Around a year ago I stumbled upon a YouTube video that completely changed the way I look at food. That video was called 'The Disturbing Reality Of Ultra-Processed Food' and I highly recommend giving it a watch. Now, prior to watching this video, I've heard a million times that "processed foods are bad for you" and I've sort of always just brushed it off and never really thought anything significant of it. But thanks to the YouTube algorithm, I ended up watching this video and it all just clicked. I realized in that moment why ultra-processed foods are actually bad for you, and left with a few key points that really stuck with me... that the vast majority of my diet (jimmy dean breakfast sandwiches, protein bars, takeout etc), consisted not of food, but of 'food-like products' and 'pre-digested foods'. I then went down a rabbit hole and watched some more videos on the topic and read a bunch of articles until I finally felt enough willpower to decide that I would make the lifestyle change of replacing all processed and ultra-processed foods in my diet with NOVA category 1 food items (unmodified, whole food items) that I would prepare myself. So I went to the supermarket a few days later and realized it was actually a lot harder than I had initially thought to identify and mentally categorize items based on their level of processing and which NOVA group they'd fall into. So I went searching online and on the app store for a scanner app that would do this for me. I found a few apps like Yuka and OpenFoodFacts, and they were helpful for classifying food items with barcodes, but only worked for food items with a barcode. Using these apps, I was unable to scan a picture of a food item directly, and could only determine a food item's level of processing if it had a label. I couldn't scan prepared meals, stuff from restaurants, or really any food item that had been removed from its packaging and put on a plate. In addition to this, the search functionality of these apps was also limited to strictly foods with a label, so I couldn't even look up a non-labeled food item to determine its level of processing. After searching and searching, I found that none of these scanner apps did what I wanted them to do, so I decided I would build my own. I then got to work and built an app that I felt was actually useful and that did what I wanted it to do, and that picks up the slack where Yuka, OpenFoodFacts and others falls short. After much trial and error, I was able to finish the initial version of my ultra-processed food scanner app, and I gave it the name 'NOVAScanner', which I felt had a nice ring to it. Unlike other scanners available, NOVAScanner uses AI, which allows the scanner to be much more flexible, smart, and accurate. NOVAScanner is able to essentially "see" the food item of an uploaded image, rather than just scan a label, and is thus able to scan any food item directly. To use it, simply snap a picture using the camera, or upload an image from camera roll, and receive back a NOVA categorization, processing score, and brief reasoning for why that food item was categorized the way it was. You can scan prepared dishes, packaged food items, and anything in between. The app helps me out a ton personally (which it should seeing as I'm the one who made it), and I hope it can help out others as well. Like anything that uses AI, the scanner can incorrectly classify food items from time to time, for example it may incorrectly classify chicken as turkey or vice versa. To mitigate this issue, I've added the ability to type in the name of the actual food item and re-scan for a correct analysis, if NOVAScanner gets it wrong the first time. Right now, NOVAScanner is available as a website, but I am also in the process of making an iOS app for it. I will continue to improve the scanner over time and am all ears to any feedback or suggestions you may have to improve the app. Try out NOVAScanner for yourself and let me know what you think :)

NOVAScanner AI

Ultra-Processed Food Detector

Ultra-Processed Food Detector