{
  "id": 427232,
  "title": "Computer Vision in Healthcare (formats) and Highlights of the Last RSNA Competition.",
  "url": "/competitions/rsna-2023-abdominal-trauma-detection/discussion/427232",
  "author_name": "Marília Prata",
  "post_date": "2023-07-27T01:10:32.280000",
  "votes": 44,
  "comment_count": 3,
  "views": 0,
  "content": "<h1>Highlights of Last RSNA Competition</h1>\n<p>Marking the first month - Summary of key discussions - By the Devastator <br>\n<a href=\"https://www.kaggle.com/competitions/rsna-breast-cancer-detection/discussion/374248\" target=\"_blank\">https://www.kaggle.com/competitions/rsna-breast-cancer-detection/discussion/374248</a></p>\n<p>6 Computer Vision tricks for faster training and better models - By Radek Osmulski<br>\n<a href=\"https://www.kaggle.com/competitions/rsna-breast-cancer-detection/discussion/369155\" target=\"_blank\">https://www.kaggle.com/competitions/rsna-breast-cancer-detection/discussion/369155</a></p>\n<p>[fast.ai starter pack] train + inference  - Code by Radek Osmulski<br>\n<a href=\"https://www.kaggle.com/code/radek1/fast-ai-starter-pack-train-inference\" target=\"_blank\">https://www.kaggle.com/code/radek1/fast-ai-starter-pack-train-inference</a></p>\n<p>Dicom -&gt; Resized PNG/JPG - Code by Theo Viel<br>\n<a href=\"https://www.kaggle.com/code/theoviel/dicom-resized-png-jpg/\" target=\"_blank\">https://www.kaggle.com/code/theoviel/dicom-resized-png-jpg/</a></p>\n<p>ROI extracted dataset - resolution 768pix and 1024pix⭐️ -&gt; 2023.01.14 = 1280 (windowing) By Remek Kinas<br>\n<a href=\"https://www.kaggle.com/competitions/rsna-breast-cancer-detection/discussion/369754\" target=\"_blank\">https://www.kaggle.com/competitions/rsna-breast-cancer-detection/discussion/369754</a></p>\n<p>Mammography - apply windowing -By David Roberts<br>\n<a href=\"https://www.kaggle.com/code/davidbroberts/mammography-apply-windowing/notebook\" target=\"_blank\">https://www.kaggle.com/code/davidbroberts/mammography-apply-windowing/notebook</a></p>\n<h1>Medical Imaging Standards Used in Computer Vision Models: DICOM &amp; NIfTI</h1>\n<p>\"Medical imaging and annotation is a specialized field. Perhaps more so than any others, accuracy is crucial. When we consider the end-users, such as healthcare professionals, and the ultimate outcomes, the impact on patients, we can see why accuracy is crucial.\"</p>\n<p>\"Computer vision models are powerful Artificial Intelligence and Machine Learning-based (AI/ML) software tools for analyzing images.\"</p>\n<h1>Difference between the DICOM format and JPEG?</h1>\n<p>‍\"One of the most common imaging formats is JPEG (Joint Photographic Experts Group), and although widely used the world over, it’s not practical or useful in a medical setting. DICOM files contain layers and layers of images, associated metadata, and links to databases and other medical systems.\"</p>\n<p>\"On the other hand, JPEG files are single-layer 2D images. Medical images in a JPEG format wouldn’t be detailed nor useful enough for medical purposes. Although you can convert DICOM and other files into JPEGs, this is usually convenient when explaining something in simple terms to a patient.\"  </p>\n<h1>Difference between DICOM and PACS?</h1>\n<p>\"In most healthcare workplaces, doctors and specialists also use the Picture Archiving and Communication System, or PACS, alongside other imaging formats. PACS is used as a medical image storage and archive system, with images being fed into by radiologists and other medical specialists. Images usually come from X-ray machines and MRI scanners.\"</p>\n<p>\"On the other hand, the DICOM format is an international communication standard for storing, communicating and transmitting medical images with layers of metadata. Medical professionals can use both, with one format supporting the other to ensure every stakeholder involved in patient care has the necessary information.\"</p>\n<h1>Best Practices for Using DICOM and NIfTI File Format in Computer Vision Models</h1>\n<ul>\n<li>Display the data correctly to allow for pixel-perfect annotations</li>\n<li>Ensure high levels of medical image annotation quality for computer vision models</li>\n<li>Make data audits granular: Mission-critical for healthcare regulatory compliance</li>\n<li>Improve image and video annotation efficiency with automation, to save radiologists valuable time</li>\n</ul>\n<p><a href=\"https://encord.com/blog/dicom-and-nifti-files-annotation-guide/\" target=\"_blank\">https://encord.com/blog/dicom-and-nifti-files-annotation-guide/</a><br>\n<a href=\"https://encord.com/healthcare/\" target=\"_blank\">https://encord.com/healthcare/</a></p>",
