Each step is independently performed and relies on arbitrary measurements. In this work, we investigate the potential of an end-to-end method fusing gene data with image features to generate synthetic image and learn radiogenomic map simultaneously. To achieve this goal, we develop a generative adversarial network (GAN) conditioned on both background images and gene expression profiles, synthesizing the corresponding image. Image and gene features are fused at different scales to ensure the realism and quality of the synthesized image. We tested our method on non-small cell lung cancer (NSCLC) dataset. Results demonstrate that the proposed method produces realistic synthetic images, and provides a promising way to find gene-image relationship in a holistic end-to-end manner.\n\n- [The RSNA-ASNR-MICCAI BraTS 2021 Benchmark on Brain Tumor Segmentation and Radiogenomic Classification](https://arxiv.org/abs/2107.02314) - The BraTS 2021 challenge celebrates its 10th anniversary and is jointly organized by the Radiological Society of North America (RSNA), the American Society of Neuroradiology (ASNR), and the Medical Image Computing and Computer Assisted Interventions (MICCAI) society. Since its inception, BraTS has been focusing on being a common benchmarking venue for brain glioma segmentation algorithms, with well-curated multi-institutional multi-parametric magnetic resonance imaging (mpMRI) data.\n\n- [Tumor Radiogenomics with Bayesian Layered Variable Selection](https://arxiv.org/abs/2106.10941) - We propose a statistical framework to integrate radiological magnetic resonance imaging (MRI) and genomic data to identify the underlying radiogenomic associations in lower grade gliomas (LGG). We devise a novel imaging phenotype by dividing the tumor region into concentric spherical layers that mimics the tumor evolution process. MRI data within each layer is represented by voxel--intensity-based probability density functions which capture the complete information about tumor heterogeneity.\n\n**Added 07/14/2021:**\n- [Machine learning approaches to study glioblastoma: A review of the last decade of applications](https://onlinelibrary.wiley.com/doi/epdf/10.1002/cnr2.1226)\n\n- [Unsupervised machine learning reveals risk stratifying glioblastoma tumor cells](https://pubmed.ncbi.nlm.nih.gov/32573435/)",
      "votes": 112
    },
    {
      "id": 1390627,
      "postDate": "2021-07-16T21:28:29.997Z",
      "content": "<p>You might also want to consider the following for:</p>\n<ul>\n<li>a broader radiogenomics review: <a href=\"https://pubmed.ncbi.nlm.nih.gov/31456318/\" target=\"_blank\">Imaging signatures of glioblastoma molecular characteristics: A radiogenomics review</a></li>\n<li>a review on MRI for glioblastoma: <a href=\"https://pubmed.ncbi.nlm.nih.gov/28841802/\" target=\"_blank\">Advanced magnetic resonance imaging in glioblastoma: a review</a></li>\n</ul>",
      "rawMarkdown": "You might also want to consider the following for:\n- a broader radiogenomics review: [Imaging signatures of glioblastoma molecular characteristics: A radiogenomics review](https://pubmed.ncbi.nlm.nih.gov/31456318/)\n- a review on MRI for glioblastoma: [Advanced magnetic resonance imaging in glioblastoma: a review](https://pubmed.ncbi.nlm.nih.gov/28841802/)\n\n ",
      "votes": 3
    },
    {
      "id": 1464637,
      "postDate": "2021-08-10T16:38:39.847Z",
      "content": "<p>MRI-Based Deep-Learning Method for Determining Glioma MGMT Promoter Methylation Status<br>\n<a href=\"https://doi.org/10.3174/ajnr.A7029\" target=\"_blank\">https://doi.org/10.3174/ajnr.A7029</a><br>\nAmerican Journal of Neuroradiology May 2021</p>\n<blockquote>\n  <p>predicting MGMT methylation status with a sensitivity and specificity of 96.31% [SD, 0.04%] and 91.66% [SD, 2.06%], respectively, and a mean area under the curve of 0.93 [SD, 0.01].</p>\n</blockquote>",
      "rawMarkdown": "MRI-Based Deep-Learning Method for Determining Glioma MGMT Promoter Methylation Status\nhttps://doi.org/10.3174/ajnr.A7029\nAmerican Journal of Neuroradiology May 2021\n\n> predicting MGMT methylation status with a sensitivity and specificity of 96.31% [SD, 0.04%] and 91.66% [SD, 2.06%], respectively, and a mean area under the curve of 0.93 [SD, 0.01].",
      "votes": 1
    },
    {
      "id": 1387101,
      "postDate": "2021-07-14T00:40:46.887Z",
      "content": "<p>Really great starting points of research.  Thank you!</p>",
      "rawMarkdown": "Really great starting points of research.  Thank you!",
      "votes": 1
    },
    {
      "id": 1387076,
      "postDate": "2021-07-13T23:58:41.980Z",
      "content": "<p>I was about to post papers, you beat me =) Nice List. </p>",
      "rawMarkdown": "I was about to post papers, you beat me =) Nice List. ",
      "votes": 1,
      "replies": [
        {
          "id": 1387081,
