Dear Host:
\nI have i legit question, looking in the data, I only saw images and labels without any genetic information, then, how are we going to be able to \"predict the status of a genetic biomarker important for brain cancer treatment\"?. My background is on molecular biology and i'm very interested on know the answer of my question.
Dear Host:
\nI have i legit question, looking in the data, I only saw images and labels without any genetic information, then, how are we going to be able to \"predict the status of a genetic biomarker important for brain cancer treatment\"?. My background is on molecular biology and i'm very interested on know the answer of my question.
Hiram, you are absolutely correct - MGMT is not a specific genetic subclass of glioblastoma like IDH1 and 1p19q, it is more of a \"epigenetic\" regulation of the key enzyme involved in the pathway that has been found to have clinical significance both in the response to therapy and prognosis.
", "rawMarkdown": "Hiram, you are absolutely correct - MGMT is not a specific genetic subclass of glioblastoma like IDH1 and 1p19q, it is more of a \"epigenetic\" regulation of the key enzyme involved in the pathway that has been found to have clinical significance both in the response to therapy and prognosis. ", "votes": 3 }, { "id": 1388792, "postDate": "2021-07-15T08:26:28.973Z", "content": "As previous participants said the aim is to make a binary classification to determine if the MGMT promotor alteration is present or not. there are several papers in the discussion posted here https://www.kaggle.com/c/rsna-miccai-brain-tumor-radiogenomic-classification/discussion/252833, hope it helps!
", "rawMarkdown": "As previous participants said the aim is to make a binary classification to determine if the MGMT promotor alteration is present or not. there are several papers in the discussion posted here https://www.kaggle.com/c/rsna-miccai-brain-tumor-radiogenomic-classification/discussion/252833, hope it helps! ", "votes": 1 }, { "id": 1387220, "postDate": "2021-07-14T03:39:27.250Z", "content": "@hiramcho : this is a paper that we published few years ago, I hope it helps: https://pubmed.ncbi.nlm.nih.gov/25476518/
", "rawMarkdown": "@hiramcho : this is a paper that we published few years ago, I hope it helps: https://pubmed.ncbi.nlm.nih.gov/25476518/", "votes": 1 }, { "id": 1387135, "postDate": "2021-07-14T01:49:55.323Z", "content": "I don't think there is a genetic component specifically. Seems to me it's a binary image classification problem. The goal is to classify tumor types/locations as either having MGMT or not.
", "rawMarkdown": "I don't think there is a genetic component specifically. Seems to me it's a binary image classification problem. The goal is to classify tumor types/locations as either having MGMT or not.", "votes": 2, "replies": [ { "id": 1387149, "postDate": "2021-07-14T02:17:11.003Z", "content": "Perhaps topic author's intention is to ask whether the macroscopic view of MRI can reveal the microscopic changes at the genetic level of MGMT gene methylation.
\nIn fact, it is currently impossible to tell from the images, but with deep learning, we may be able to grasp something that even humans cannot understand.
Hiram, you are absolutely correct - MGMT is not a specific genetic subclass of glioblastoma like IDH1 and 1p19q, it is more of a \"epigenetic\" regulation of the key enzyme involved in the pathway that has been found to have clinical significance both in the response to therapy and prognosis.
", "votes": 3, "replies": [] }, { "id": 1388792, "author_name": "Ranafago", "author_url": "", "post_date": "2021-07-15T08:26:28.973000", "content": "As previous participants said the aim is to make a binary classification to determine if the MGMT promotor alteration is present or not. there are several papers in the discussion posted here https://www.kaggle.com/c/rsna-miccai-brain-tumor-radiogenomic-classification/discussion/252833, hope it helps!
", "votes": 1, "replies": [] }, { "id": 1387220, "author_name": "Elbanan", "author_url": "", "post_date": "2021-07-14T03:39:27.250000", "content": "@hiramcho : this is a paper that we published few years ago, I hope it helps: https://pubmed.ncbi.nlm.nih.gov/25476518/
", "votes": 1, "replies": [] }, { "id": 1387135, "author_name": "David Roberts", "author_url": "", "post_date": "2021-07-14T01:49:55.323000", "content": "I don't think there is a genetic component specifically. Seems to me it's a binary image classification problem. The goal is to classify tumor types/locations as either having MGMT or not.
", "votes": 2, "replies": [ { "id": 1387149, "author_name": "dlbsabu", "author_url": "", "post_date": "2021-07-14T02:17:11.003000", "content": "Perhaps topic author's intention is to ask whether the macroscopic view of MRI can reveal the microscopic changes at the genetic level of MGMT gene methylation.
\nIn fact, it is currently impossible to tell from the images, but with deep learning, we may be able to grasp something that even humans cannot understand.