what does automating nucleus detection means and how it is useful? I know less about biology but how detecting nucleus would help in curing? If their is any deformity or irregular shape then it is understandable. Can anyone explain in detail what we have to detect and how it is useful? Thanks
", "messages": [ { "id": "292549", "postDate": "03/08/2018 06:19:05", "content": "what does automating nucleus detection means and how it is useful? I know less about biology but how detecting nucleus would help in curing? If their is any deformity or irregular shape then it is understandable. Can anyone explain in detail what we have to detect and how it is useful? Thanks
", "rawMarkdown": "what does automating nucleus detection means and how it is useful? I know less about biology but how detecting nucleus would help in curing? If their is any deformity or irregular shape then it is understandable. Can anyone explain in detail what we have to detect and how it is useful? Thanks", "votes": null }, { "id": "292946", "postDate": "03/08/2018 23:46:56", "content": "We are detecting nuclei simply to count them. I have done lots of microscopy and quantitative image analysis and, I assure you, obtaining an accurate cell count is not easy (mostly because image data are not perfect, or standardized). One of the easiest ways to cure cancer is to detect it in its earliest stages. Cancer is characterized by rapid, uncontrolled cell division. This means that areas with many, densely packed nuclei are more likely to be cancerous than areas which have sparse nuclei that can be much larger (but still overlap). Both of these area types in a tissue can produce a similar level of signal in terms of size and intensity. So, how do you tell large overlapping nuclei from small densely packed nuclei without a trained pathologist actually looking at them? This is where the competition will assist with curing... through assisting with early detection of cancer- by identifying the number of nuclei in a given region.
", "rawMarkdown": "We are detecting nuclei simply to count them. I have done lots of microscopy and quantitative image analysis and, I assure you, obtaining an accurate cell count is not easy (mostly because image data are not perfect, or standardized). One of the easiest ways to cure cancer is to detect it in its earliest stages. Cancer is characterized by rapid, uncontrolled cell division. This means that areas with many, densely packed nuclei are more likely to be cancerous than areas which have sparse nuclei that can be much larger (but still overlap). Both of these area types in a tissue can produce a similar level of signal in terms of size and intensity. So, how do you tell large overlapping nuclei from small densely packed nuclei without a trained pathologist actually looking at them? This is where the competition will assist with curing... through assisting with early detection of cancer- by identifying the number of nuclei in a given region.", "votes": null }, { "id": "293414", "postDate": "03/09/2018 19:38:10", "content": "Take a look at the intro video: https://datasciencebowl.com/2018dsbtutorial/
", "rawMarkdown": "Take a look at the intro video: https://datasciencebowl.com/2018dsbtutorial/", "votes": null } ], "comments": [ { "id": 292946, "author_name": "bleuknight", "author_url": "", "post_date": "03/08/2018 23:46:56", "content": "We are detecting nuclei simply to count them. I have done lots of microscopy and quantitative image analysis and, I assure you, obtaining an accurate cell count is not easy (mostly because image data are not perfect, or standardized). One of the easiest ways to cure cancer is to detect it in its earliest stages. Cancer is characterized by rapid, uncontrolled cell division. This means that areas with many, densely packed nuclei are more likely to be cancerous than areas which have sparse nuclei that can be much larger (but still overlap). Both of these area types in a tissue can produce a similar level of signal in terms of size and intensity. So, how do you tell large overlapping nuclei from small densely packed nuclei without a trained pathologist actually looking at them? This is where the competition will assist with curing... through assisting with early detection of cancer- by identifying the number of nuclei in a given region.
", "votes": null, "replies": [] }, { "id": 293414, "author_name": "drannecarpenter", "author_url": "", "post_date": "03/09/2018 19:38:10", "content": "Take a look at the intro video: https://datasciencebowl.com/2018dsbtutorial/
", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "292549": "what does automating nucleus detection means and how it is useful? I know less about biology but how detecting nucleus would help in curing? If their is any deformity or irregular shape then it is understandable. Can anyone explain in detail what we have to detect and how it is useful? Thanks", "292946": "We are detecting nuclei simply to count them. I have done lots of microscopy and quantitative image analysis and, I assure you, obtaining an accurate cell count is not easy (mostly because image data are not perfect, or standardized). One of the easiest ways to cure cancer is to detect it in its earliest stages. Cancer is characterized by rapid, uncontrolled cell division. This means that areas with many, densely packed nuclei are more likely to be cancerous than areas which have sparse nuclei that can be much larger (but still overlap). Both of these area types in a tissue can produce a similar level of signal in terms of size and intensity. So, how do you tell large overlapping nuclei from small densely packed nuclei without a trained pathologist actually looking at them? This is where the competition will assist with curing... through assisting with early detection of cancer- by identifying the number of nuclei in a given region.", "293414": "Take a look at the intro video: https://datasciencebowl.com/2018dsbtutorial/" }, "source": "meta" }