run_on_idc_data.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "view-in-github"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/bamf-health/aimi-liver-ct/blob/main/run_on_idc_data.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "N4yyM1h5mlWB"
      },
      "source": [
        "Run AI segmentation on the TCGA-LIHC collection.\n",
        "\n",
        "Be sure to run this in a runtime with an attached GPU\n",
        "\n",
        "\n",
        "Querying and Download the CT scans is based on the [IDC  tutorial cookbook](https://github.com/ImagingDataCommons/IDC-Tutorials/blob/master/notebooks/cookbook.ipynb).\n",
        "\n",
        "## Prerequisites\n",
        "Please complete the prerequisites as described in this [documentation page](https://learn.canceridc.dev/introduction/getting-started-with-gcp).\n",
        "\n",
        "Insert that project ID in the cell below."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "CnaY1qR4FQxm"
      },
      "outputs": [],
      "source": [
        "#@title Enter your Project ID and authenticate with GCP\n",
        "# initialize this variable with your Google Cloud Project ID!\n",
        "my_ProjectID = ''\n",
        "\n",
        "import os\n",
        "os.environ[\"GCP_PROJECT_ID\"] = my_ProjectID\n",
        "\n",
        "from google.colab import auth\n",
        "auth.authenticate_user()\n",
        "\n",
        "import pandas as pd\n",
        "import subprocess\n",
        "from pathlib import Path\n",
        "from tempfile import TemporaryDirectory\n",
        "import shutil\n",
        "from tqdm.auto import tqdm"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "upzH2aPmtzV_"
      },
      "source": [
        "Install the `s5cmd` tool for efficient manifest downloads"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rvYExdWZoFpK",
        "outputId": "bf59da8a-c513-4dc7-de19-2db0e9ea0e6a"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "--2023-09-27 19:18:11--  https://github.com/peak/s5cmd/releases/download/v2.2.2/s5cmd_2.2.2_Linux-64bit.tar.gz\n",
            "Resolving github.com (github.com)... 140.82.121.3\n",
            "Connecting to github.com (github.com)|140.82.121.3|:443... connected.\n",
            "HTTP request sent, awaiting response... 302 Found\n",
            "Location: https://objects.githubusercontent.com/github-production-release-asset-2e65be/73909333/e095ae85-9acf-4dcc-b744-128b3311849c?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIAIWNJYAX4CSVEH53A%2F20230927%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Date=20230927T191811Z&X-Amz-Expires=300&X-Amz-Signature=bcb23068c7132b163aacff0e8085514d73581873d1a116e099db30b299ac41cd&X-Amz-SignedHeaders=host&actor_id=0&key_id=0&repo_id=73909333&response-content-disposition=attachment%3B%20filename%3Ds5cmd_2.2.2_Linux-64bit.tar.gz&response-content-type=application%2Foctet-stream [following]\n",
            "--2023-09-27 19:18:11--  https://objects.githubusercontent.com/github-production-release-asset-2e65be/73909333/e095ae85-9acf-4dcc-b744-128b3311849c?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIAIWNJYAX4CSVEH53A%2F20230927%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Date=20230927T191811Z&X-Amz-Expires=300&X-Amz-Signature=bcb23068c7132b163aacff0e8085514d73581873d1a116e099db30b299ac41cd&X-Amz-SignedHeaders=host&actor_id=0&key_id=0&repo_id=73909333&response-content-disposition=attachment%3B%20filename%3Ds5cmd_2.2.2_Linux-64bit.tar.gz&response-content-type=application%2Foctet-stream\n",
            "Resolving objects.githubusercontent.com (objects.githubusercontent.com)... 185.199.110.133, 185.199.109.133, 185.199.111.133, ...\n",
            "Connecting to objects.githubusercontent.com (objects.githubusercontent.com)|185.199.110.133|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 4821452 (4.6M) [application/octet-stream]\n",
            "Saving to: ‘s5cmd_2.2.2_Linux-64bit.tar.gz’\n",
            "\n",
            "s5cmd_2.2.2_Linux-6 100%[===================>]   4.60M  --.-KB/s    in 0.1s    \n",
            "\n",
            "2023-09-27 19:18:11 (35.0 MB/s) - ‘s5cmd_2.2.2_Linux-64bit.tar.gz’ saved [4821452/4821452]\n",
            "\n"
          ]
        },
        {
          "data": {
            "text/plain": []
          },
          "execution_count": 2,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "%%shell\n",