  "messages": [
    {
      "id": 2360736,
      "postDate": "2023-07-27T01:10:32.280Z",
      "content": "<h1>Highlights of Last RSNA Competition</h1>\n<p>Marking the first month - Summary of key discussions - By the Devastator <br>\n<a href=\"https://www.kaggle.com/competitions/rsna-breast-cancer-detection/discussion/374248\" target=\"_blank\">https://www.kaggle.com/competitions/rsna-breast-cancer-detection/discussion/374248</a></p>\n<p>6 Computer Vision tricks for faster training and better models - By Radek Osmulski<br>\n<a href=\"https://www.kaggle.com/competitions/rsna-breast-cancer-detection/discussion/369155\" target=\"_blank\">https://www.kaggle.com/competitions/rsna-breast-cancer-detection/discussion/369155</a></p>\n<p>[fast.ai starter pack] train + inference  - Code by Radek Osmulski<br>\n<a href=\"https://www.kaggle.com/code/radek1/fast-ai-starter-pack-train-inference\" target=\"_blank\">https://www.kaggle.com/code/radek1/fast-ai-starter-pack-train-inference</a></p>\n<p>Dicom -&gt; Resized PNG/JPG - Code by Theo Viel<br>\n<a href=\"https://www.kaggle.com/code/theoviel/dicom-resized-png-jpg/\" target=\"_blank\">https://www.kaggle.com/code/theoviel/dicom-resized-png-jpg/</a></p>\n<p>ROI extracted dataset - resolution 768pix and 1024pix⭐️ -&gt; 2023.01.14 = 1280 (windowing) By Remek Kinas<br>\n<a href=\"https://www.kaggle.com/competitions/rsna-breast-cancer-detection/discussion/369754\" target=\"_blank\">https://www.kaggle.com/competitions/rsna-breast-cancer-detection/discussion/369754</a></p>\n<p>Mammography - apply windowing -By David Roberts<br>\n<a href=\"https://www.kaggle.com/code/davidbroberts/mammography-apply-windowing/notebook\" target=\"_blank\">https://www.kaggle.com/code/davidbroberts/mammography-apply-windowing/notebook</a></p>\n<h1>Medical Imaging Standards Used in Computer Vision Models: DICOM &amp; NIfTI</h1>\n<p>\"Medical imaging and annotation is a specialized field. Perhaps more so than any others, accuracy is crucial. When we consider the end-users, such as healthcare professionals, and the ultimate outcomes, the impact on patients, we can see why accuracy is crucial.\"</p>\n<p>\"Computer vision models are powerful Artificial Intelligence and Machine Learning-based (AI/ML) software tools for analyzing images.\"</p>\n<h1>Difference between the DICOM format and JPEG?</h1>\n<p>‍\"One of the most common imaging formats is JPEG (Joint Photographic Experts Group), and although widely used the world over, it’s not practical or useful in a medical setting. DICOM files contain layers and layers of images, associated metadata, and links to databases and other medical systems.\"</p>\n<p>\"On the other hand, JPEG files are single-layer 2D images. Medical images in a JPEG format wouldn’t be detailed nor useful enough for medical purposes. Although you can convert DICOM and other files into JPEGs, this is usually convenient when explaining something in simple terms to a patient.\"  </p>\n<h1>Difference between DICOM and PACS?</h1>\n<p>\"In most healthcare workplaces, doctors and specialists also use the Picture Archiving and Communication System, or PACS, alongside other imaging formats. PACS is used as a medical image storage and archive system, with images being fed into by radiologists and other medical specialists. Images usually come from X-ray machines and MRI scanners.\"</p>\n<p>\"On the other hand, the DICOM format is an international communication standard for storing, communicating and transmitting medical images with layers of metadata. Medical professionals can use both, with one format supporting the other to ensure every stakeholder involved in patient care has the necessary information.\"</p>\n<h1>Best Practices for Using DICOM and NIfTI File Format in Computer Vision Models</h1>\n<ul>\n<li>Display the data correctly to allow for pixel-perfect annotations</li>\n<li>Ensure high levels of medical image annotation quality for computer vision models</li>\n<li>Make data audits granular: Mission-critical for healthcare regulatory compliance</li>\n<li>Improve image and video annotation efficiency with automation, to save radiologists valuable time</li>\n</ul>\n<p><a href=\"https://encord.com/blog/dicom-and-nifti-files-annotation-guide/\" target=\"_blank\">https://encord.com/blog/dicom-and-nifti-files-annotation-guide/</a><br>\n<a href=\"https://encord.com/healthcare/\" target=\"_blank\">https://encord.com/healthcare/</a></p>",
