Alzheimer's_Disease.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "T4"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "source": [
        "import kagglehub\n",
        "\n",
        "# Download latest version\n",
        "path = kagglehub.dataset_download(\"marcopinamonti/alzheimer-mri-4-classes-dataset\")\n",
        "\n",
        "print(\"Path to dataset files:\", path)"
      ],
      "metadata": {
        "id": "OUOWY8No9-2R",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "c33bbfa7-142c-414c-b81d-7f2a6825d437"
      },
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading from https://www.kaggle.com/api/v1/datasets/download/marcopinamonti/alzheimer-mri-4-classes-dataset?dataset_version_number=1...\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 34.1M/34.1M [00:00<00:00, 52.5MB/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Extracting files...\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Path to dataset files: /root/.cache/kagglehub/datasets/marcopinamonti/alzheimer-mri-4-classes-dataset/versions/1\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Import necessary libraries\n",
        "import zipfile\n",
        "import os\n",
        "\n",
        "# Specify the path to the uploaded zip file\n",
        "zip_file_path = \"/content/archive.zip\"  # Update the file name if different\n",
        "\n",
        "# Extract the zip file\n",
        "with zipfile.ZipFile(zip_file_path, 'r') as zip_ref:\n",
        "    zip_ref.extractall(\"/content/alzheimers_dataset\")  # Destination folder\n",
        "\n",
        "# Verify extraction\n",
        "print(\"Dataset extracted to /content/alzheimers_dataset\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "L62x_2t6xNTA",
        "outputId": "4e30737f-7ed1-4212-ab9b-83da699f9287"
      },
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Dataset extracted to /content/alzheimers_dataset\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import os\n",
        "import numpy as np\n",
        "import cv2\n",
        "import matplotlib.pyplot as plt\n",
        "from tensorflow.keras.models import Model\n",
        "from tensorflow.keras.layers import Input, Conv2D, MaxPooling2D, Flatten, Dense, Dropout, concatenate\n",
        "from tensorflow.keras.optimizers import Adam\n",
        "from tensorflow.keras.utils import to_categorical\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import classification_report, confusion_matrix, roc_auc_score"
      ],
      "metadata": {
        "id": "b5gGSyxxyFrw"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "DATASET_DIR = \"/content/alzheimers_dataset/Alzheimer_MRI_4_classes_dataset\"\n",
        "CLASSES = ['MildDemented', 'ModerateDemented', 'NonDemented', 'VeryMildDemented']\n",
        "IMG_SIZE = (224, 224)"
      ],
      "metadata": {
        "id": "i2EGhp-kyi94"
      },
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def load_images_from_directory(dataset_dir, classes, img_size):\n",
        "    images, labels = [], []\n",
        "    for idx, label in enumerate(classes):\n",
        "        class_dir = os.path.join(dataset_dir, label)\n",
        "        for img_name in os.listdir(class_dir):\n",
        "            img_path = os.path.join(class_dir, img_name)\n",
        "            try:\n",
        "                img = cv2.imread(img_path, cv2.IMREAD_GRAYSCALE)  # Load as grayscale\n",
        "                img = cv2.resize(img, img_size)  # Resize to 224x224\n",
        "                images.append(img)\n",
        "                labels.append(idx)  # Numeric label\n",
        "            except Exception as e:\n",
        "                print(f\"Error loading image {img_path}: {e}\")\n",
        "    return np.array(images), np.array(labels)\n"
      ],
      "metadata": {
        "id": "fYVvrVJ0yr9S"
      },
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X, y = load_images_from_directory(DATASET_DIR, CLASSES, IMG_SIZE)\n",
        "\n",
        "# Normalize pixel values\n",
        "X = X / 255.0  # Scale to [0, 1]\n",
        "\n",
        "# Reshape for CNN input\n",
        "X = X.reshape(-1, IMG_SIZE[0], IMG_SIZE[1], 1)  # Add channel dimension\n",
        "\n",
        "# One-hot encode labels\n",
        "y = to_categorical(y, num_classes=len(CLASSES))\n",
        "\n",
        "# Split dataset\n",
        "X_train, X_temp, y_train, y_temp = train_test_split(X, y, test_size=0.3, random_state=42)\n",
