Post by Umair Arshad

🚀 𝗘𝘅𝗰𝗶𝘁𝗲𝗱 𝘁𝗼 𝗦𝗵𝗮𝗿𝗲 𝗢𝘂𝗿 𝗙𝗶𝗻𝗮𝗹 𝗬𝗲𝗮𝗿 𝗣𝗿𝗼𝗷𝗲𝗰𝘁! After months of research, development, and experimentation, I am proud to present our Final Year Project: 𝗔𝗻 𝗘𝘆𝗲-𝗟𝗲𝘃𝗲𝗹 𝗠𝘂𝗹𝘁𝗶-𝗧𝗮𝘀𝗸 𝗗𝗲𝗲𝗽 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝗙𝗿𝗮𝗺𝗲𝘄𝗼𝗿𝗸 𝗳𝗼𝗿 𝗖𝗹𝗶𝗻𝗶𝗰𝗮𝗹 𝗗𝗶𝗮𝗴𝗻𝗼𝘀𝗶𝘀 𝗼𝗳 𝗘𝘆𝗲 𝗗𝗶𝘀𝗲𝗮𝘀𝗲𝘀 Our solution leverages 𝗔𝗿𝘁𝗶𝗳𝗶𝗰𝗶𝗮𝗹 𝗜𝗻𝘁𝗲𝗹𝗹𝗶𝗴𝗲𝗻𝗰𝗲, 𝗗𝗲𝗲𝗽 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴, 𝗠𝘂𝗹𝘁𝗶-𝗧𝗮𝘀𝗸 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴, and 𝗞𝗻𝗼𝘄𝗹𝗲𝗱𝗴𝗲 𝗗𝗶𝘀𝘁𝗶𝗹𝗹𝗮𝘁𝗶𝗼𝗻 to simultaneously detect 𝗗𝗶𝗮𝗯𝗲𝘁𝗶𝗰 𝗥𝗲𝘁𝗶𝗻𝗼𝗽𝗮𝘁𝗵𝘆, 𝗖𝗮𝘁𝗮𝗿𝗮𝗰𝘁, and 𝗚𝗹𝗮𝘂𝗰𝗼𝗺𝗮 from a single retinal fundus image, providing a faster and more efficient approach to eye disease diagnosis. 🔹 𝗞𝗲𝘆 𝗙𝗲𝗮𝘁𝘂𝗿𝗲𝘀 • 𝗠𝘂𝗹𝘁𝗶-𝗧𝗮𝘀𝗸 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴 (MTL) • 𝗞𝗻𝗼𝘄𝗹𝗲𝗱𝗴𝗲 𝗗𝗶𝘀𝘁𝗶𝗹𝗹𝗮𝘁𝗶𝗼𝗻 (Teacher–Student Framework) • 𝗡𝗼𝘃𝗲𝗹 𝗠𝗮𝘀𝗸𝗲𝗱 𝗟𝗼𝘀𝘀 𝗦𝘁𝗿𝗮𝘁𝗲𝗴𝘆 • 𝗚𝗿𝗮𝗱-𝗖𝗔𝗠 𝗘𝘅𝗽𝗹𝗮𝗶𝗻𝗮𝗯𝗹𝗲 𝗔𝗜 (XAI) • 𝗥𝗲𝗮𝗹-𝗧𝗶𝗺𝗲 𝗪𝗲𝗯 𝗔𝗽𝗽𝗹𝗶𝗰𝗮𝘁𝗶𝗼𝗻 𝗗𝗲𝗽𝗹𝗼𝘆𝗺𝗲𝗻𝘁 🛠️ 𝗧𝗲𝗰𝗵𝗻𝗼𝗹𝗼𝗴𝗶𝗲𝘀 𝗨𝘀𝗲𝗱 𝗣𝘆𝘁𝗵𝗼𝗻, 𝗧𝗲𝗻𝘀𝗼𝗿𝗙𝗹𝗼𝘄, 𝗞𝗲𝗿𝗮𝘀, 𝗗𝗲𝗻𝘀𝗲𝗡𝗲𝘁𝟭𝟮𝟭, 𝗢𝗽𝗲𝗻𝗖𝗩, 𝗥𝗲𝗮𝗰𝘁.𝗷𝘀, 𝗦𝘁𝗿𝗲𝗮𝗺𝗹𝗶𝘁, 𝗡𝘂𝗺𝗣𝘆, 𝗣𝗮𝗻𝗱𝗮𝘀, 𝗦𝗰𝗶𝗸𝗶𝘁-𝗟𝗲𝗮𝗿𝗻, and 𝗚𝗿𝗮𝗱-𝗖𝗔𝗠. 📊 𝗥𝗲𝘀𝘂𝗹𝘁𝘀 • 𝗗𝗶𝗮𝗯𝗲𝘁𝗶𝗰 𝗥𝗲𝘁𝗶𝗻𝗼𝗽𝗮𝘁𝗵𝘆: 𝟴𝟰% Accuracy • 𝗖𝗮𝘁𝗮𝗿𝗮𝗰𝘁 𝗗𝗲𝘁𝗲𝗰𝘁𝗶𝗼𝗻: 𝟵𝟴% Accuracy • 𝗚𝗹𝗮𝘂𝗰𝗼𝗺𝗮 𝗗𝗲𝘁𝗲𝗰𝘁𝗶𝗼𝗻: 𝟵𝟰% Accuracy • 𝗔𝘃𝗲𝗿𝗮𝗴𝗲 𝗔𝗰𝗰𝘂𝗿𝗮𝗰𝘆: 𝟵𝟮% • 𝗜𝗻𝗳𝗲𝗿𝗲𝗻𝗰𝗲 𝗧𝗶𝗺𝗲: 𝟬.𝟬𝟴𝘀 per image • 𝟯.𝟮× 𝗙𝗮𝘀𝘁𝗲𝗿 than separate disease-specific models A special thanks to Sir Engr. Muhammad Talha Jahangir for his continuous guidance and mentorship throughout this journey. 👥 𝗧𝗲𝗮𝗺 𝗠𝗲𝗺𝗯𝗲𝗿𝘀. Ahmad Ali Zain Sami Ullah Saad Ali Hamid 𝗔𝗹𝗶𝘀𝗵𝗯𝗮 𝗦𝗮𝗱𝗶𝗾 🎥 𝗣𝗿𝗼𝗷𝗲𝗰𝘁 𝗗𝗲𝗺𝗼𝗻𝘀𝘁𝗿𝗮𝘁𝗶𝗼𝗻 𝗩𝗶𝗱𝗲𝗼 𝗔𝘁𝘁𝗮𝗰𝗵𝗲𝗱. #ArtificialIntelligence #AI #MachineLearning #DeepLearning #ComputerVision #HealthcareAI #MedicalAI #DataScience #Innovation #Technology #TensorFlow #Python #ReactJS #FinalYearProject #FYP #MNSUET

Video Content