      "rawMarkdown": "#Highlights of Last RSNA Competition\n\nMarking the first month - Summary of key discussions - By the Devastator \nhttps://www.kaggle.com/competitions/rsna-breast-cancer-detection/discussion/374248\n\n6 Computer Vision tricks for faster training and better models - By Radek Osmulski\nhttps://www.kaggle.com/competitions/rsna-breast-cancer-detection/discussion/369155\n\n[fast.ai starter pack] train + inference  - Code by Radek Osmulski\nhttps://www.kaggle.com/code/radek1/fast-ai-starter-pack-train-inference\n\nDicom -> Resized PNG/JPG - Code by Theo Viel\nhttps://www.kaggle.com/code/theoviel/dicom-resized-png-jpg/\n\nROI extracted dataset - resolution 768pix and 1024pix⭐️ -> 2023.01.14 = 1280 (windowing) By Remek Kinas\nhttps://www.kaggle.com/competitions/rsna-breast-cancer-detection/discussion/369754\n\nMammography - apply windowing -By David Roberts\nhttps://www.kaggle.com/code/davidbroberts/mammography-apply-windowing/notebook\n\n#Medical Imaging Standards Used in Computer Vision Models: DICOM & NIfTI\n\n\"Medical imaging and annotation is a specialized field. Perhaps more so than any others, accuracy is crucial. When we consider the end-users, such as healthcare professionals, and the ultimate outcomes, the impact on patients, we can see why accuracy is crucial.\"\n\n\"Computer vision models are powerful Artificial Intelligence and Machine Learning-based (AI/ML) software tools for analyzing images.\"\n\n#Difference between the DICOM format and JPEG?\n\n‍\"One of the most common imaging formats is JPEG (Joint Photographic Experts Group), and although widely used the world over, it’s not practical or useful in a medical setting. DICOM files contain layers and layers of images, associated metadata, and links to databases and other medical systems.\"\n\n\"On the other hand, JPEG files are single-layer 2D images. Medical images in a JPEG format wouldn’t be detailed nor useful enough for medical purposes. Although you can convert DICOM and other files into JPEGs, this is usually convenient when explaining something in simple terms to a patient.\"  \n\n#Difference between DICOM and PACS?\n\n\"In most healthcare workplaces, doctors and specialists also use the Picture Archiving and Communication System, or PACS, alongside other imaging formats. PACS is used as a medical image storage and archive system, with images being fed into by radiologists and other medical specialists. Images usually come from X-ray machines and MRI scanners.\"\n\n\"On the other hand, the DICOM format is an international communication standard for storing, communicating and transmitting medical images with layers of metadata. Medical professionals can use both, with one format supporting the other to ensure every stakeholder involved in patient care has the necessary information.\"\n\n#Best Practices for Using DICOM and NIfTI File Format in Computer Vision Models\n\n- Display the data correctly to allow for pixel-perfect annotations\n- Ensure high levels of medical image annotation quality for computer vision models\n- Make data audits granular: Mission-critical for healthcare regulatory compliance\n- Improve image and video annotation efficiency with automation, to save radiologists valuable time\n\nhttps://encord.com/blog/dicom-and-nifti-files-annotation-guide/\nhttps://encord.com/healthcare/",
      "votes": 44
    },
    {
      "id": 2658860,
      "postDate": "2024-02-19T13:10:02.460Z",
      "content": "<p>I was searching for 'computer vision applications in the healthcare industry' and found your post.  <br>\nThank you for sharing..!!</p>",
      "rawMarkdown": " I was searching for 'computer vision applications in the healthcare industry' and found your post.  \nThank you for sharing..!!"