          "postDate": "2021-07-14T00:04:35.767Z",
          "content": "<p>Thank you! Feel free to add more!</p>",
          "rawMarkdown": "Thank you! Feel free to add more!",
          "votes": 1
        }
      ]
    },
    {
      "id": 1390470,
      "postDate": "2021-07-16T17:07:22.080Z",
      "content": "<p><a href=\"https://www.kaggle.com/crained\" target=\"_blank\">@crained</a> - great list - before I saw your discussion, I posted an abstract from and a link to a paper discussing <strong>Automatic Prediction of MGMT Status in Glioblastoma</strong></p>\n<p>Here is the link to the discussion.</p>\n<p><a href=\"https://www.kaggle.com/c/rsna-miccai-brain-tumor-radiogenomic-classification/discussion/253451\" target=\"_blank\">https://www.kaggle.com/c/rsna-miccai-brain-tumor-radiogenomic-classification/discussion/253451</a></p>",
      "rawMarkdown": "@crained - great list - before I saw your discussion, I posted an abstract from and a link to a paper discussing **Automatic Prediction of MGMT Status in Glioblastoma**\n\nHere is the link to the discussion.\n\nhttps://www.kaggle.com/c/rsna-miccai-brain-tumor-radiogenomic-classification/discussion/253451\n\n",
      "votes": 2
    },
    {
      "id": 1389476,
      "postDate": "2021-07-15T18:23:16.500Z",
      "content": "<p>Hi,</p>\n<p>I found this one. May be interesting:</p>\n<p><a href=\"https://www.nature.com/articles/s41598-021-90428-8\" target=\"_blank\">https://www.nature.com/articles/s41598-021-90428-8</a></p>",
      "rawMarkdown": "Hi,\n\nI found this one. May be interesting:\n\nhttps://www.nature.com/articles/s41598-021-90428-8\n\n",
      "votes": 2
    },
    {
      "id": 1387131,
      "postDate": "2021-07-14T01:43:56.450Z",
      "content": "<p>Thanks for sharing …can we share paid papers ?? just asking ..</p>",
      "rawMarkdown": "Thanks for sharing ...can we share paid papers ?? just asking ..",
      "votes": 2,
      "replies": [
        {
          "id": 1387668,
          "postDate": "2021-07-14T11:00:03.083Z",
          "content": "<p>Do you have an example of one? Is it one you own or?</p>",
          "rawMarkdown": "Do you have an example of one? Is it one you own or?",
          "votes": 1
        },
        {
          "id": 1387830,
          "postDate": "2021-07-14T13:01:04.997Z",
          "content": "<p>No, it's not my own. The papers which I have downloaded from IEEE Xplore after Subscribing for Individual Access <a href=\"https://www.kaggle.com/crained\" target=\"_blank\">@crained</a> </p>",
          "rawMarkdown": "No, it's not my own. The papers which I have downloaded from IEEE Xplore after Subscribing for Individual Access @crained "
        },
        {
          "id": 1387862,
          "postDate": "2021-07-14T13:28:42.147Z",
          "content": "<p>I'd check with the licensing but my guess is you can't share the actual paper but you could share the link and just say you have to pay for it. That way it just sets expectations for everyone. </p>",
          "rawMarkdown": "I'd check with the licensing but my guess is you can't share the actual paper but you could share the link and just say you have to pay for it. That way it just sets expectations for everyone. ",
          "votes": 2
        },
        {
          "id": 1388001,
          "postDate": "2021-07-14T15:16:41.820Z",
          "content": "<p>Thanks <a href=\"https://www.kaggle.com/crained\" target=\"_blank\">@crained</a> for your information… </p>",
          "rawMarkdown": "Thanks @crained for your information... "
        },
        {
          "id": 1388112,
          "postDate": "2021-07-14T16:50:51.890Z",
          "content": "<p>Sharing a paid content can have licensing concerns. But, the community would highly appreciate it if you can make a post on your learnings from those papers. Any awesome content is always welcomed </p>",
          "rawMarkdown": "Sharing a paid content can have licensing concerns. But, the community would highly appreciate it if you can make a post on your learnings from those papers. Any awesome content is always welcomed ",
          "votes": 3
        },
        {
          "id": 1404410,
          "postDate": "2021-07-29T20:46:01.773Z",
          "content": "<p>sci-hub ;)</p>",
          "rawMarkdown": "sci-hub ;)"
        }
      ]
    },
    {
      "id": 1495705,
      "postDate": "2021-08-29T17:47:27.137Z",
      "content": "<p><a href=\"https://www.kaggle.com/crained\" target=\"_blank\">@crained</a> Excellent Share, i was searching for some papers Thanks 👍</p>",
      "rawMarkdown": "@crained Excellent Share, i was searching for some papers Thanks 👍"
    },
    {
      "id": 1495631,
      "postDate": "2021-08-29T16:55:47.890Z",
      "content": "<p>Thanks a lot for sharing - very useful indeed.</p>",
      "rawMarkdown": "Thanks a lot for sharing - very useful indeed."
