femishajan / detect_melanoma

Detect Melanoma

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.

Conclusions

• Conclusion 1 from the analysis - Aaccuracy of the model for the Training data set is about 79%. But the Validation accuracy doesnt appear to be similar as it only about 55%. The validation loss as observed is very high. This indicates that there is overfit in the model
• Conclusion 2 from the analysis - The model accuracy for Train data is about 35% and the accuracy for the Validation set is also at 36%. Much better model compared to the previous two models as there seems to be No Overfit.And this proves that Data Augmentation has improved the model performance.
• Conclusion 3 from the analysis - The training accuracy is about 92%. The validation accuracy is nearly 76%. We have solved the overfitting , underfitting andclass imbalance issue. Much better models could be built or tried out using more epochs and more layer

Technologies Used

pathlib
tensorflow
matplotlib
numpy
pandas
pathlib
matplotlib
glob 

Contact

Created by [@femishajan] - feel free to contact me!