{
  "id": 376378,
  "title": "Previous research by Nan Wu and others",
  "url": "/competitions/rsna-breast-cancer-detection/discussion/376378",
  "author_name": "Tanrei(nama)",
  "post_date": "2023-01-06T01:27:33.251000",
  "votes": 0,
  "comment_count": 1,
  "views": 0,
  "content": "<p>I found Previous research of Breast Cancer Detection</p>\n<p>Unfortunately I couldn't find the dataset, but there is a trained model on GitHub</p>\n<p><a href=\"https://github.com/nyukat/breast_cancer_classifier\" target=\"_blank\">https://github.com/nyukat/breast_cancer_classifier</a></p>\n<p>and the peper<br>\n<a href=\"https://ieeexplore.ieee.org/document/8861376\" target=\"_blank\">https://ieeexplore.ieee.org/document/8861376</a></p>\n<p>and test run in Kaggle kernel.<br>\n<a href=\"https://www.kaggle.com/tanreinama/nyukat-breast-cancer-classifier\" target=\"_blank\">https://www.kaggle.com/tanreinama/nyukat-breast-cancer-classifier</a></p>\n<p>Since there are many test images in the competition, it is not possible to perform it in a perfect form. Object detection and clipping steps are omitted.<br>\nStrangely, the leaderboard scores are very low. Do you know why?</p>",
  "messages": [
    {
      "id": 2088252,
      "postDate": "2023-01-06T07:50:51.903Z",
      "content": "<p>I tried that on train.  The results seems pretty good, but you're right, too slow.</p>\n<p>Not so sound flippant, truly, but perhaps the scores you see are low because you omitted some key steps?  Or am I missing something?</p>",
      "rawMarkdown": "I tried that on train.  The results seems pretty good, but you're right, too slow.\n\nNot so sound flippant, truly, but perhaps the scores you see are low because you omitted some key steps?  Or am I missing something?"
    },
    {
      "id": 2088003,
      "postDate": "2023-01-06T01:27:33.250Z",
      "content": "<p>I found Previous research of Breast Cancer Detection</p>\n<p>Unfortunately I couldn't find the dataset, but there is a trained model on GitHub</p>\n<p><a href=\"https://github.com/nyukat/breast_cancer_classifier\" target=\"_blank\">https://github.com/nyukat/breast_cancer_classifier</a></p>\n<p>and the peper<br>\n<a href=\"https://ieeexplore.ieee.org/document/8861376\" target=\"_blank\">https://ieeexplore.ieee.org/document/8861376</a></p>\n<p>and test run in Kaggle kernel.<br>\n<a href=\"https://www.kaggle.com/tanreinama/nyukat-breast-cancer-classifier\" target=\"_blank\">https://www.kaggle.com/tanreinama/nyukat-breast-cancer-classifier</a></p>\n<p>Since there are many test images in the competition, it is not possible to perform it in a perfect form. Object detection and clipping steps are omitted.<br>\nStrangely, the leaderboard scores are very low. Do you know why?</p>",
      "rawMarkdown": "I found Previous research of Breast Cancer Detection\n\nUnfortunately I couldn't find the dataset, but there is a trained model on GitHub\n\nhttps://github.com/nyukat/breast_cancer_classifier\n\nand the peper\nhttps://ieeexplore.ieee.org/document/8861376\n\nand test run in Kaggle kernel.\nhttps://www.kaggle.com/tanreinama/nyukat-breast-cancer-classifier\n\nSince there are many test images in the competition, it is not possible to perform it in a perfect form. Object detection and clipping steps are omitted.\nStrangely, the leaderboard scores are very low. Do you know why?\n"
    }
  ],
  "comments": [
    {
      "id": 2088252,
      "author_name": "@kaggleqrdl",
      "author_url": "",
      "post_date": "2023-01-06T07:50:51.903000",
      "content": "<p>I tried that on train.  The results seems pretty good, but you're right, too slow.</p>\n<p>Not so sound flippant, truly, but perhaps the scores you see are low because you omitted some key steps?  Or am I missing something?</p>",
      "votes": 0,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2088252": "I tried that on train.  The results seems pretty good, but you're right, too slow.\n\nNot so sound flippant, truly, but perhaps the scores you see are low because you omitted some key steps?  Or am I missing something?",
    "2088003": "I found Previous research of Breast Cancer Detection\n\nUnfortunately I couldn't find the dataset, but there is a trained model on GitHub\n\nhttps://github.com/nyukat/breast_cancer_classifier\n\nand the peper\nhttps://ieeexplore.ieee.org/document/8861376\n\nand test run in Kaggle kernel.\nhttps://www.kaggle.com/tanreinama/nyukat-breast-cancer-classifier\n\nSince there are many test images in the competition, it is not possible to perform it in a perfect form. Object detection and clipping steps are omitted.\nStrangely, the leaderboard scores are very low. Do you know why?\n"
  }
}