    },
    {
      "id": 1455594,
      "postDate": "2021-08-06T16:07:45.557Z",
      "content": "<p>Let me paste this paragraph:</p>\n<p>Thus, we restricted the current analysis in GBM patients<br>\nwith wild-type IDH. The purpose of our study was to seek<br>\ncertain variables derived from conventional structural<br>\nimage features including <strong>multifocal, tumor cross midline,\ntumor location, enhancement, cyst, necrosis, edema, side</strong>,<br>\nwhich may reflect MGMT promoter methylation status.</p>\n<p>from the following paper:</p>\n<p><a href=\"https://pubmed.ncbi.nlm.nih.gov/29467012/\" target=\"_blank\">https://pubmed.ncbi.nlm.nih.gov/29467012/</a></p>\n<p>Hope you find it useful.</p>",
      "rawMarkdown": "Let me paste this paragraph:\n\nThus, we restricted the current analysis in GBM patients\nwith wild-type IDH. The purpose of our study was to seek\ncertain variables derived from conventional structural\nimage features including **multifocal, tumor cross midline,\ntumor location, enhancement, cyst, necrosis, edema, side**,\nwhich may reflect MGMT promoter methylation status.\n\nfrom the following paper:\n\nhttps://pubmed.ncbi.nlm.nih.gov/29467012/\n\nHope you find it useful."
    },
    {
      "id": 1390004,
      "postDate": "2021-07-16T09:21:52.173Z",
      "content": "<p>great work! thx for sharing </p>",
      "rawMarkdown": "great work! thx for sharing "
    },
    {
      "id": 1388433,
      "postDate": "2021-07-15T00:12:04.697Z",
      "content": "<p>Good day, <br>\nI think this is some good back ground reading and source if I posted in the wrong forum please advise as I’m new to platform.<br>\n<a href=\"https://scholar.google.com/scholar?q=MGMT+promoter+methylation&amp;hl=en&amp;as_sdt=0&amp;as_vis=1&amp;oi=scholart#d=gs_qabs&amp;u=%23p%3D4uQBUNP3C1sJ\" target=\"_blank\">https://scholar.google.com/scholar?q=MGMT+promoter+methylation&amp;hl=en&amp;as_sdt=0&amp;as_vis=1&amp;oi=scholart#d=gs_qabs&amp;u=%23p%3D4uQBUNP3C1sJ</a></p>",
      "rawMarkdown": "Good day, \nI think this is some good back ground reading and source if I posted in the wrong forum please advise as I’m new to platform.\nhttps://scholar.google.com/scholar?q=MGMT+promoter+methylation&hl=en&as_sdt=0&as_vis=1&oi=scholart#d=gs_qabs&u=%23p%3D4uQBUNP3C1sJ"
    },
    {
      "id": 1388367,
      "postDate": "2021-07-14T21:39:29.267Z",
      "content": "<p>Updated with a few more papers. I'll add dates so if you come back you can see what are new papers. I have never done this before so hope it is helpful. Let me know. Enjoy!</p>",
      "rawMarkdown": "Updated with a few more papers. I'll add dates so if you come back you can see what are new papers. I have never done this before so hope it is helpful. Let me know. Enjoy!"
    },
    {
      "id": 1388299,
      "postDate": "2021-07-14T19:39:22.220Z",
      "content": "<p>R Packages mentioned in practitioner articles, open access:</p>\n<p><a href=\"https://academic.oup.com/noa/article/2/1/vdaa117/5983632\" target=\"_blank\">MGMT promoter methylation</a><br>\n<a href=\"https://www.mdpi.com/2072-6694/13/6/1377\" target=\"_blank\">Pediatric Neuro-oncology diagnostic workflow implementation France</a></p>",
      "rawMarkdown": "R Packages mentioned in practitioner articles, open access:\n\n[MGMT promoter methylation](https://academic.oup.com/noa/article/2/1/vdaa117/5983632)\n[Pediatric Neuro-oncology diagnostic workflow implementation France](https://www.mdpi.com/2072-6694/13/6/1377)\n\n"
    },
    {
      "id": 1387999,
      "postDate": "2021-07-14T15:12:27.017Z",
      "content": "<p>Great One…</p>",
      "rawMarkdown": "Great One..."
