{
  "id": 65831,
  "title": "Pneumonia Detection without Neural Networks",
  "url": "/competitions/rsna-pneumonia-detection-challenge/discussion/65831",
  "author_name": "",
  "post_date": "2018-09-15T09:35:24.151218100Z",
  "votes": 1,
  "comment_count": 2,
  "views": 0,
  "content": "<p>Hi there! Is it possible to detect pneumonia without neural networks? I just get curious and want to compare NN and non-NN algorithms. I think it is possible to do it with Random Forest (maybe) but how data should be prepared for it?</p>",
  "messages": [
    {
      "id": "387623",
      "postDate": "09/15/2018 09:35:24",
      "content": "<p>Hi there! Is it possible to detect pneumonia without neural networks? I just get curious and want to compare NN and non-NN algorithms. I think it is possible to do it with Random Forest (maybe) but how data should be prepared for it?</p>",
      "rawMarkdown": "Hi there! Is it possible to detect pneumonia without neural networks? I just get curious and want to compare NN and non-NN algorithms. I think it is possible to do it with Random Forest (maybe) but how data should be prepared for it?",
      "votes": null
    },
    {
      "id": "388003",
      "postDate": "09/16/2018 04:53:17",
      "content": "<p>Yes, it is! Here's a hint: you want to build features by taking a sliding window. You may also want to incorporate the output of a NN in your RF model as a feature. For example, do lung segmentation and incorporate that output in the RF model.</p>",
      "rawMarkdown": "Yes, it is! Here's a hint: you want to build features by taking a sliding window. You may also want to incorporate the output of a NN in your RF model as a feature. For example, do lung segmentation and incorporate that output in the RF model.",
      "votes": null
    },
    {
      "id": "388067",
      "postDate": "09/16/2018 08:19:20",
      "content": "<p>Thanks a lot. I will try it</p>",
      "rawMarkdown": "Thanks a lot. I will try it",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 388003,
      "author_name": "puremath86",
      "author_url": "",
      "post_date": "09/16/2018 04:53:17",
      "content": "<p>Yes, it is! Here's a hint: you want to build features by taking a sliding window. You may also want to incorporate the output of a NN in your RF model as a feature. For example, do lung segmentation and incorporate that output in the RF model.</p>",
      "votes": null,
      "replies": [
        {
          "id": 388067,
          "author_name": "alpamys",
          "author_url": "",
          "post_date": "09/16/2018 08:19:20",
          "content": "<p>Thanks a lot. I will try it</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "387623": "Hi there! Is it possible to detect pneumonia without neural networks? I just get curious and want to compare NN and non-NN algorithms. I think it is possible to do it with Random Forest (maybe) but how data should be prepared for it?",
    "388003": "Yes, it is! Here's a hint: you want to build features by taking a sliding window. You may also want to incorporate the output of a NN in your RF model as a feature. For example, do lung segmentation and incorporate that output in the RF model.",
    "388067": "Thanks a lot. I will try it"
  },
  "source": "meta"
}