        "X_val, X_test, y_val, y_test = train_test_split(X_temp, y_temp, test_size=0.5, random_state=42)\n",
        "\n",
        "print(f\"Training samples: {X_train.shape[0]}\")\n",
        "print(f\"Validation samples: {X_val.shape[0]}\")\n",
        "print(f\"Test samples: {X_test.shape[0]}\")\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6A2BVTuNyt1g",
        "outputId": "e8a7f781-adda-4a8a-864b-d8e3eb7b4636"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Training samples: 4480\n",
            "Validation samples: 960\n",
            "Test samples: 960\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def create_cnn(input_shape):\n",
        "    input_layer = Input(shape=input_shape)\n",
        "\n",
        "    # Convolutional Layers\n",
        "    x = Conv2D(32, (3, 3), activation='relu', padding='same')(input_layer)\n",
        "    x = MaxPooling2D((2, 2))(x)\n",
        "    x = Conv2D(64, (3, 3), activation='relu', padding='same')(x)\n",
        "    x = MaxPooling2D((2, 2))(x)\n",
        "\n",
        "    # Flatten and Dense Layers\n",
        "    x = Flatten()(x)\n",
        "    x = Dense(128, activation='relu')(x)\n",
        "    x = Dropout(0.3)(x)\n",
        "    return input_layer, x"
      ],
      "metadata": {
        "id": "JK3tHGbGyvHu"
      },
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def build_ann(input_shape, num_classes):\n",
        "    cnn_input, cnn_output = create_cnn(input_shape)\n",
        "\n",
        "    # Output layer\n",
        "    output = Dense(num_classes, activation='softmax')(cnn_output)\n",
        "\n",
        "    # Model\n",
        "    model = Model(inputs=cnn_input, outputs=output)\n",
        "    return model\n"
      ],
      "metadata": {
        "id": "ViGBw8gJzP_D"
      },
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Define input shape\n",
        "input_shape = (IMG_SIZE[0], IMG_SIZE[1], 1)  # 224x224 grayscale images\n",
        "\n",
        "# Build the model\n",
        "model = build_ann(input_shape=input_shape, num_classes=len(CLASSES))\n",
        "\n",
        "# Compile\n",
        "model.compile(optimizer=Adam(learning_rate=0.001),\n",
        "              loss='categorical_crossentropy',\n",
        "              metrics=['accuracy'])\n",
        "\n",
        "model.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 417
        },
        "id": "b9COP8RxzRnf",
        "outputId": "94262f56-3371-43b6-c4d9-83fb862d1625"
      },
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1mModel: \"functional\"\u001b[0m\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"functional\"</span>\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\n",
              "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                        \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape               \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m        Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩\n",
              "│ input_layer (\u001b[38;5;33mInputLayer\u001b[0m)             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m1\u001b[0m)         │               \u001b[38;5;34m0\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ conv2d (\u001b[38;5;33mConv2D\u001b[0m)                      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m32\u001b[0m)        │             \u001b[38;5;34m320\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ max_pooling2d (\u001b[38;5;33mMaxPooling2D\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m32\u001b[0m)        │               \u001b[38;5;34m0\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ conv2d_1 (\u001b[38;5;33mConv2D\u001b[0m)                    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m64\u001b[0m)        │          \u001b[38;5;34m18,496\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ max_pooling2d_1 (\u001b[38;5;33mMaxPooling2D\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m64\u001b[0m)          │               \u001b[38;5;34m0\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ flatten (\u001b[38;5;33mFlatten\u001b[0m)                    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m200704\u001b[0m)              │               \u001b[38;5;34m0\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense (\u001b[38;5;33mDense\u001b[0m)                        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m)                 │      \u001b[38;5;34m25,690,240\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dropout (\u001b[38;5;33mDropout\u001b[0m)                    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m)                 │               \u001b[38;5;34m0\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense_1 (\u001b[38;5;33mDense\u001b[0m)                      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m)                   │             \u001b[38;5;34m516\u001b[0m │\n",