        "VERSION=\"s5cmd_2.2.2_Linux-64bit\"\n",
        "wget https://github.com/peak/s5cmd/releases/download/v2.2.2/${VERSION}.tar.gz\n",
        "tar zxf ${VERSION}.tar.gz\n",
        "mv s5cmd /usr/bin"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gGJGZb_it_W9"
      },
      "source": [
        "Install dcm2niix for dicom conversion. Use the prebuild version because it was jpeg support."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "OeGOu3ePuMBd",
        "outputId": "1318d075-83f3-4796-8890-1fa37ece90e0"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current\n",
            "                                 Dload  Upload   Total   Spent    Left  Speed\n",
            "  0     0    0     0    0     0      0      0 --:--:-- --:--:-- --:--:--     0\n",
            "  0     0    0     0    0     0      0      0 --:--:-- --:--:-- --:--:--     0\n",
            "100  855k  100  855k    0     0  1013k      0 --:--:-- --:--:-- --:--:-- 1013k\n",
            "Archive:  dcm2niix_lnx.zip\n",
            "  inflating: /usr/bin/dcm2niix       \n"
          ]
        },
        {
          "data": {
            "text/plain": []
          },
          "execution_count": 3,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "%%shell\n",
        "curl -fLO https://github.com/rordenlab/dcm2niix/releases/latest/download/dcm2niix_lnx.zip\n",
        "unzip -o dcm2niix_lnx.zip -d /usr/bin"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HWklCU-UvSER"
      },
      "source": [
        "We can build a download manifest. To reproduce our results, get a list of the SeriesInstanceUIDs from the `qa-results.csv` file. This csv is in the liver-ct.zip file at https://zenodo.org/record/8352041. Alternativly, we can get a copy directly from the github repo"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "W08brbi1w-Yw",
        "outputId": "636e1d32-b8d5-41b8-de37-3e040b299ff6"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "--2023-09-27 19:18:16--  https://github.com/bamf-health/aimi-liver-ct/raw/main/qa-results/qa-results.csv\n",
            "Resolving github.com (github.com)... 140.82.121.3\n",
            "Connecting to github.com (github.com)|140.82.121.3|:443... connected.\n",
            "HTTP request sent, awaiting response... 302 Found\n",
            "Location: https://raw.githubusercontent.com/bamf-health/aimi-liver-ct/main/qa-results/qa-results.csv [following]\n",
            "--2023-09-27 19:18:17--  https://raw.githubusercontent.com/bamf-health/aimi-liver-ct/main/qa-results/qa-results.csv\n",
            "Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.108.133, 185.199.109.133, 185.199.110.133, ...\n",
            "Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.108.133|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 30208 (30K) [text/plain]\n",
            "Saving to: ‘qa-results.csv’\n",
            "\n",
            "qa-results.csv      100%[===================>]  29.50K  --.-KB/s    in 0s      \n",
            "\n",
            "2023-09-27 19:18:17 (64.1 MB/s) - ‘qa-results.csv’ saved [30208/30208]\n",
            "\n"
          ]
        },
        {
          "data": {
            "text/plain": []
          },
          "execution_count": 4,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "%%shell\n",
        "wget https://github.com/bamf-health/aimi-liver-ct/raw/main/qa-results/qa-results.csv"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "id": "sY_R2n98xWNm"
      },
      "outputs": [],
      "source": [
        "qa_df = pd.read_csv('qa-results.csv')\n",
        "series_uids = qa_df.SeriesInstanceUID.tolist()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "id": "DbqG5yBEEEiy"
      },
      "outputs": [],
      "source": [
        "# python API is the most flexible way to query IDC BigQuery metadata tables\n",
        "from google.cloud import bigquery\n",
        "bq_client = bigquery.Client(my_ProjectID)\n",
        "\n",
        "# enclose series_uids in quotes for use in sql query\n",
        "series_uids = [f\"'{x}'\" for x in series_uids]\n",
        "\n",
        "selection_query =f\"\"\"\n",
        "SELECT\n",
        "  # Organize the files in-place right after downloading\n",
        "  ANY_VALUE(CONCAT(\"cp s3\",REGEXP_SUBSTR(gcs_url, \"(://.*)/\"),\"/* \",collection_id,\"/\",PatientID,\"/\",StudyInstanceUID,\"/\",SeriesInstanceUID)) AS s5cmd_command\n",
