added theoretical introduction to WSI DICOM format

Author: DanielaSchacherer

notebooks/pathomics/DICOM_WSI_overview.png

@@ -23,7 +23,7 @@
         "Besides radiology, slide microscopy is the second major imaging modality in the IDC. Slide microscopy images show **thin sections of tissue samples** (e.g., from a resected tumor) at microscopic resolution. They provide a unique glimpse into cellular architecture and function that is essential for diagnosing complex diseases like cancer. Computerized analysis makes the assessment of slide microscopy images more reproducible and less time consuming and it enables the extraction of novel digital biomarkers from tissue images.\n",
         "\n",
         "This tutorial provides a **15-minute** introduction to the way slide microscopy data is organized within the IDC and demonstrates how to **examine available data and build a data cohort** for further analysis.\n",
-        "It also provides an example of the **application of deep learning (DL)** to computational pathology analysis. In particular, a pre-trained model made available in the context of the publication [The NCI Imaging Data Commons as a platform for reproducible research in computational pathology](https://doi.org/10.1016/j.cmpb.2023.107839) is used to **classify lung tissue regions** into either non-neoplastic (normal), adenocarcinoma (LUAD), or squamous cell carcinoma (LSCC) tissue. A quite similar notebook was part of a [tutorial given at ISBI 2024](https://github.com/ImagingDataCommons/IDC-Tutorials/wiki/ISBI2024) and can be found [here](https://github.com/ImagingDataCommons/IDC-Tutorials/blob/master/notebooks/labs/idc_isbi2024.ipynb).\n",
+        "It also provides an example of the **application of deep learning (DL)** to computational pathology analysis. In particular, a pre-trained model made available in the context of the publication [The NCI Imaging Data Commons as a platform for reproducible research in computational pathology](https://doi.org/10.1016/j.cmpb.2023.107839) is used to **classify lung tissue regions** into either non-neoplastic (normal), adenocarcinoma (LUAD), or squamous cell carcinoma (LSCC) tissue. A quite similar notebook with an acompanying slide deck was part of a tutorial given at ISBI 2024 and can be found [here](https://github.com/ImagingDataCommons/IDC-Tutorials/wiki/ISBI2024). \n",
         "\n",
         "For a more comprehensive tutorial including training of a tissue classification model on IDC-hosted slide microscopy data, see the Github repository [idc-comppath-reproducibility](https://github.com/ImagingDataCommons/idc-comppath-reproducibility).  \n",
         "  \n",
@@ -183,28 +183,23 @@
     },
     {
       "cell_type": "markdown",
-      "metadata": {
-        "id": "Px14ELwJXj9q"
-      },
+      "metadata": {},
       "source": [
-        "# Determinism\n",
-        "Even in smaller experiments it is useful to ensure reproducibility of operations. An important part to this is setting random seeds and enabling TensorFlow operations to be deterministic.\n",
-        "\n",
-        "To learn more about reproducible computational pathology workflows using the IDC, you can read our publication [The NCI Imaging Data Commons as a platform for reproducible research in computational pathology](https://doi.org/10.1016/j.cmpb.2023.107839)."
+        "# Part 1: Introduction to whole slide imaging with DICOM"
       ]
     },
     {
-      "cell_type": "code",
-      "execution_count": 5,
-      "metadata": {
-        "id": "wlRZ3mjuXkIB"
-      },
-      "outputs": [],
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "The IDC uses the [DICOM standard](https://www.dicomstandard.org/) for data representation. As shown in the figure below, a WSI pyramid corresponds to a `Series` of DICOM (SOP) instances, that each represent the slide at a different resolution. Each DICOM instance is stored as a separate DICOM file. Multiple DICOM `Series` can be organized in a DICOM `Study`, and similarly multiple DICOM `Studies` can belong to one `Patient`. "
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
       "source": [
-        "random.seed(0)\n",
-        "np.random.seed(0)\n",
-        "tf.random.set_seed(0)\n",
-        "tf.config.experimental.enable_op_determinism()"
+        "<img src=\"DICOM_WSI_overview.png\" alt=\"Description\" width=\"1000\"/>"
       ]
     },
     {
@@ -213,7 +208,7 @@
         "id": "OaLcUqjY39FU"
       },
       "source": [
-        "# Part 1: How to search whole-slide data on IDC"
+        "# Part 2: How to search whole slide data on IDC"
       ]
     },
     {
@@ -224,7 +219,7 @@
       "source": [
         "For most computational pathology experiments, the first step is to select a cohort of WSIs by filtering for the desired metadata attributes.\n",
         "\n",
-        "The IDC uses the [DICOM standard](https://www.dicomstandard.org/) for data representation. Here, a WSI corresponds to a `Series` of DICOM image objects, each representing the slide at a different resolution. Each DICOM object is stored as a separate DICOM file. Cohort selection is done easiest by executing SQL-like statements using the Python package `idc-index` against different index tables, which list all available DICOM `Series` or files (rows) with the corresponding available metadata atrributes (columns). For getting started with the idc-index, we refer to this [introductory notebook](https://github.com/ImagingDataCommons/IDC-Tutorials/blob/master/notebooks/getting_started/part2_searching_basics.ipynb).\n",
+        "Cohort selection is done easiest by executing SQL-like statements using the Python package `idc-index` against different index tables, which list all available DICOM `Series` or files (rows) with the corresponding available metadata atrributes (columns). For getting started with the idc-index, we refer to this [introductory notebook](https://github.com/ImagingDataCommons/IDC-Tutorials/blob/master/notebooks/getting_started/part2_searching_basics.ipynb).\n",
         "\n",
         "If you can't find your attribute of interest in the idc-index, check out [this notebook](https://github.com/ImagingDataCommons/IDC-Tutorials/blob/master/notebooks/getting_started/part3_exploring_cohorts.ipynb) to learn how to search **all** of the metadata accompanying IDC using Google BigQuery.\n",
         "\n",