https://github.com/assasinator/segmentation-in-skin-lesion-classification-
Melanoma is one of the most dangerous types of skin cancers. So, early detection of this cancer can help the patient recover better. This is a course project where we tried to implement several classification algorithms both on segmented and non-segmented images to check the effect of segmentation on the classification. We also tried to explore the performance metrics of different classification algorithms.Our research has shown that, for small dataset the traditional CNN works better when the image is segmented and for the images without segmented the transfer learning algorithm works better compared to other classification algorithms.
https://github.com/assasinator/segmentation-in-skin-lesion-classification-
classification-algorithm deep-neural-networks image-processing machine-learning-algorithms medical-imaging segmentation
Last synced: 12 months ago
JSON representation
Melanoma is one of the most dangerous types of skin cancers. So, early detection of this cancer can help the patient recover better. This is a course project where we tried to implement several classification algorithms both on segmented and non-segmented images to check the effect of segmentation on the classification. We also tried to explore the performance metrics of different classification algorithms.Our research has shown that, for small dataset the traditional CNN works better when the image is segmented and for the images without segmented the transfer learning algorithm works better compared to other classification algorithms.
- Host: GitHub
- URL: https://github.com/assasinator/segmentation-in-skin-lesion-classification-
- Owner: assasinator
- License: mit
- Created: 2022-01-16T13:58:18.000Z (over 4 years ago)
- Default Branch: main
- Last Pushed: 2022-01-16T14:11:09.000Z (over 4 years ago)
- Last Synced: 2025-05-31T11:44:57.548Z (about 1 year ago)
- Topics: classification-algorithm, deep-neural-networks, image-processing, machine-learning-algorithms, medical-imaging, segmentation
- Language: Jupyter Notebook
- Homepage:
- Size: 2.01 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0