        "FROM\n",
        "  `bigquery-public-data.idc_current.dicom_all`\n",
        "WHERE\n",
        "  SeriesInstanceUID IN ({','.join(series_uids)})\n",
        "GROUP BY\n",
        "  SeriesInstanceUID\n",
        "\"\"\"\n",
        "\n",
        "selection_result = bq_client.query(selection_query)\n",
        "selection_df = selection_result.result().to_dataframe()\n",
        "\n",
        "selection_df.to_csv(\"/content/s5cmd_gcp_manifest.txt\", header=False, index=False)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LdyZlcWitX5z"
      },
      "source": [
        "Download manifest with the `s5cmd` tool."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 32,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dyfDhOqLoIK7",
        "outputId": "69c90e3c-05aa-486a-a94e-5ce7c3afee2e"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "dicoms already downloaded\n"
          ]
        },
        {
          "data": {
            "text/plain": []
          },
          "execution_count": 32,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "%%shell\n",
        "# check if dicoms have already been downloaded\n",
        "if test -n \"$(find dcms -name '*.dcm' -print -quit)\"\n",
        "then\n",
        "    echo \"dicoms already downloaded\"\n",
        "else\n",
        "  mkdir -p dcms\n",
        "  cd dcms && s5cmd --no-sign-request --endpoint-url https://storage.googleapis.com run ../s5cmd_gcp_manifest.txt\n",
        "  cd -\n",
        "fi"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "AH0HdxH25FAx"
      },
      "source": [
        "# Run model on TCGA-LIHC\n",
        "You can run the model on the downloaded scans with the below code.\n",
        "\n",
        ">If you want to run the model locally, this code is containerized in the project. Refer to the [readme](https://github.com/bamf-health/aimi-liver-ct/tree/main#running-inference) for instructions on running the container locally."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UJMKUYXt5N-Y"
      },
      "source": [
        "Install `nnunet` python package"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 21,
      "metadata": {
        "id": "zuuKeVfX17Hn"
      },
      "outputs": [],
      "source": [
        "%%capture\n",
        "!pip install nnunet"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 23,
      "metadata": {
        "id": "N3KFef1g7Wva"
      },
      "outputs": [],
      "source": [
        "# setup nnunet paths\n",
        "os.environ[\"nnUNet_raw_data_base\"] =\"/nnunet_data/nnUNet_raw_data_base/\"\n",
        "os.environ[\"nnUNet_preprocessed\"] =\"/nnunet_data/nnUNet_preprocessed/\"\n",
        "os.environ[\"RESULTS_FOLDER\"] =\"/nnunet_data/nnUNet_trained_models/\"\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "12o-4JNS559K"
      },
      "source": [
        "download model weights from zenodo"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 33,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ll5e4zv_6sTi",
        "outputId": "42276700-66b3-469c-be16-e99f7c439a1a"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "model weights already downloaded\n"
          ]
        },
        {
          "data": {
            "text/plain": []
          },
          "execution_count": 33,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "%%shell\n",
        "# check if weights exist before starting a large downloaded\n",
        "if [ ! -f ${RESULTS_FOLDER}nnUNet/3d_fullres/Task773_Liver/nnUNetTrainerV2__nnUNetPlansv2.1/fold_0/model_final_checkpoint.model ]\n",
        "then\n",
        "    mkdir -p ${nnUNet_raw_data_base}\n",
        "    mkdir -p ${nnUNet_preprocessed}\n",
        "    mkdir -p ${RESULTS_FOLDER}nnUNet/\n",
        "    echo \"Downloading model weights\"\n",
        "    wget https://zenodo.org/record/8274976/files/Task773_Liver.zip\n",
        "    unzip Task773_Liver.zip -d ${RESULTS_FOLDER}nnUNet/\n",
        "else\n",
        "    echo \"model weights already downloaded\"\n",
        "fi"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CMPDFav5BAfk"
      },
      "source": [
        "Install binaries for `itkimage2segimage` executable"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 34,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NUjGUqV57ngz",
