Melanoma-Detection-Assignment

The project uses a Convolutional Neural Network based model to assess the skin lesions and determine whether the person is affected with Melanoma or other diseases.

Problem Statement

To build a CNN based model which can accurately detect melanoma. Melanoma is a type of cancer that can be deadly if not detected early. It accounts for 75% of skin cancer deaths. A solution that can evaluate images and alert dermatologists about the presence of melanoma has the potential to reduce a lot of manual effort needed in diagnosis.

Output is classified into:

1. Actinic keratosis
2. Basal cell carcinoma
3. Dermatofibroma
4. Melanoma
5. Nevus
6. Pigmented benign keratosis
7. Seborrheic keratosis
8. Squamous cell carcinoma
9. Vascular lesion

Project Pipeline :

1. Loading of Data
2. Data Visualisation
3. Creation of base model
4. Choosing an Augmentation strategy and building the model
5. Training and evaluating the augmented model
6. Dealing with Class Imbalance using Augmentator
7. BUilding the final model and Evaluating it.

Results

The final model achieved an accuracy of 91% with a loss of 0.25. The Validation accuracy we achieved with model is 86%. The model is a good fit. Using the Augmentor and dealing with class imbalance helped the model to achieve better accuracy.