              "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\n",
              "┃<span style=\"font-weight: bold\"> Layer (type)                         </span>┃<span style=\"font-weight: bold\"> Output Shape                </span>┃<span style=\"font-weight: bold\">         Param # </span>┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩\n",
              "│ input_layer (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">InputLayer</span>)             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">224</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">224</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>)         │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ conv2d (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)                      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">224</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">224</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)        │             <span style=\"color: #00af00; text-decoration-color: #00af00\">320</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ max_pooling2d (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)         │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)        │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ conv2d_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)                    │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)        │          <span style=\"color: #00af00; text-decoration-color: #00af00\">18,496</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ max_pooling2d_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">56</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">56</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)          │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ flatten (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Flatten</span>)                    │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">200704</span>)              │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)                 │      <span style=\"color: #00af00; text-decoration-color: #00af00\">25,690,240</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dropout (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)                    │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)                 │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">4</span>)                   │             <span style=\"color: #00af00; text-decoration-color: #00af00\">516</span> │\n",
              "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m25,709,572\u001b[0m (98.07 MB)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">25,709,572</span> (98.07 MB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m25,709,572\u001b[0m (98.07 MB)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">25,709,572</span> (98.07 MB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "history = model.fit(\n",
        "    X_train, y_train,\n",
        "    validation_data=(X_val, y_val),\n",
        "    epochs=50,\n",
        "    batch_size=32,\n",
        "    verbose=1\n",
        ")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "JG1L8fTuzTiM",
        "outputId": "df927f6f-fcc1-45d4-be98-462af848737f"
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 50ms/step - accuracy: 0.4748 - loss: 1.6532 - val_accuracy: 0.5365 - val_loss: 0.9418\n",
            "Epoch 2/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 38ms/step - accuracy: 0.5208 - loss: 0.9788 - val_accuracy: 0.5979 - val_loss: 0.8787\n",
            "Epoch 3/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 39ms/step - accuracy: 0.6070 - loss: 0.8462 - val_accuracy: 0.7302 - val_loss: 0.6738\n",
            "Epoch 4/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 39ms/step - accuracy: 0.7485 - loss: 0.6185 - val_accuracy: 0.8323 - val_loss: 0.4389\n",