        "outputId": "1688c6cc-1eae-41c7-97f2-99d825843341"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "--2023-09-27 19:41:15--  https://github.com/QIICR/dcmqi/releases/download/v1.2.5/dcmqi-1.2.5-linux.tar.gz\n",
            "Resolving github.com (github.com)... 140.82.121.3\n",
            "Connecting to github.com (github.com)|140.82.121.3|:443... connected.\n",
            "HTTP request sent, awaiting response... 302 Found\n",
            "Location: https://objects.githubusercontent.com/github-production-release-asset-2e65be/50675718/79d3ad95-9f0c-42a4-a1c5-bf5a63461894?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIAIWNJYAX4CSVEH53A%2F20230927%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Date=20230927T194115Z&X-Amz-Expires=300&X-Amz-Signature=f6b80a4fa33ac7144b4817f815faaab84bdfa11899fc3183bd946c54d2917010&X-Amz-SignedHeaders=host&actor_id=0&key_id=0&repo_id=50675718&response-content-disposition=attachment%3B%20filename%3Ddcmqi-1.2.5-linux.tar.gz&response-content-type=application%2Foctet-stream [following]\n",
            "--2023-09-27 19:41:16--  https://objects.githubusercontent.com/github-production-release-asset-2e65be/50675718/79d3ad95-9f0c-42a4-a1c5-bf5a63461894?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIAIWNJYAX4CSVEH53A%2F20230927%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Date=20230927T194115Z&X-Amz-Expires=300&X-Amz-Signature=f6b80a4fa33ac7144b4817f815faaab84bdfa11899fc3183bd946c54d2917010&X-Amz-SignedHeaders=host&actor_id=0&key_id=0&repo_id=50675718&response-content-disposition=attachment%3B%20filename%3Ddcmqi-1.2.5-linux.tar.gz&response-content-type=application%2Foctet-stream\n",
            "Resolving objects.githubusercontent.com (objects.githubusercontent.com)... 185.199.108.133, 185.199.109.133, 185.199.110.133, ...\n",
            "Connecting to objects.githubusercontent.com (objects.githubusercontent.com)|185.199.108.133|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 21102129 (20M) [application/octet-stream]\n",
            "Saving to: ‘dcmqi-1.2.5-linux.tar.gz’\n",
            "\n",
            "dcmqi-1.2.5-linux.t 100%[===================>]  20.12M  53.3MB/s    in 0.4s    \n",
            "\n",
            "2023-09-27 19:41:16 (53.3 MB/s) - ‘dcmqi-1.2.5-linux.tar.gz’ saved [21102129/21102129]\n",
            "\n",
            "dcmqi-1.2.5-linux/bin/\n",
            "dcmqi-1.2.5-linux/bin/itkimage2segimage\n",
            "dcmqi-1.2.5-linux/bin/tid1500reader\n",
            "dcmqi-1.2.5-linux/bin/tid1500reader.xml\n",
            "dcmqi-1.2.5-linux/bin/itkimage2segimage.xml\n",
            "dcmqi-1.2.5-linux/bin/itkimage2paramap.xml\n",
            "dcmqi-1.2.5-linux/bin/itkimage2paramap\n",
            "dcmqi-1.2.5-linux/bin/segimage2itkimage.xml\n",
            "dcmqi-1.2.5-linux/bin/segimage2itkimage\n",
            "dcmqi-1.2.5-linux/bin/tid1500writer.xml\n",
            "dcmqi-1.2.5-linux/bin/tid1500writer\n",
            "dcmqi-1.2.5-linux/bin/paramap2itkimage\n",
            "dcmqi-1.2.5-linux/bin/paramap2itkimage.xml\n",
            "dcmqi-1.2.5-linux/share/\n",
            "dcmqi-1.2.5-linux/share/doc/\n",
            "dcmqi-1.2.5-linux/share/doc/ITK-4.10/\n",
            "dcmqi-1.2.5-linux/share/doc/ITK-4.10/itksys/\n",
            "dcmqi-1.2.5-linux/share/doc/ITK-4.10/itksys/Copyright.txt\n"
          ]
        },
        {
          "data": {
            "text/plain": []
          },
          "execution_count": 34,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "%%shell\n",
        "# Install binaries for itkimage2segimage package\n",
        "mkdir -p /app\n",
        "PACKAGE_TAR=\"dcmqi-1.2.5-linux.tar.gz\"\n",
        "ITKIMAGE2SEGIMAGE_URL=https://github.com/QIICR/dcmqi/releases/download/v1.2.5/${PACKAGE_TAR}\n",
        "wget ${ITKIMAGE2SEGIMAGE_URL} --no-check-certificate\n",
        "tar -zxvf ${PACKAGE_TAR} -C /app\n",
        "rm ${PACKAGE_TAR}"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nre07laxzfKp"
      },
      "source": [
        "Run inference on scans.\n",
        "\n",
        "First download  `run.py` and `ai-dicom-seg-meta.json` from the git repo"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 22,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "p0hn7XjE6I6j",
        "outputId": "d6fdd697-2e9a-451b-9994-176fe094c000"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "--2023-09-27 19:35:21--  https://github.com/bamf-health/aimi-liver-ct/raw/main/container/app/src/run.py\n",