    },
    {
      "id": 1387813,
      "postDate": "2021-07-14T12:46:15.300Z",
      "content": "<p>Nice sharing</p>",
      "rawMarkdown": "Nice sharing"
    },
    {
      "id": 1387453,
      "postDate": "2021-07-14T07:40:28.877Z",
      "content": "<p>Awesome listing! Thx for sharing :) </p>",
      "rawMarkdown": "Awesome listing! Thx for sharing :) "
    },
    {
      "id": 1389926,
      "postDate": "2021-07-16T08:15:18.193Z",
      "rawMarkdown": "",
      "isDeleted": true
    },
    {
      "id": 1388413,
      "postDate": "2021-07-14T23:27:23.340Z",
      "rawMarkdown": "",
      "isDeleted": true
    },
    {
      "id": 1495910,
      "postDate": "2021-08-29T22:03:44.033Z",
      "content": "<p>Very helpful! Thank u for sharing =))</p>",
      "rawMarkdown": "Very helpful! Thank u for sharing =))",
      "votes": 1
    },
    {
      "id": 1388526,
      "postDate": "2021-07-15T03:40:13.097Z",
      "content": "<p>Thanks for sharing</p>",
      "rawMarkdown": "Thanks for sharing",
      "votes": 1
    },
    {
      "id": 1495524,
      "postDate": "2021-08-29T16:00:36.013Z",
      "content": "<p>Thank for sharing!</p>",
      "rawMarkdown": "Thank for sharing!"
    },
    {
      "id": 1471064,
      "postDate": "2021-08-13T22:44:15.700Z",
      "content": "<p>Thank you for sharing!</p>",
      "rawMarkdown": "Thank you for sharing!"
    },
    {
      "id": 1457395,
      "postDate": "2021-08-07T11:37:37.740Z",
      "content": "<p>Thanks for sharing <a href=\"https://www.kaggle.com/crained\" target=\"_blank\">@crained</a> </p>",
      "rawMarkdown": "Thanks for sharing @crained "
    },
    {
      "id": 1455801,
      "postDate": "2021-08-06T17:32:22.933Z",
      "content": "<p>Nice research thank you </p>",
      "rawMarkdown": "Nice research thank you "
    },
    {
      "id": 1401731,
      "postDate": "2021-07-27T14:20:43.437Z",
      "content": "<p>Great work Thanks for sharing !</p>",
      "rawMarkdown": "Great work Thanks for sharing !"
    },
    {
      "id": 1392168,
      "postDate": "2021-07-18T12:02:37.007Z",
      "content": "<p>Great Thanks for sharing !</p>",
      "rawMarkdown": "Great Thanks for sharing !"
    },
    {
      "id": 1387900,
      "postDate": "2021-07-14T13:59:12.753Z",
      "content": "<p>Awesome list. Thanks for sharing 👍</p>",
      "rawMarkdown": "Awesome list. Thanks for sharing 👍"
    }
  ],
  "comments": [
    {
      "id": 1390627,
      "author_name": "Spyridon Bakas",
      "author_url": "",
      "post_date": "2021-07-16T21:28:29.997000",
      "content": "<p>You might also want to consider the following for:</p>\n<ul>\n<li>a broader radiogenomics review: <a href=\"https://pubmed.ncbi.nlm.nih.gov/31456318/\" target=\"_blank\">Imaging signatures of glioblastoma molecular characteristics: A radiogenomics review</a></li>\n<li>a review on MRI for glioblastoma: <a href=\"https://pubmed.ncbi.nlm.nih.gov/28841802/\" target=\"_blank\">Advanced magnetic resonance imaging in glioblastoma: a review</a></li>\n</ul>",
      "votes": 3,
      "replies": []
    },
    {
      "id": 1464637,
      "author_name": "wafflebufflo",
      "author_url": "",
      "post_date": "2021-08-10T16:38:39.847000",
      "content": "<p>MRI-Based Deep-Learning Method for Determining Glioma MGMT Promoter Methylation Status<br>\n<a href=\"https://doi.org/10.3174/ajnr.A7029\" target=\"_blank\">https://doi.org/10.3174/ajnr.A7029</a><br>\nAmerican Journal of Neuroradiology May 2021</p>\n<blockquote>\n  <p>predicting MGMT methylation status with a sensitivity and specificity of 96.31% [SD, 0.04%] and 91.66% [SD, 2.06%], respectively, and a mean area under the curve of 0.93 [SD, 0.01].</p>\n</blockquote>",
      "votes": 1,
      "replies": []
    },
    {
      "id": 1387101,
      "author_name": "ElenaEB",
      "author_url": "",
      "post_date": "2021-07-14T00:40:46.887000",
      "content": "<p>Really great starting points of research.  Thank you!</p>",
      "votes": 1,
      "replies": []
    },
    {
      "id": 1387076,
      "author_name": "DrHB",
      "author_url": "",
      "post_date": "2021-07-13T23:58:41.980000",
      "content": "<p>I was about to post papers, you beat me =) Nice List. </p>",
      "votes": 1,
      "replies": [
        {
          "id": 1387081,
          "author_name": "Charlie Craine",
          "author_url": "",
          "post_date": "2021-07-14T00:04:35.767000",
          "content": "<p>Thank you! Feel free to add more!</p>",
          "votes": 1,
          "replies": []
        }
      ]
    },
    {
      "id": 1390470,
      "author_name": "Ken Miller",
      "author_url": "",
      "post_date": "2021-07-16T17:07:22.080000",
