AI-Driven Skin Cancer Detection Application

Project Objective

To develop a skin cancer detection application that can assist dermatologists and individuals in identifying potential skin cancer lesions from images of moles. The goal is to create a reliable and accessible tool for early detection and diagnosis, which can potentially save lives through early intervention.

90% accuracy on the validation set

37 dermatologists in focus group

Partner's Profile

Accuracy and Metrics

The trained skin cancer detection model achieved an accuracy from 53% to over 90% on the validation set. It demonstrated high precision and recall for both benign and malignant cases, indicating its effectiveness in diagnosis.

The web-based application received positive feedback from 37 dermatologists and users alike. Its user-friendly interface and accurate predictions contributed to its adoption and usage for preliminary assessments of skin lesions.

Ongoing Improvement

By providing an accessible tool for early skin cancer detection, the application has the potential to aid in early intervention and improve the chances of successful treatment.

The model continues to be updated with new data to enhance its accuracy and keep up with evolving medical knowledge. User feedback and additional features, such as a skin lesion severity score, are being considered for future application iterations.

Solution

The combination of Scikit-learn, Keras, and the ResNet50 model was used to develop an effective skin cancer detection application. Our team built an AI-powered web application that aligns to leverage machine learning and deep learning for early disease detection and highlights the importance of ethical considerations and continuous improvement in healthcare applications.

Web-Based Application

Creation of a user-friendly web-based interface using Django. Integration of the trained model into the application. Ability for users to upload mole images for classification. Displaying the classification results with an indication of the likelihood of malignancy.

Feature Extraction with ResNet50

Utilization of the ResNet50 model pre-trained on a large dataset to extract high-level features from moles images. Removal of the top classification layers to retain the feature extraction capability.

Custom Classifier

Development of a custom neural network classifier using Keras. The classifier takes ResNet50 features as input and outputs skin cancer classifications (benign or malignant). Training the classifier on the preprocessed dataset.

Model Evaluation

Evaluation of the model's performance using metrics such as accuracy, precision, recall, and F1-score on the validation set. Hyperparameter tuning to optimize the model's architecture and training parameters.

Data Collection and Preprocessing

A curated dataset of moles images obtained from various sources. Splitting the dataset into training, validation, and test sets. Preprocess the images through resizing (500x500), format normalization (converting to RGB), and data augmentation to improve model generalization.

Technologies

Dataset

A dataset of moles images labeled as benign or malignant

Machine Learning Frameworks

Scikit-learn and Keras

Pre-trained Model

ResNet50 as a feature extractor

Programming Language

Python

Deployment

Django for creating a web-based user interface

Projects

View all

Diabetic Self-Monitoring App — DiaHero™

for improving the management and treatment of diabetes.

View case

Medication Database — QRmed

for checking the authenticity of medicines in one click.

View case

Patient Web Portal — Medcard24™

for a convenient search of doctors and health facilities.

View case

EHR Mobile App — Medcard24™

top-rated app 4.8/5 ⭐️ 500K downloads

View case

Medical Information System — Medstar™

for organized medical data storage and processing.

View case

Remote Patient Monitoring App

for remote medical diagnostics and health monitoring.

View case

Medicine Registry Software

for information sourcing of all medications in Ukraine.

View case

e-Prescribing App

for easy medical prescription issues and management.

View case

Civic Technologies Web App

the web platform for civic engagement.

View case

Learning Management System

for teaching children English.

View case

GIS Platform — KidsVisitor

for exploring the nearby events for children.

View case

Food Delivery Service App

food delivery service platform

View case

Taxi Dispatch Software

solution for dispatching and management of taxi

View case

Maritime Logistics Software

for ease and efficiency in fleet management.

View case

Travel App — World Map App

for a holiday destination finding and travel planning.

View case

SaaS for Travel Advisers

Cloud hotel booking software for travel advisors

View case

ERP for Landscaping Business — ServiceKit

for automated billing of your clients on time.

View case

Beach Management App — Loungeron

for occupancy control on the beaches.

View case

GIS Web App for Agronomists — Lanscript

a digital version of field planting history.

View case

SaaS for Gyms — Gymistration

for the ultimate sports business management.

View case

SaaS for Radiologists — Radiolance®

professional platform for interaction with radiologists

View case

Cloud Medical Diagnostics Platform — IDIS™

for an intelligent patient’s diagnosis remotely.

View case

IoT App for Heating Devices Control — Terneo™

for remote controlling of smart heating system.

View case

MVP of Astrology App — Starmantic

for personalized horoscope from astrology enthusiasts

View case

MVP of Sake App

sake beverage catalog

View case

Flutter App for Specialists in Accounting

specialized social network for accountants.

View case

MVP App — Stylist

for automated creating fashionable looks.

View case

Patent Documentation Platform — PatentPad

helps inventors create a successful patent application for their invention.

View case

AI-Driven GIS Mobile App

to automate bicycle quests

View case

Virtual Try-On Room with SMPL Anthropometry

fit clothing onto SMPL models.

View case