BearHack

This past weekend, I had the opportunity to participate in BearHack 2024 with my roommates Osvaldo Ortiz, Omar Nosseir and Jacob Hensley. A few days before the hackathon, we sat down and brainstormed ideas while exploring the possible tracks: Best Beginner, Best UI/UX, Best Wearable, and Best Start-Up Idea. We generated numerous ideas for each track and then realized, why not compete in all? We pooled our creativity and conceived SkinVue! SkinVue is a wearable device and platform designed to scan skin lesions and classify them as benign or malignant. In cases of malignancy, it generates a distribution indicating the likelihood of specific diseases or cancers. Our wearable device ensures consistently high-quality images, which are transmitted to our server housing a convolutional neural network for classification. Following classification, the image and its categorization are stored in the user's account for historical reference and future tracking. For the tech enthusiasts, I leveraged my favorite frontend tech stack, comprising of React, TailwindCSS, Framer Motion, and Redux. Additionally, I opted for an AWS EC2 instance to host our Flask server, housing our CNN model developed and trained using Python and Tensorflow. For our datastore, a PostgresSQL database called Supabase was chosen. Jacob and Osvaldo worked with the Arduino Uno R3 and ArduCam Mini for our hardware setup. Omar crafted the 3D model using Fusion 360 and coordinated its printing with our off-campus 3D printing expert. Jacob and Omar came to a victory at about 2 am fixing an incompatibility with the hardware components giving us the final piece of hope we were looking for. Our diverse skill sets converged seamlessly, enabling us to execute every component—from frontend and backend development to AI modeling, 3D printing, and hardware integration—within 24 hours. Despite the sleepless marathon, we knew it was a venture worth undertaking. We started at 10 am and finished at about 9 am the next day. Within hours, we would come to find out our project had won the Best Overall Hack award! Check out the Github repo here: https://lnkd.in/gYrnD5DK

This past weekend, I had the opportunity to participate in BearHack 2024 with my roommates Osvaldo Ortiz, Omar Nosseir and Jacob Hensley. A few days before the hackathon, we sat down and brainstormed ideas while exploring the possible tracks: Best Beginner, Best UI/UX, Best Wearable, and Best Start-Up Idea. We generated numerous ideas for each track and then realized, why not compete in all? We pooled our creativity and conceived SkinVue! SkinVue is a wearable device and platform designed to scan skin lesions and classify them as benign or malignant. In cases of malignancy, it generates a distribution indicating the likelihood of specific diseases or cancers. Our wearable device ensures consistently high-quality images, which are transmitted to our server housing a convolutional neural network for classification. Following classification, the image and its categorization are stored in the user's account for historical reference and future tracking. For the tech enthusiasts, I leveraged my favorite frontend tech stack, comprising of React, TailwindCSS, Framer Motion, and Redux. Additionally, I opted for an AWS EC2 instance to host our Flask server, housing our CNN model developed and trained using Python and Tensorflow. For our datastore, a PostgresSQL database called Supabase was chosen. Jacob and Osvaldo worked with the Arduino Uno R3 and ArduCam Mini for our hardware setup. Omar crafted the 3D model using Fusion 360 and coordinated its printing with our off-campus 3D printing expert. Jacob and Omar came to a victory at about 2 am fixing an incompatibility with the hardware components giving us the final piece of hope we were looking for. Our diverse skill sets converged seamlessly, enabling us to execute every component—from frontend and backend development to AI modeling, 3D printing, and hardware integration—within 24 hours. Despite the sleepless marathon, we knew it was a venture worth undertaking. We started at 10 am and finished at about 9 am the next day. Within hours, we would come to find out our project had won the Best Overall Hack award! Check out the Github repo here: https://lnkd.in/gYrnD5DK