      "content": "<p><a href=\"https://www.kaggle.com/crained\" target=\"_blank\">@crained</a> - great list - before I saw your discussion, I posted an abstract from and a link to a paper discussing <strong>Automatic Prediction of MGMT Status in Glioblastoma</strong></p>\n<p>Here is the link to the discussion.</p>\n<p><a href=\"https://www.kaggle.com/c/rsna-miccai-brain-tumor-radiogenomic-classification/discussion/253451\" target=\"_blank\">https://www.kaggle.com/c/rsna-miccai-brain-tumor-radiogenomic-classification/discussion/253451</a></p>",
      "votes": 2,
      "replies": []
    },
    {
      "id": 1389476,
      "author_name": "LucaMTB",
      "author_url": "",
      "post_date": "2021-07-15T18:23:16.500000",
      "content": "<p>Hi,</p>\n<p>I found this one. May be interesting:</p>\n<p><a href=\"https://www.nature.com/articles/s41598-021-90428-8\" target=\"_blank\">https://www.nature.com/articles/s41598-021-90428-8</a></p>",
      "votes": 2,
      "replies": []
    },
    {
      "id": 1387131,
      "author_name": "laxman kusuma",
      "author_url": "",
      "post_date": "2021-07-14T01:43:56.450000",
      "content": "<p>Thanks for sharing …can we share paid papers ?? just asking ..</p>",
      "votes": 2,
      "replies": [
        {
          "id": 1387668,
          "author_name": "Charlie Craine",
          "author_url": "",
          "post_date": "2021-07-14T11:00:03.083000",
          "content": "<p>Do you have an example of one? Is it one you own or?</p>",
          "votes": 1,
          "replies": []
        },
        {
          "id": 1387830,
          "author_name": "laxman kusuma",
          "author_url": "",
          "post_date": "2021-07-14T13:01:04.997000",
          "content": "<p>No, it's not my own. The papers which I have downloaded from IEEE Xplore after Subscribing for Individual Access <a href=\"https://www.kaggle.com/crained\" target=\"_blank\">@crained</a> </p>",
          "votes": 0,
          "replies": []
        },
        {
          "id": 1387862,
          "author_name": "Charlie Craine",
          "author_url": "",
          "post_date": "2021-07-14T13:28:42.147000",
          "content": "<p>I'd check with the licensing but my guess is you can't share the actual paper but you could share the link and just say you have to pay for it. That way it just sets expectations for everyone. </p>",
          "votes": 2,
          "replies": []
        },
        {
          "id": 1388001,
          "author_name": "laxman kusuma",
          "author_url": "",
          "post_date": "2021-07-14T15:16:41.820000",
          "content": "<p>Thanks <a href=\"https://www.kaggle.com/crained\" target=\"_blank\">@crained</a> for your information… </p>",
          "votes": 0,
          "replies": []
        },
        {
          "id": 1388112,
          "author_name": "Shahebaz Mohammad",
          "author_url": "",
          "post_date": "2021-07-14T16:50:51.890000",
          "content": "<p>Sharing a paid content can have licensing concerns. But, the community would highly appreciate it if you can make a post on your learnings from those papers. Any awesome content is always welcomed </p>",
          "votes": 3,
          "replies": []
        },
        {
          "id": 1404410,
          "author_name": "Sagar",
          "author_url": "",
          "post_date": "2021-07-29T20:46:01.773000",
          "content": "<p>sci-hub ;)</p>",
          "votes": 0,
          "replies": []
        }
      ]
    },
    {
      "id": 1495705,
      "author_name": "Susant_Achary🎲",
      "author_url": "",
      "post_date": "2021-08-29T17:47:27.137000",
      "content": "<p><a href=\"https://www.kaggle.com/crained\" target=\"_blank\">@crained</a> Excellent Share, i was searching for some papers Thanks 👍</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1495631,
      "author_name": "santiagonasar",
      "author_url": "",
      "post_date": "2021-08-29T16:55:47.890000",
      "content": "<p>Thanks a lot for sharing - very useful indeed.</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1455594,
      "author_name": "ademdum",
      "author_url": "",
      "post_date": "2021-08-06T16:07:45.557000",
      "content": "<p>Let me paste this paragraph:</p>\n<p>Thus, we restricted the current analysis in GBM patients<br>\nwith wild-type IDH. The purpose of our study was to seek<br>\ncertain variables derived from conventional structural<br>\nimage features including <strong>multifocal, tumor cross midline,\ntumor location, enhancement, cyst, necrosis, edema, side</strong>,<br>\nwhich may reflect MGMT promoter methylation status.</p>\n<p>from the following paper:</p>\n<p><a href=\"https://pubmed.ncbi.nlm.nih.gov/29467012/\" target=\"_blank\">https://pubmed.ncbi.nlm.nih.gov/29467012/</a></p>\n<p>Hope you find it useful.</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1390004,
      "author_name": "Evgeniy Alkhovik",
      "author_url": "",