            "Resolving github.com (github.com)... 140.82.121.4\n",
            "Connecting to github.com (github.com)|140.82.121.4|:443... connected.\n",
            "HTTP request sent, awaiting response... 302 Found\n",
            "Location: https://raw.githubusercontent.com/bamf-health/aimi-liver-ct/main/container/app/src/run.py [following]\n",
            "--2023-09-27 19:35:21--  https://raw.githubusercontent.com/bamf-health/aimi-liver-ct/main/container/app/src/run.py\n",
            "Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.108.133, 185.199.109.133, 185.199.111.133, ...\n",
            "Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.108.133|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 7372 (7.2K) [text/plain]\n",
            "Saving to: ‘run.py’\n",
            "\n",
            "run.py              100%[===================>]   7.20K  --.-KB/s    in 0s      \n",
            "\n",
            "2023-09-27 19:35:22 (74.7 MB/s) - ‘run.py’ saved [7372/7372]\n",
            "\n",
            "--2023-09-27 19:35:22--  https://github.com/bamf-health/aimi-liver-ct/raw/main/container/app/dcm-meta/ai-dicom-seg-meta.json\n",
            "Resolving github.com (github.com)... 140.82.121.4\n",
            "Connecting to github.com (github.com)|140.82.121.4|:443... connected.\n",
            "HTTP request sent, awaiting response... 302 Found\n",
            "Location: https://raw.githubusercontent.com/bamf-health/aimi-liver-ct/main/container/app/dcm-meta/ai-dicom-seg-meta.json [following]\n",
            "--2023-09-27 19:35:22--  https://raw.githubusercontent.com/bamf-health/aimi-liver-ct/main/container/app/dcm-meta/ai-dicom-seg-meta.json\n",
            "Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.108.133, 185.199.109.133, 185.199.110.133, ...\n",
            "Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.108.133|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 1004 [text/plain]\n",
            "Saving to: ‘ai-dicom-seg-meta.json’\n",
            "\n",
            "ai-dicom-seg-meta.j 100%[===================>]    1004  --.-KB/s    in 0s      \n",
            "\n",
            "2023-09-27 19:35:22 (58.2 MB/s) - ‘ai-dicom-seg-meta.json’ saved [1004/1004]\n",
            "\n"
          ]
        },
        {
          "data": {
            "text/plain": []
          },
          "execution_count": 22,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "%%shell\n",
        "wget https://github.com/bamf-health/aimi-liver-ct/raw/main/container/app/src/run.py\n",
        "wget https://github.com/bamf-health/aimi-liver-ct/raw/main/container/app/dcm-meta/ai-dicom-seg-meta.json"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "caUGeHpZLJOX"
      },
      "source": [
        "Then run model on all downloaded dicom series"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 35,
      "metadata": {
        "id": "C7aJrrT9B6sH"
      },
      "outputs": [],
      "source": [
        "from run import main_dicom"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 36,
      "metadata": {
        "id": "00qfwZogIE4C"
      },
      "outputs": [],
      "source": [
        "dcm_dir = Path('dcms')\n",
        "seg_dir = Path('preds')\n",
        "seg_meta = Path('ai-dicom-seg-meta.json')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "LnGIjdcIH-NZ"
      },
      "outputs": [],
      "source": [
        "main_dicom(dcm_dir, seg_dir, seg_meta)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gLoCXJJJKZYO"
      },
      "source": [
        "Download segmentations"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "-i3j3i3mLpmz"
      },
      "outputs": [],
      "source": [
        "%%shell\n",
        "zip -r preds.zip preds"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "_MiV7-vcOFTh"
      },
      "outputs": [],
      "source": [
        "from google.colab import files\n",
        "files.download('preds.zip')"
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "authorship_tag": "ABX9TyNlEsagoeqJ8j0lQC7+/lzL",
      "gpuType": "T4",
      "include_colab_link": true,
      "provenance": [],
      "toc_visible": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python",
      "version": "3.10.12"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}