Bone_Age_Classification_Task.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyM1rulBVAv4fYZdJr62Dn9z",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/Armin-Abdollahi/Bone-Age-Estimation/blob/main/Bone_Age_Classification_Task.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Classification task for bone age estimation"
      ],
      "metadata": {
        "id": "t5547yfNYDMY"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "metadata": {
        "id": "IpoMwrcEYN5A"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "9hTGgbSJX78N"
      },
      "outputs": [],
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "import tensorflow as tf\n",
        "import matplotlib.pyplot as plt\n",
        "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import Conv2D, MaxPooling2D, Flatten, Dense, Dropout\n",
        "from tensorflow.keras.optimizers import Adam\n",
        "from sklearn.naive_bayes import MultinomialNB\n",
        "from sklearn.metrics import confusion_matrix, classification_report\n",
        "from sklearn.preprocessing import LabelEncoder"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Load the labels\n",
        "labels_path = '/content/drive/MyDrive/Digital_Hand_Atlas_Dataset/Total_Images/Bordered_XRay_Images/Classification_Task_Labels.csv'\n",
        "labels = pd.read_csv(labels_path)"
      ],
      "metadata": {
        "id": "vn-lsRK9YPx1"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "labels['Case ID'] = labels['Case ID'].apply(lambda x: str(x)+'.jpg')\n",
        "\n",
        "labels.head()"
      ],
      "metadata": {
        "id": "higcSYB_YRKr"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Map bone age to age groups (Group A, B, C)\n",
        "def map_to_age_group(bone_age):\n",
        "    if bone_age <= 5:\n",
        "        return 'Group A'\n",
        "    elif 6 <= bone_age <= 12:\n",
        "        return 'Group B'\n",
        "    else:\n",
        "        return 'Group C'\n",
        "\n",
        "labels['Age Group'] = labels['Bone Age'].apply(map_to_age_group)"
      ],
      "metadata": {
        "id": "3PCnu-2mYSmP"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Ensure the 'Case ID' column is of type string\n",
        "labels['Case ID'] = labels['Case ID'].astype(str)"
      ],
      "metadata": {
        "id": "mcpcNiNRYUFz"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Create the image data generator\n",
        "datagen = ImageDataGenerator(rescale=1./255, validation_split=0.2)"
      ],
      "metadata": {
        "id": "X4sn2lTdYVYc"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Create the train and validation generators\n",
        "train_generator = datagen.flow_from_dataframe(\n",
        "    dataframe=labels,\n",
        "    directory='/content/drive/MyDrive/Digital_Hand_Atlas_Dataset/Total_Images/Bordered_XRay_Images',\n",
        "    x_col='Case ID',\n",
        "    y_col='Age Group',  # Use the new 'Age Group' column\n",
        "    target_size=(256, 256),\n",
        "    color_mode='grayscale',\n",
        "    class_mode='categorical',  # Use categorical mode for classification\n",
        "    batch_size=32,\n",
        "    subset='training'\n",
        ")\n",
        "\n",
        "validation_generator = datagen.flow_from_dataframe(\n",
        "    dataframe=labels,\n",
        "    directory='/content/drive/MyDrive/Digital_Hand_Atlas_Dataset/Total_Images/Bordered_XRay_Images',\n",
        "    x_col='Case ID',\n",
        "    y_col='Age Group',\n",
        "    target_size=(256, 256),\n",
        "    color_mode='grayscale',\n",
        "    class_mode='categorical',\n",
        "    batch_size=32,\n",
        "    subset='validation'\n",
        ")"
      ],
      "metadata": {
        "id": "8S2q2PeSYWvl"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Build the modified CNN model\n",
        "model = Sequential([\n",
        "    Conv2D(32, (3,3), activation='relu', input_shape=(256, 256, 1)),\n",
        "    MaxPooling2D(2, 2),\n",
        "    Conv2D(64, (3,3), activation='relu'),\n",
        "    MaxPooling2D(2,2),\n",
        "    Conv2D(128, (3,3), activation='relu'),\n",
        "    MaxPooling2D(2,2),\n",
        "    Flatten(),\n",
        "    Dense(512, activation='relu'),\n",
        "    Dropout(0.5),\n",
        "    Dense(3, activation='softmax')  # Three output neurons for age groups\n",
        "])"
      ],
      "metadata": {
        "id": "uVatypfFYYPh"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Compile the model\n",
        "model.compile(optimizer=Adam(), loss='mean_squared_error', metrics=['mae', 'accuracy'])"
      ],
      "metadata": {
        "id": "xMU_Mv5eYZBv"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Train the model\n",
        "history = model.fit(\n",
        "    train_generator,\n",
        "    steps_per_epoch=train_generator.samples // 32,\n",
        "    validation_data=validation_generator,\n",
        "    validation_steps=validation_generator.samples // 32,\n",
        "    epochs=10\n",
        ")"
      ],
      "metadata": {
        "id": "n1V_7vJDYgVq"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Plot training & validation MAE values\n",
        "plt.figure(figsize=(18, 4))\n",
        "\n",
        "plt.subplot(1, 3, 1)\n",
        "plt.plot(history.history['mae'])\n",
        "plt.plot(history.history['val_mae'])\n",
        "plt.title('Model MAE')\n",
        "plt.ylabel('MAE')\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['Train', 'Validation'], loc='upper left')\n",
        "\n",
        "# Plot training & validation loss values\n",
        "plt.subplot(1, 3, 2)\n",
        "plt.plot(history.history['loss'])\n",
