shivam3052 / assignment_cnn_skin_cancer

Skin Cancer Detection using CNN

Overview

• This project aims to develop a Convolutional Neural Network (CNN) model for the accurate detection of skin cancer.
• The dataset used in this project contains images of various skin cancer types.
• The goal is to create a model that can assist dermatologists in diagnosing skin cancer at an early stage.

Data Collection

• The dataset is accessed by mounting Google Drive using Google Colab.
• It contains images of various skin cancer types, with separate subdirectories for each class.

Data Preprocessing

Loading Data

• The data is loaded using TensorFlow's tf.keras.utils.image_dataset_from_directory function.
• The dataset is split into training and validation datasets, with an 80% - 20% ratio.

Data Augmentation

• Data augmentation is applied to address class imbalance and improve model generalization.
• The Augmentor library is used to add more samples to classes with fewer images.
• Augmentation techniques include random rotation, flipping, and zooming.

Model Creation

• A CNN model is created to classify skin cancer types.
• The model architecture consists of convolutional layers, max-pooling layers, and fully connected layers.
• Batch normalization is applied to some layers to improve training stability.

Model Training

• The model is compiled with the Adam optimizer and sparse categorical cross-entropy loss function.
• It is trained for 50 epochs to ensure convergence.
• The training process is visualized with accuracy and loss plots.

Model Evaluation

• The model's performance is evaluated based on training and validation accuracy and loss.
• Results are visualized to assess the model's training progress and identify potential issues.

Findings

• Initial training of the model shows signs of underfitting.
• Class imbalance is identified as an issue, which prompted the use of data augmentation to balance the classes.
• Data augmentation techniques significantly improve the model's performance and reduce underfitting.
• The final model demonstrates improved accuracy and reduced loss, indicating successful training.

Conclusion

• This project highlights the importance of data preprocessing, data augmentation, and model training to address class imbalance and improve skin cancer detection using a CNN.
• The code and findings can serve as a foundation for further research and development in medical image analysis.

Instructions

• The complete code and dataset can be found in the Jupyter Notebook provided in the repository.
• Feel free to experiment with different augmentation techniques, model architectures, and hyperparameters to further enhance the model's performance.