      "post_date": "2021-07-16T09:21:52.173000",
      "content": "<p>great work! thx for sharing </p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1388433,
      "author_name": "captbullett",
      "author_url": "",
      "post_date": "2021-07-15T00:12:04.697000",
      "content": "<p>Good day, <br>\nI think this is some good back ground reading and source if I posted in the wrong forum please advise as I’m new to platform.<br>\n<a href=\"https://scholar.google.com/scholar?q=MGMT+promoter+methylation&amp;hl=en&amp;as_sdt=0&amp;as_vis=1&amp;oi=scholart#d=gs_qabs&amp;u=%23p%3D4uQBUNP3C1sJ\" target=\"_blank\">https://scholar.google.com/scholar?q=MGMT+promoter+methylation&amp;hl=en&amp;as_sdt=0&amp;as_vis=1&amp;oi=scholart#d=gs_qabs&amp;u=%23p%3D4uQBUNP3C1sJ</a></p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1388367,
      "author_name": "Charlie Craine",
      "author_url": "",
      "post_date": "2021-07-14T21:39:29.267000",
      "content": "<p>Updated with a few more papers. I'll add dates so if you come back you can see what are new papers. I have never done this before so hope it is helpful. Let me know. Enjoy!</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1388299,
      "author_name": "Chris Engish",
      "author_url": "",
      "post_date": "2021-07-14T19:39:22.220000",
      "content": "<p>R Packages mentioned in practitioner articles, open access:</p>\n<p><a href=\"https://academic.oup.com/noa/article/2/1/vdaa117/5983632\" target=\"_blank\">MGMT promoter methylation</a><br>\n<a href=\"https://www.mdpi.com/2072-6694/13/6/1377\" target=\"_blank\">Pediatric Neuro-oncology diagnostic workflow implementation France</a></p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1387999,
      "author_name": "Amol Ambkar",
      "author_url": "",
      "post_date": "2021-07-14T15:12:27.017000",
      "content": "<p>Great One…</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1387813,
      "author_name": "Joshua Muwanguzi",
      "author_url": "",
      "post_date": "2021-07-14T12:46:15.300000",
      "content": "<p>Nice sharing</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1387453,
      "author_name": "Sangwook Kim",
      "author_url": "",
      "post_date": "2021-07-14T07:40:28.877000",
      "content": "<p>Awesome listing! Thx for sharing :) </p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1389926,
      "author_name": "",
      "author_url": "",
      "post_date": "2021-07-16T08:15:18.193000",
      "content": "",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1388413,
      "author_name": "",
      "author_url": "",
      "post_date": "2021-07-14T23:27:23.340000",
      "content": "",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1495910,
      "author_name": "Fuco",
      "author_url": "",
      "post_date": "2021-08-29T22:03:44.033000",
      "content": "<p>Very helpful! Thank u for sharing =))</p>",
      "votes": 1,
      "replies": []
    },
    {
      "id": 1388526,
      "author_name": "M.M. Sabbir Islam",
      "author_url": "",
      "post_date": "2021-07-15T03:40:13.097000",
      "content": "<p>Thanks for sharing</p>",
      "votes": 1,
      "replies": []
    },
    {
      "id": 1495524,
      "author_name": "namtran",
      "author_url": "",
      "post_date": "2021-08-29T16:00:36.013000",
      "content": "<p>Thank for sharing!</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1471064,
      "author_name": "Lilkoke",
      "author_url": "",
      "post_date": "2021-08-13T22:44:15.700000",
      "content": "<p>Thank you for sharing!</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1457395,
      "author_name": "Dr. Amritpal Singh",
      "author_url": "",
      "post_date": "2021-08-07T11:37:37.740000",
      "content": "<p>Thanks for sharing <a href=\"https://www.kaggle.com/crained\" target=\"_blank\">@crained</a> </p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1455801,
      "author_name": "A7_H",
      "author_url": "",
      "post_date": "2021-08-06T17:32:22.933000",
      "content": "<p>Nice research thank you </p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1401731,
      "author_name": "Alexey Bragin",
      "author_url": "",
      "post_date": "2021-07-27T14:20:43.437000",
      "content": "<p>Great work Thanks for sharing !</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1392168,
      "author_name": "Mohamed Hany",
      "author_url": "",
      "post_date": "2021-07-18T12:02:37.007000",
      "content": "<p>Great Thanks for sharing !</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1387900,
      "author_name": "Sani Kamal",
      "author_url": "",
      "post_date": "2021-07-14T13:59:12.753000",
      "content": "<p>Awesome list. Thanks for sharing 👍</p>",