            "Epoch 5/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 39ms/step - accuracy: 0.8705 - loss: 0.3467 - val_accuracy: 0.8906 - val_loss: 0.2583\n",
            "Epoch 6/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 39ms/step - accuracy: 0.9459 - loss: 0.1599 - val_accuracy: 0.9458 - val_loss: 0.1615\n",
            "Epoch 7/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 38ms/step - accuracy: 0.9695 - loss: 0.0943 - val_accuracy: 0.9615 - val_loss: 0.1102\n",
            "Epoch 8/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 39ms/step - accuracy: 0.9809 - loss: 0.0601 - val_accuracy: 0.9521 - val_loss: 0.1226\n",
            "Epoch 9/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 39ms/step - accuracy: 0.9885 - loss: 0.0415 - val_accuracy: 0.9719 - val_loss: 0.0828\n",
            "Epoch 10/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 39ms/step - accuracy: 0.9928 - loss: 0.0244 - val_accuracy: 0.9760 - val_loss: 0.0836\n",
            "Epoch 11/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 39ms/step - accuracy: 0.9911 - loss: 0.0281 - val_accuracy: 0.9698 - val_loss: 0.1012\n",
            "Epoch 12/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 39ms/step - accuracy: 0.9958 - loss: 0.0163 - val_accuracy: 0.9771 - val_loss: 0.0580\n",
            "Epoch 13/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 40ms/step - accuracy: 0.9932 - loss: 0.0202 - val_accuracy: 0.9740 - val_loss: 0.0759\n",
            "Epoch 14/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 39ms/step - accuracy: 0.9922 - loss: 0.0249 - val_accuracy: 0.9740 - val_loss: 0.0768\n",
            "Epoch 15/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 42ms/step - accuracy: 0.9932 - loss: 0.0200 - val_accuracy: 0.9771 - val_loss: 0.0655\n",
            "Epoch 16/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 39ms/step - accuracy: 0.9936 - loss: 0.0179 - val_accuracy: 0.9750 - val_loss: 0.0894\n",
            "Epoch 17/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 39ms/step - accuracy: 0.9963 - loss: 0.0100 - val_accuracy: 0.9802 - val_loss: 0.0657\n",
            "Epoch 18/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 39ms/step - accuracy: 0.9990 - loss: 0.0068 - val_accuracy: 0.9792 - val_loss: 0.0729\n",
            "Epoch 19/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 40ms/step - accuracy: 0.9946 - loss: 0.0176 - val_accuracy: 0.9802 - val_loss: 0.0617\n",
            "Epoch 20/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 39ms/step - accuracy: 0.9934 - loss: 0.0191 - val_accuracy: 0.9792 - val_loss: 0.0766\n",
            "Epoch 21/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 40ms/step - accuracy: 0.9942 - loss: 0.0174 - val_accuracy: 0.9750 - val_loss: 0.0942\n",
            "Epoch 22/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 39ms/step - accuracy: 0.9947 - loss: 0.0170 - val_accuracy: 0.9688 - val_loss: 0.1429\n",
            "Epoch 23/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 40ms/step - accuracy: 0.9962 - loss: 0.0191 - val_accuracy: 0.9771 - val_loss: 0.0625\n",
            "Epoch 24/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 40ms/step - accuracy: 0.9964 - loss: 0.0143 - val_accuracy: 0.9802 - val_loss: 0.0945\n",
            "Epoch 25/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 40ms/step - accuracy: 0.9942 - loss: 0.0195 - val_accuracy: 0.9750 - val_loss: 0.1001\n",
            "Epoch 26/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 40ms/step - accuracy: 0.9939 - loss: 0.0162 - val_accuracy: 0.9729 - val_loss: 0.1479\n",
            "Epoch 27/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 40ms/step - accuracy: 0.9967 - loss: 0.0091 - val_accuracy: 0.9740 - val_loss: 0.0972\n",
            "Epoch 28/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 40ms/step - accuracy: 0.9936 - loss: 0.0179 - val_accuracy: 0.9823 - val_loss: 0.0863\n",
            "Epoch 29/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 40ms/step - accuracy: 0.9985 - loss: 0.0056 - val_accuracy: 0.9802 - val_loss: 0.0941\n",
            "Epoch 30/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 40ms/step - accuracy: 0.9968 - loss: 0.0122 - val_accuracy: 0.9771 - val_loss: 0.1026\n",