    },
    {
      "id": 2361217,
      "postDate": "2023-07-27T09:13:03.220Z",
      "rawMarkdown": "",
      "isDeleted": true
    },
    {
      "id": 2368669,
      "postDate": "2023-08-01T09:58:12.170Z",
      "content": "<p>Thanks for the link and suggestion.</p>",
      "rawMarkdown": "Thanks for the link and suggestion."
    }
  ],
  "comments": [
    {
      "id": 2658860,
      "author_name": "Anmol Arora",
      "author_url": "",
      "post_date": "2024-02-19T13:10:02.460000",
      "content": "<p>I was searching for 'computer vision applications in the healthcare industry' and found your post.  <br>\nThank you for sharing..!!</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 2361217,
      "author_name": "",
      "author_url": "",
      "post_date": "2023-07-27T09:13:03.220000",
      "content": "",
      "votes": 0,
      "replies": []
    },
    {
      "id": 2368669,
      "author_name": "Yang.Xu21",
      "author_url": "",
      "post_date": "2023-08-01T09:58:12.170000",
      "content": "<p>Thanks for the link and suggestion.</p>",
      "votes": 0,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2360736": "#Highlights of Last RSNA Competition\n\nMarking the first month - Summary of key discussions - By the Devastator \nhttps://www.kaggle.com/competitions/rsna-breast-cancer-detection/discussion/374248\n\n6 Computer Vision tricks for faster training and better models - By Radek Osmulski\nhttps://www.kaggle.com/competitions/rsna-breast-cancer-detection/discussion/369155\n\n[fast.ai starter pack] train + inference  - Code by Radek Osmulski\nhttps://www.kaggle.com/code/radek1/fast-ai-starter-pack-train-inference\n\nDicom -> Resized PNG/JPG - Code by Theo Viel\nhttps://www.kaggle.com/code/theoviel/dicom-resized-png-jpg/\n\nROI extracted dataset - resolution 768pix and 1024pix⭐️ -> 2023.01.14 = 1280 (windowing) By Remek Kinas\nhttps://www.kaggle.com/competitions/rsna-breast-cancer-detection/discussion/369754\n\nMammography - apply windowing -By David Roberts\nhttps://www.kaggle.com/code/davidbroberts/mammography-apply-windowing/notebook\n\n#Medical Imaging Standards Used in Computer Vision Models: DICOM & NIfTI\n\n\"Medical imaging and annotation is a specialized field. Perhaps more so than any others, accuracy is crucial. When we consider the end-users, such as healthcare professionals, and the ultimate outcomes, the impact on patients, we can see why accuracy is crucial.\"\n\n\"Computer vision models are powerful Artificial Intelligence and Machine Learning-based (AI/ML) software tools for analyzing images.\"\n\n#Difference between the DICOM format and JPEG?\n\n‍\"One of the most common imaging formats is JPEG (Joint Photographic Experts Group), and although widely used the world over, it’s not practical or useful in a medical setting. DICOM files contain layers and layers of images, associated metadata, and links to databases and other medical systems.\"\n\n\"On the other hand, JPEG files are single-layer 2D images. Medical images in a JPEG format wouldn’t be detailed nor useful enough for medical purposes. Although you can convert DICOM and other files into JPEGs, this is usually convenient when explaining something in simple terms to a patient.\"  \n\n#Difference between DICOM and PACS?\n\n\"In most healthcare workplaces, doctors and specialists also use the Picture Archiving and Communication System, or PACS, alongside other imaging formats. PACS is used as a medical image storage and archive system, with images being fed into by radiologists and other medical specialists. Images usually come from X-ray machines and MRI scanners.\"\n\n\"On the other hand, the DICOM format is an international communication standard for storing, communicating and transmitting medical images with layers of metadata. Medical professionals can use both, with one format supporting the other to ensure every stakeholder involved in patient care has the necessary information.\"\n\n#Best Practices for Using DICOM and NIfTI File Format in Computer Vision Models\n\n- Display the data correctly to allow for pixel-perfect annotations\n- Ensure high levels of medical image annotation quality for computer vision models\n- Make data audits granular: Mission-critical for healthcare regulatory compliance\n- Improve image and video annotation efficiency with automation, to save radiologists valuable time\n\nhttps://encord.com/blog/dicom-and-nifti-files-annotation-guide/\nhttps://encord.com/healthcare/",
    "2658860": " I was searching for 'computer vision applications in the healthcare industry' and found your post.  \nThank you for sharing..!!",
    "2361217": "",
    "2368669": "Thanks for the link and suggestion."
  }
}