      "votes": 0,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1387070": "Hey everyone!\n\nI wanted to start a papers thread and build on it, and hope others share as well, papers to gain domain knowledge. I have no domain knowledge in this area so I have downloaded papers that appeared to be relevant after reading their abstracts. I'll be reading these over the coming days and commenting more as I go through all of them. Hope this helps!\n\n**Research Papers:**\n\n- [Machine Learning and Glioblastoma: Treatment Response Monitoring Biomarkers in 2021](https://arxiv.org/abs/2104.08072) - The aim of the systematic review was to assess recently published studies on diagnostic test accuracy of glioblastoma treatment response monitoring biomarkers in adults, developed through machine learning (ML). Articles were searched for using MEDLINE, EMBASE, and the Cochrane Register. Included study participants were adult patients with high grade glioma who had undergone standard treatment (maximal resection, radiotherapy with concomitant and adjuvant temozolomide) and subsequently underwent follow-up imaging to determine treatment response status.\n\n- [Glioblastoma Multiforme Prognosis: MRI Missing Modality Generation, Segmentation and Radiogenomic Survival Prediction](https://arxiv.org/abs/2104.01149) -  The proposed approaches are evaluated by comparative experiments with state-of-the-art models in synthesis, segmentation, and overall survival (OS) prediction. We observe that adding missing MRI modality improves the segmentation prediction, and expression levels of gene markers have a high contribution in the GBM prognosis prediction, and fused radiogenomic features boost the OS estimation.\n\n- [Feature Selection for Imbalanced Data with Deep Sparse Autoencoders Ensemble](https://arxiv.org/abs/2103.11678) - Class imbalance is a common issue in many domain applications of learning algorithms. Oftentimes, in the same domains it is much more relevant to correctly classify and profile minority class observations. This need can be addressed by Feature Selection (FS), that offers several further advantages, s.a. decreasing computational costs, aiding inference and interpretability. However, traditional FS techniques may become sub-optimal in the presence of strongly imbalanced data. To achieve FS advantages in this setting, we propose a filtering FS algorithm ranking feature importance on the basis of the Reconstruction Error of a Deep Sparse AutoEncoders Ensemble (DSAEE).\n\n- [Spatial-And-Context aware (SpACe) \"virtual biopsy\" radiogenomic maps to target tumor mutational status on structural MRI](https://arxiv.org/abs/2006.09878) - With growing emphasis on personalized cancer-therapies,radiogenomics has shown promise in identifying target tumor mutational status on routine imaging (i.e. MRI) scans. These approaches fall into 2 categories: (1) deep-learning/radiomics (context-based), using image features from the entire tumor to identify the gene mutation status, or (2) atlas (spatial)-based to obtain likelihood of gene mutation status based on population statistics. While many genes (i.e. EGFR, MGMT) are spatially variant, a significant challenge in reliable assessment of gene mutation status on imaging has been the lack of available co-localized ground truth for training the models.\n\n- [Correlation via synthesis: end-to-end nodule image generation and radiogenomic map learning based on generative adversarial network](https://arxiv.org/abs/1907.03728) - Radiogenomic map linking image features and gene expression profiles is useful for noninvasively identifying molecular properties of a particular type of disease. Conventionally, such map is produced in three separate steps: 1) gene-clustering to \"metagenes\", 2) image feature extraction, and 3) statistical correlation between metagenes and image features. Each step is independently performed and relies on arbitrary measurements. In this work, we investigate the potential of an end-to-end method fusing gene data with image features to generate synthetic image and learn radiogenomic map simultaneously. To achieve this goal, we develop a generative adversarial network (GAN) conditioned on both background images and gene expression profiles, synthesizing the corresponding image. Image and gene features are fused at different scales to ensure the realism and quality of the synthesized image. We tested our method on non-small cell lung cancer (NSCLC) dataset. Results demonstrate that the proposed method produces realistic synthetic images, and provides a promising way to find gene-image relationship in a holistic end-to-end manner.