            "Epoch 31/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 41ms/step - accuracy: 0.9982 - loss: 0.0066 - val_accuracy: 0.9708 - val_loss: 0.1259\n",
            "Epoch 32/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 40ms/step - accuracy: 0.9956 - loss: 0.0150 - val_accuracy: 0.9802 - val_loss: 0.0931\n",
            "Epoch 33/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 40ms/step - accuracy: 0.9971 - loss: 0.0087 - val_accuracy: 0.9812 - val_loss: 0.0975\n",
            "Epoch 34/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 40ms/step - accuracy: 0.9982 - loss: 0.0059 - val_accuracy: 0.9729 - val_loss: 0.1298\n",
            "Epoch 35/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 40ms/step - accuracy: 0.9961 - loss: 0.0083 - val_accuracy: 0.9698 - val_loss: 0.1457\n",
            "Epoch 36/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 41ms/step - accuracy: 0.9971 - loss: 0.0092 - val_accuracy: 0.9760 - val_loss: 0.1017\n",
            "Epoch 37/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 43ms/step - accuracy: 0.9967 - loss: 0.0097 - val_accuracy: 0.9688 - val_loss: 0.1415\n",
            "Epoch 38/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 40ms/step - accuracy: 0.9946 - loss: 0.0143 - val_accuracy: 0.9729 - val_loss: 0.1104\n",
            "Epoch 39/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 40ms/step - accuracy: 0.9939 - loss: 0.0109 - val_accuracy: 0.9729 - val_loss: 0.1222\n",
            "Epoch 40/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 40ms/step - accuracy: 0.9979 - loss: 0.0069 - val_accuracy: 0.9760 - val_loss: 0.0802\n",
            "Epoch 41/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 40ms/step - accuracy: 0.9980 - loss: 0.0055 - val_accuracy: 0.9802 - val_loss: 0.1128\n",
            "Epoch 42/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 41ms/step - accuracy: 0.9966 - loss: 0.0111 - val_accuracy: 0.9698 - val_loss: 0.1401\n",
            "Epoch 43/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 40ms/step - accuracy: 0.9984 - loss: 0.0047 - val_accuracy: 0.9812 - val_loss: 0.1014\n",
            "Epoch 44/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 40ms/step - accuracy: 0.9985 - loss: 0.0051 - val_accuracy: 0.9740 - val_loss: 0.1262\n",
            "Epoch 45/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 41ms/step - accuracy: 0.9963 - loss: 0.0093 - val_accuracy: 0.9812 - val_loss: 0.0847\n",
            "Epoch 46/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 40ms/step - accuracy: 0.9976 - loss: 0.0056 - val_accuracy: 0.9781 - val_loss: 0.1049\n",
            "Epoch 47/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 40ms/step - accuracy: 0.9972 - loss: 0.0088 - val_accuracy: 0.9844 - val_loss: 0.0870\n",
            "Epoch 48/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 40ms/step - accuracy: 0.9925 - loss: 0.0220 - val_accuracy: 0.9781 - val_loss: 0.0983\n",
            "Epoch 49/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 40ms/step - accuracy: 0.9991 - loss: 0.0040 - val_accuracy: 0.9823 - val_loss: 0.0925\n",
            "Epoch 50/50\n",
            "\u001b[1m140/140\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 40ms/step - accuracy: 0.9998 - loss: 0.0036 - val_accuracy: 0.9719 - val_loss: 0.1633\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Evaluate on test data\n",
        "test_loss, test_accuracy = model.evaluate(X_test, y_test, verbose=1)\n",
        "print(f\"Test Accuracy: {test_accuracy * 100:.2f}%\")\n",
        "\n",
        "# Classification report\n",
        "y_pred = model.predict(X_test)\n",
        "y_pred_classes = np.argmax(y_pred, axis=1)\n",
        "y_true = np.argmax(y_test, axis=1)\n",
        "\n",
        "print(\"Classification Report:\")\n",
        "print(classification_report(y_true, y_pred_classes, target_names=CLASSES))\n",
        "\n",
        "# Confusion matrix\n",
        "conf_matrix = confusion_matrix(y_true, y_pred_classes)\n",
        "print(\"Confusion Matrix:\")\n",
        "print(conf_matrix)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "oT030iEDzVfn",
        "outputId": "64cc4d02-5a7c-4728-944f-39aefca6d3b3"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m30/30\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 12ms/step - accuracy: 0.9710 - loss: 0.1652\n",
            "Test Accuracy: 97.29%\n",
            "\u001b[1m30/30\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step\n",
