Glaucoma-Detection-NSGA-II
Deep learning–based Glaucoma detection using CNN, Transfer Learning, and NSGA-II optimization, with a Flask web app for real-time predictions and Grad-CAM visualizations.
- 4
- Python
- 15 mo ago
codedbyasim
I am a Software Engineer
Built from codedbyasim's public GitHub profile, Rawalpindi, Pakistan. Not on Take My Repo yet, so there are no posts or updates here.
Deep learning–based Glaucoma detection using CNN, Transfer Learning, and NSGA-II optimization, with a Flask web app for real-time predictions and Grad-CAM visualizations.
Extract and summarise data from PDFs and images using OCR + LLMs. Built with Python, OpenCV, HuggingFace, and Flask.
Recursive carbon accounting platform built with Sanity Graph-Relational Content Lake and Next.js 15. Real-time supply chain visualization, "What-If" impact simulations, and automated CSRD compliance auditing.
DeskPilot: Your autonomous AI co-pilots for desktop productivity, document generation, and workflow automation — powered by Amazon Bedrock.
Predictive hyper-local early warning for smog spikes, urban heat islands, and flash flood risk across Lahore District.
A Devvit mod tool that turns user behavior into trust signals, helping moderators prioritize queues, reward helpful members, and prevent reputation farming.
🤖 AI-powered lead generation system for freelancers. Automated lead discovery, AI scoring, contact extraction, and personalized pitches for AI & web services.
A comprehensive, role-based Learning Management System (LMS) built with Django 5.0 and PostgreSQL for university and academic management.
A Chrome Extension + Flask backend that detects phishing websites using Regex rules + ML model. Features: suspicious pattern detection, ML-based prediction, and real-time Red/Green safety indicator.
An advanced Diabetic Retinopathy screening and AI consultation system powered by Vision Transformers (ViT) and local LLMs for early retinal care diagnostics.
A Fiverr-style online service marketplace built with Django, featuring sellers, clients, service listings, orders, messaging, reviews, dashboards, and deployment setup.
This project implements a federated learning system to predict diabetes outcomes using the Pima Indians Diabetes dataset. By leveraging Flower and PyTorch, the model is trained across multiple simulated clients without ever centralizing patient data. This ensures data privacy while enabling collaborative machine learning in the healthcare domain.
A Flask-based web app that detects spam emails/SMS using Multinomial Naive Bayes and TF-IDF. Built with NLP, Scikit-learn, and NLTK for high-accuracy classification.
ThirstCap: AI scheduler that flattens data-center water spikes on the hottest days. 72h forecast-driven optimizer with time-shift, site-shift, water-power swap and tank control under a utility water cap. Xylem Global Student Innovation Challenge 2026, Water Quantity.
TERRA-SONIC: The NASA Earth Information Jukebox (2026 NASA Space Apps Challenge by TeamX) — Real-time Computer Vision & 4-bus Web Audio API sonification workstation translating NASA SVS Earth visualizations into accessible soundscapes for Blind & Low-Vision (BLV) explorers.
QuoteCage compares supplier prices for a small-business shopping list. A buyer pastes a list such as “A4 paper 10 reams, toner HP 26A x2, office chairs x4.” Each vendor is opened in a fresh browser. The run returns prices, stock, shipping, screenshots, and a recommendation
MEHFIL is a wedding-week look planner for South Asian celebrations. One portrait becomes Mehndi, Baraat, and Walima: the outfit and the grooming in a single frame, plus an event-day skin goal and a shoppable shade list with no brand names.
A synthetic-match desk that turns a 90-minute event stream into a broadcast. One NumPy engine measures the match. Four agents share a blackboard. A grounding auditor deletes any number the stream cannot prove. The screen is a live studio: pitch, lower-third, personas, and a trace of what each agent actually took.
Never let documentation drift break your AI. Autonomous web scraping, automated scraper self-healing with human-in-the-loop gates, and token-accurate RAG Q&A.