Fitterfly collab with Google Cloud to launch AI feature for diabetic people

Fitterfly Klik uses AI computer vision paired with the Fitterfly Nutrition Database, comprising over 37,000 Indian foods, to recognize, analyze, and break down food

Fitterfly, a leading healthtech company in India, has partnered with Google Cloud to launch its latest AI feature - Fitterfly Klik, which will enable people with diabetes and other non-communicable diseases (NCDs) to track and analyse their meals effortlessly.

Nutrition planning remains the most important bulwark for managing type 2 diabetes as well as for weight loss and heart disease. However, traditional meal tracking methods - be it through manual journaling or entering in a text format in an app - are cumbersome and discouraging for many.

Fitterfly’s latest feature, Klik, changes all this by leveraging the power of Gemini Flash 1.5 on Google Cloud’s Vertex AI platform. This transforms meal tracking into a quick and insightful experience.

“Klik makes meal logging possible in a few seconds, even with multiple dishes on a plate,” said Ammar Jagirdar, Head of Fitterfly X-Labs. “Multimodal models today are surpassing traditional image classification methods in accuracy, even on a complex problem like food.”

Fitterfly Klik uses AI computer vision paired with the Fitterfly Nutrition Database, comprising over 37,000 Indian foods, to recognize, analyse, and break down food.

Within the Fitterfly app, users can access Klik within the meal-tracking food diary by clicking on the camera icon. A user can also access Klik through a chat with JEDi, Fitterfly’s AI Coach. A user can Klik while having a meal or upload images from the picture gallery in one go in the evening. It will instantly provide feedback on that food item, its portion size and give detailed nutritional breakdowns like calorie count, macronutrient and micronutrient distribution.

Klik’s advanced algorithms analyse the food’s image and its nutritional data and directly sync it to the user’s Meal Diary on the Fitterfly app. Users can also add additional items from the same plate if all dishes aren't visible in the initial photo.

“Indians eat more than 32 cuisines and a ‘food cam AI’ was difficult to develop for both identification and serving size estimation,” said Dr. Arbinder Singal, Fitterfly's Co-founder and CEO. “Klik is a step towards helping millions of people make better choices and enabling better health outcomes.”

Bikram Singh Bedi, Vice President and Country MD at Google Cloud India said that combined with Fitterfly's dataset and Gemini's multimodal capabilities, both teams collaborated to identify the right model to help bring this powerful idea to life in an extremely short duration.

The initial outcomes from the Fitterfly Klik beta phase are encouraging: users save significant time and effort logging meals, as evidenced by a 90 per cent reduction in clicks. Validation tests confirm that Klik can precisely identify a wide range of food items and their portion sizes. This ensures that users have the most accurate nutritional information.