\n\n- [The RSNA-ASNR-MICCAI BraTS 2021 Benchmark on Brain Tumor Segmentation and Radiogenomic Classification](https://arxiv.org/abs/2107.02314) - The BraTS 2021 challenge celebrates its 10th anniversary and is jointly organized by the Radiological Society of North America (RSNA), the American Society of Neuroradiology (ASNR), and the Medical Image Computing and Computer Assisted Interventions (MICCAI) society. Since its inception, BraTS has been focusing on being a common benchmarking venue for brain glioma segmentation algorithms, with well-curated multi-institutional multi-parametric magnetic resonance imaging (mpMRI) data.\n\n- [Tumor Radiogenomics with Bayesian Layered Variable Selection](https://arxiv.org/abs/2106.10941) - We propose a statistical framework to integrate radiological magnetic resonance imaging (MRI) and genomic data to identify the underlying radiogenomic associations in lower grade gliomas (LGG). We devise a novel imaging phenotype by dividing the tumor region into concentric spherical layers that mimics the tumor evolution process. MRI data within each layer is represented by voxel--intensity-based probability density functions which capture the complete information about tumor heterogeneity.\n\n**Added 07/14/2021:**\n- [Machine learning approaches to study glioblastoma: A review of the last decade of applications](https://onlinelibrary.wiley.com/doi/epdf/10.1002/cnr2.1226)\n\n- [Unsupervised machine learning reveals risk stratifying glioblastoma tumor cells](https://pubmed.ncbi.nlm.nih.gov/32573435/)",
    "1390627": "You might also want to consider the following for:\n- a broader radiogenomics review: [Imaging signatures of glioblastoma molecular characteristics: A radiogenomics review](https://pubmed.ncbi.nlm.nih.gov/31456318/)\n- a review on MRI for glioblastoma: [Advanced magnetic resonance imaging in glioblastoma: a review](https://pubmed.ncbi.nlm.nih.gov/28841802/)\n\n ",
    "1464637": "MRI-Based Deep-Learning Method for Determining Glioma MGMT Promoter Methylation Status\nhttps://doi.org/10.3174/ajnr.A7029\nAmerican Journal of Neuroradiology May 2021\n\n> predicting MGMT methylation status with a sensitivity and specificity of 96.31% [SD, 0.04%] and 91.66% [SD, 2.06%], respectively, and a mean area under the curve of 0.93 [SD, 0.01].",
    "1387101": "Really great starting points of research.  Thank you!",
    "1387076": "I was about to post papers, you beat me =) Nice List. ",
    "1390470": "@crained - great list - before I saw your discussion, I posted an abstract from and a link to a paper discussing **Automatic Prediction of MGMT Status in Glioblastoma**\n\nHere is the link to the discussion.\n\nhttps://www.kaggle.com/c/rsna-miccai-brain-tumor-radiogenomic-classification/discussion/253451\n\n",
    "1389476": "Hi,\n\nI found this one. May be interesting:\n\nhttps://www.nature.com/articles/s41598-021-90428-8\n\n",
    "1387131": "Thanks for sharing ...can we share paid papers ?? just asking ..",
    "1495705": "@crained Excellent Share, i was searching for some papers Thanks 👍",
    "1495631": "Thanks a lot for sharing - very useful indeed.",
    "1455594": "Let me paste this paragraph:\n\nThus, we restricted the current analysis in GBM patients\nwith wild-type IDH. The purpose of our study was to seek\ncertain variables derived from conventional structural\nimage features including **multifocal, tumor cross midline,\ntumor location, enhancement, cyst, necrosis, edema, side**,\nwhich may reflect MGMT promoter methylation status.\n\nfrom the following paper:\n\nhttps://pubmed.ncbi.nlm.nih.gov/29467012/\n\nHope you find it useful.",
    "1390004": "great work! thx for sharing ",
    "1388433": "Good day, \nI think this is some good back ground reading and source if I posted in the wrong forum please advise as I’m new to platform.\nhttps://scholar.google.com/scholar?q=MGMT+promoter+methylation&hl=en&as_sdt=0&as_vis=1&oi=scholart#d=gs_qabs&u=%23p%3D4uQBUNP3C1sJ",
    "1388367": "Updated with a few more papers. I'll add dates so if you come back you can see what are new papers. I have never done this before so hope it is helpful. Let me know. Enjoy!",
    "1388299": "R Packages mentioned in practitioner articles, open access:\n\n[MGMT promoter methylation](https://academic.oup.com/noa/article/2/1/vdaa117/5983632)\n[Pediatric Neuro-oncology diagnostic workflow implementation France](https://www.mdpi.com/2072-6694/13/6/1377)\n\n",
    "1387999": "Great One...",
    "1387813": "Nice sharing",
    "1387453": "Awesome listing! Thx for sharing :) ",
    "1389926": "",
    "1388413": "",
    "1495910": "Very helpful! Thank u for sharing =))",
    "1388526": "Thanks for sharing",
    "1495524": "Thank for sharing!",
    "1471064": "Thank you for sharing!",
    "1457395": "Thanks for sharing @crained ",
    "1455801": "Nice research thank you ",
    "1401731": "Great work Thanks for sharing !",
    "1392168": "Great Thanks for sharing !",
    "1387900": "Awesome list. Thanks for sharing 👍"
  }
}