Real Time Skin Cancer Detection App using Transfer Learning
Skin cancer is one of the most world-wide diseases that cause death. It is the abnormal growth of skin cells, most often develops on skin exposed to the sun. But this common form of cancer can also occur on areas of your skin not ordinarily exposed to sunlight. There are three major types
Project Title
Real Time Skin Cancer Detection App using Transfer Learning
Project Area of Specialization
Project Summary
Skin cancer is one of the most world-wide diseases that
cause death. It is the abnormal growth of skin cells, most often develops on skin exposed to the sun. But this common form of cancer can also occur on areas of your skin not ordinarily exposed to sunlight.
There are three major types of skin cancer
- Basal cell carcinoma
- Squamous cell carcinoma
- Melanoma
Note: two types of carcinoma are very near to each other so usually initial categorization comes in melanoma and non-melanoma.
Early detection of these lesions may increase the curing rate to 90%. The high similarity
between different types of skin lesions makes the visual examination hard and may lead to wrong investigation. Therefore, an automated system is required for skin lesion classification.
Skin cancer is the cancer you can see. Unlike cancers that develop inside the body, skin cancers form on the outside and are usually visible. That’s why skin exams, both at home and with a dermatologist, are especially vital. Early detection saves lives. This project gives the power to detect cancer early when it’s easiest to cure, before it can become dangerous, disfiguring or deadly.
Melanoma is the deadly cancer type that even take the lives of the people immediately. The early detection of melanoma is very important to save the lives of the peole. Therefore, in this project we will develop an android app that will detect the melanoma skin cancer on real-time and can save the lives of the patients.
Project Objectives
The objective of this project is to develop a smart android app that is intelligent to detect skin cancer on real-time scenerio.
The output of this project will
- The primary objective of this Project is to develop a framework that is capable of analyzing Melanoma Skin Cancer.
- The second objective is to measure or evaluate the risk of melanoma using Skin lesion images or dermatological images.
- Finally, to detect and classify the Melanoma or Benign (Non- Melanoma).
- To reduce the gap between the dermotologists and skin cancer patients.
- To save the lives of the patients having melanoma at very early stages using this app.
- Detection and Analysis of Skin Cancer
- Binary classification of Melanoma and Non-Melanoma skin cancers
Project Implementation Method
We have used 6 transfer learning models in our project.
1. MobileNetV2
2.VGG16
3.InceptionV3
4.MobileNetV2 with Fine tuning
5. VGG16 with Fine tuning
6.InceptionV3 with Fine tuning
Our MobileNetV2 with finetuning models gives us 99.99% accuracy on melanoma detection as compared to other models.
We have used MobileNetV2 model to make our android app because it is a light weighted and more efficient model for mobile app development.
Benefits of the Project
The benefit of the Skin Cancer Detection App is to detect the pre-cancerous symptoms at an initial stage. The idea of this project is to help prevent the skin cancer or to avoid fatal complications. The advantage is to improve the condition at the very initial stage.
From this project, the people will use this easy-to-use android app that can save thier lives which is one of the most important benefit of our project.
The Application can avoid severe complications and improves the health of the person. In this way, it will impact people positively.
Technical Details of Final Deliverable
Final Deliverable:
Software Application (Android app)
Documentations
Reports
Proposals
Gantt Charts and Log books
Research Paper (Published)
Front-end development
Back-end development
Models building
Models Live testing
User Interface
Live training and testing
Model integration in android app
Result testing and validation
Final Deliverable of the Project
Core Industry
Other Industries
Core Technology
Other Technologies
Sustainable Development Goals
Required Resources
| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Documentation and Printing | Miscellaneous | 20 | 500 | 10000 |
| Graphic card | Equipment | 1 | 16035 | 16035 |
| Publish App on Playstore | Equipment | 1 | 4643 | 4643 |
| Total in (Rs) | 30678 |
If you need this project, please contact me on contact@adikhanofficial.com
Low Cost Speed Monitoring and Recognition System
Road accidents in university are a big problem which is faced by almost every week in the...
SOP\'s Enforcement Model for Pandemic Covid-19
According to the current situation, covid cases has been increasing day by day. This is du...
Development of Vehicle Control System based on Artificial Intelligence...
World is moving towards control and automation and we are living in era of 4th industrial...
Implementation and Evaluation of Experimental Setup For Knee Prosthesi...
Human gait is unique for every person on the globe. Every person is recognized by their ga...
Image Classification and retrieval on the Basis of fusion of features
In this study project, our aim is to explore the research in the area of image classificat...