        "plt.plot(history.history['val_loss'])\n",
        "plt.title('Model Loss')\n",
        "plt.ylabel('Loss')\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['Train', 'Validation'], loc='upper left')\n",
        "\n",
        "# Plot training & validation accuracy values\n",
        "plt.subplot(1, 3, 3)\n",
        "plt.plot(history.history['accuracy'])\n",
        "plt.plot(history.history['val_accuracy'])\n",
        "plt.title('Model Accuracy')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['Train', 'Validation'], loc='upper left')\n",
        "\n",
        "plt.tight_layout()\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "a83mtm3TYrxH"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Train the model\n",
        "history = model.fit(\n",
        "    train_generator,\n",
        "    steps_per_epoch=train_generator.samples // 32,\n",
        "    validation_data=validation_generator,\n",
        "    validation_steps=validation_generator.samples // 32,\n",
        "    epochs=18\n",
        ")"
      ],
      "metadata": {
        "id": "9J1loSG3YuMR"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Plot training & validation MAE values\n",
        "plt.figure(figsize=(18, 4))\n",
        "\n",
        "plt.subplot(1, 3, 1)\n",
        "plt.plot(history.history['mae'])\n",
        "plt.plot(history.history['val_mae'])\n",
        "plt.title('Model MAE')\n",
        "plt.ylabel('MAE')\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['Train', 'Validation'], loc='upper left')\n",
        "\n",
        "# Plot training & validation loss values\n",
        "plt.subplot(1, 3, 2)\n",
        "plt.plot(history.history['loss'])\n",
        "plt.plot(history.history['val_loss'])\n",
        "plt.title('Model Loss')\n",
        "plt.ylabel('Loss')\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['Train', 'Validation'], loc='upper left')\n",
        "\n",
        "# Plot training & validation accuracy values\n",
        "plt.subplot(1, 3, 3)\n",
        "plt.plot(history.history['accuracy'])\n",
        "plt.plot(history.history['val_accuracy'])\n",
        "plt.title('Model Accuracy')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['Train', 'Validation'], loc='upper left')\n",
        "\n",
        "plt.tight_layout()\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "U9CzgoXmYv1C"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Save the model\n",
        "model.save('/content/drive/MyDrive/bone_age_classification_model.h5')"
      ],
      "metadata": {
        "id": "RWe2rQ69ZeBa"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Confusion Matrix"
      ],
      "metadata": {
        "id": "neNKCECpZG2l"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "from sklearn.naive_bayes import MultinomialNB\n",
        "from sklearn.metrics import confusion_matrix, classification_report\n",
        "from sklearn.preprocessing import LabelEncoder"
      ],
      "metadata": {
        "id": "Rdjn8suKZGEo"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Load the labels\n",
        "labels_path = '/content/drive/MyDrive/Digital_Hand_Atlas_Dataset/Total_Images/Bordered_XRay_Images/Classification_Task_Labels.csv'\n",
        "labels = pd.read_csv(labels_path)"
      ],
      "metadata": {
        "id": "Tbuj8iqAZLmY"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Map bone age to age groups (Group A, B, C)\n",
        "def map_to_age_group(bone_age):\n",
        "    if bone_age <= 5:\n",
        "        return 'Group A'\n",
        "    elif 6 <= bone_age <= 12:\n",
        "        return 'Group B'\n",
        "    else:\n",
        "        return 'Group C'\n",
        "\n",
        "labels['Age Group'] = labels['Bone Age'].apply(map_to_age_group)"
      ],
      "metadata": {
        "id": "L95gek8eZND0"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Encode age groups\n",
        "le = LabelEncoder()\n",
        "labels['Age Group Encoded'] = le.fit_transform(labels['Age Group'])"
      ],
      "metadata": {
        "id": "4Nc6L_ewZOrB"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Split data into train and validation sets\n",
        "train_size = int(0.8 * len(labels))\n",
        "train_data = labels.iloc[:train_size]\n",
        "validation_data = labels.iloc[train_size:]"
      ],
      "metadata": {
        "id": "4ekmP_7sZQL5"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Create features (X) and target labels (y)\n",
        "X_train = train_data['Case ID']\n",
        "y_train = train_data['Age Group Encoded']\n",
        "X_val = validation_data['Case ID']\n",
        "y_val = validation_data['Age Group Encoded']"
      ],
      "metadata": {
        "id": "ynQwdgsoZRfD"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Initialize and train the Naive Bayes classifier\n",
        "nb_classifier = MultinomialNB()\n",
        "nb_classifier.fit(X_train.to_frame(), y_train)"
      ],
      "metadata": {
        "id": "MHYv2wGHZS8F"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Predictions on validation set\n",
        "y_pred = nb_classifier.predict(X_val.to_frame())"
      ],
      "metadata": {
        "id": "4QHiAbKAZUe6"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Confusion matrix\n",
        "conf_matrix = confusion_matrix(y_val, y_pred)\n",
        "print(\"Confusion Matrix:\\n\", conf_matrix)"
      ],
      "metadata": {
        "id": "WL9kOEm7ZVx7"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Classification report\n",
        "class_report = classification_report(y_val, y_pred)\n",
        "print(\"Classification Report:\\n\", class_report)"
      ],
      "metadata": {
        "id": "O4rMNXfCZXOJ"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}