About Our Project

”Dermatologist Assistant” is an mobile application and computer-aided diagnosis (CAD) tool with an objective of speeding up the detection of lesion areas in skin. Our proposed system is available for all users types that are concerned with dermatology (Skin Cancer) issues. All that appears as options through a friendly GUI, after capturing any lesion are and request classification for it.
We will implement a non-existing system with a high accuracy, speed differentiation and detection of various type of cancers . Our cancer detector goes through three essential stages. Firstly, uploading image for skin. Secondly, testing the image after applying different algorithms. Finally, classifying whether the disease is (melomna, basic cell cariona, nevus ).
Starting with pre-processing which will be performed on the uploaded image from mobile phone that will be made on the mobile . Then, segmentation by applying threshold which will be made on the server . After that is feature extraction, mainly extracting 11 features which are (Mean , Standard Deviation ,Skewness ,Kurtosis ,Smoothness, Homogenity , Contrast , Energy,Entropy,surf,sift). Final step is classification using different algorithms to train and test the dataset.
The system stores medical records to save patients medical history And the system have a doctor sign up that is made by the admin and the patient can register easily . It also has the ability of learning from miss-classified tests to enhance the future accuracy of the system.


Download Proposal Document

Download Proposal Presentation

SRS (Software Recruitment Specification)

Download SRS Document

Download SRS Presentation

Research Paper

Download Research Paper Document

Download Research Paper Presentation *** Still not finished

SDD (Software Design Document)

Download SDD Document *** Still not finished

Download SDD Presentation *** Still not finished


Class Diagram

Context Diagram

Block Diagram

About Diseases

Basal Cell Carcinoma

Researches Summaries

Four Class Classification of Skin Lesions with Task Decomposition Strategy

Deep Residual Neural Networks for Automated Basal Cell Carcinoma Detection

Basal Cell Carcinoma Detection in Full Field OCT Images Using Convolutional Neural

Automatic detection of basal cell carcinoma using vascular-extracted features from dermoscopy

Automated Detection and Segmentation of Vascular Structures of Skin Lesions Seen in Dermoscopy, with an application to Basal Cell Carcinoma Classification

About Our Team

Hagar Maher

012 266 186 57



Ehab Mustafa

015 505 601 48



Alaa Muhamed

010 951 858 79