            "Classification Report:\n",
            "                  precision    recall  f1-score   support\n",
            "\n",
            "    MildDemented       1.00      0.97      0.98       156\n",
            "ModerateDemented       1.00      1.00      1.00         6\n",
            "     NonDemented       0.97      0.99      0.98       479\n",
            "VeryMildDemented       0.97      0.95      0.96       319\n",
            "\n",
            "        accuracy                           0.97       960\n",
            "       macro avg       0.98      0.98      0.98       960\n",
            "    weighted avg       0.97      0.97      0.97       960\n",
            "\n",
            "Confusion Matrix:\n",
            "[[151   0   1   4]\n",
            " [  0   6   0   0]\n",
            " [  0   0 474   5]\n",
            " [  0   0  16 303]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Plot training history\n",
        "plt.figure(figsize=(12, 5))\n",
        "\n",
        "# Accuracy\n",
        "plt.subplot(1, 2, 1)\n",
        "plt.plot(history.history['accuracy'], label='Training Accuracy')\n",
        "plt.plot(history.history['val_accuracy'], label='Validation Accuracy')\n",
        "plt.title('Accuracy Over Epochs')\n",
        "plt.xlabel('Epochs')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.legend()\n",
        "\n",
        "# Loss\n",
        "plt.subplot(1, 2, 2)\n",
        "plt.plot(history.history['loss'], label='Training Loss')\n",
        "plt.plot(history.history['val_loss'], label='Validation Loss')\n",
        "plt.title('Loss Over Epochs')\n",
        "plt.xlabel('Epochs')\n",
        "plt.ylabel('Loss')\n",
        "plt.legend()\n",
        "\n",
        "plt.tight_layout()\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 476
        },
        "id": "AxtAoGz704_D",
        "outputId": "945a1f21-03f4-4c26-8d32-2532e0fda2f5"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x500 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install tf-keras-vis\n",
        "from tensorflow.keras import backend as K\n",
        "from tf_keras_vis.utils.scores import CategoricalScore\n",
        "from tf_keras_vis.gradcam import Gradcam\n",
        "\n",
        "# Grad-CAM visualization\n",
        "def gradcam_visualization(image, model, class_index):\n",
        "    gradcam = Gradcam(model=model)\n",
        "    score = CategoricalScore(class_index)  # Class index for Grad-CAM\n",
        "    cam = gradcam(score, image[np.newaxis])\n",
        "    plt.imshow(cam[0], cmap='jet')\n",
        "    plt.title(f'Grad-CAM for class {class_index}')\n",
        "    plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qOf5i9Mt0-Gs",
        "outputId": "3128ea76-ea6d-454e-8bf1-6a7e466ea941"
      },
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Collecting tf-keras-vis\n",
            "  Downloading tf_keras_vis-0.8.7-py3-none-any.whl.metadata (10 kB)\n",
            "Requirement already satisfied: scipy in /usr/local/lib/python3.10/dist-packages (from tf-keras-vis) (1.13.1)\n",
            "Requirement already satisfied: pillow in /usr/local/lib/python3.10/dist-packages (from tf-keras-vis) (11.0.0)\n",
            "Requirement already satisfied: deprecated in /usr/local/lib/python3.10/dist-packages (from tf-keras-vis) (1.2.15)\n",
            "Requirement already satisfied: imageio in /usr/local/lib/python3.10/dist-packages (from tf-keras-vis) (2.36.0)\n",
            "Requirement already satisfied: packaging in /usr/local/lib/python3.10/dist-packages (from tf-keras-vis) (24.2)\n",
            "Requirement already satisfied: wrapt<2,>=1.10 in /usr/local/lib/python3.10/dist-packages (from deprecated->tf-keras-vis) (1.16.0)\n",
            "Requirement already satisfied: numpy in /usr/local/lib/python3.10/dist-packages (from imageio->tf-keras-vis) (1.26.4)\n",
            "Downloading tf_keras_vis-0.8.7-py3-none-any.whl (52 kB)\n",
            "\u001b[?25l   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m0.0/52.5 kB\u001b[0m \u001b[31m?\u001b[0m eta \u001b[36m-:--:--\u001b[0m\r\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m52.5/52.5 kB\u001b[0m \u001b[31m2.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hInstalling collected packages: tf-keras-vis\n",
            "Successfully installed tf-keras-vis-0.8.7\n"
          ]
        }
      ]
    }
  ]
}