Aflatosafe – Smart Aflatoxin Detection and Management

While food is essential for survival, it can also become a medium for silent and deadly threats.

In Kenya, one such threat has emerged in the form of aflatoxins, highly toxic secondary metabolites produced primarily by the fungi Aspergillus flavus and A. parasiticus. These compounds have been detected in maize samples at concentrations as high as 58,000 μg/kg, posing a significant public health risk.

Historically, aflatoxin contamination has had dire consequences in Kenya, most notably during the 2004 outbreak of aflatoxicosis, which resulted in 317 reported cases and 125 deaths. The contamination tends to occur when crops experience drought stress followed by rainfall near harvest time, conditions that favor fungal growth and toxin production.

The health implications of aflatoxin exposure are severe and wide-ranging, including:

  • Acute hepatitis resulting from short-term, high-dose exposure
  • Hepatocellular carcinoma (liver cancer) linked to long-term intake
  • Stunted growth and impaired development in children
  • Premature births and fatal abnormalities

Amid such a critical situation, we introduce Aflatosafe, a one stop platform for detection and prevention of Aflatoxin. Aflatosafe stands as an end to end solution that addresses key aspects of the Aflatoxin crisis while providing relevant insights to the government for efficient resource distribution and creating a community platform for farmers in Kenya to connect and discuss on the issues they face.

Key Aspects –

We are here to provide a mobile app that will have features based on the user using it.

  1. If the user is a farmer – 

i. The farmer will be able to input the 15 farm related features to the app and machine learning algorithms in the app will be capable of predicting if there is contamination in the soil or not based on inputs from farmers.If there is it provides suggestions to the farmer on how it can be improved.

ii. The farmer will be connected with the other farmers, through the community feature of the app, to discuss with other farmers about pressing issues they face and collaborate to find solutions to those issues.

  1. If the user is a Government of Kenya – 

The Government will be accessing the interactive dashboards that we will provide to deliver actionable insights based on the aflatoxin contamination data collected through the app. 

  1. If the user is a Buyer or Consumer of crops – 

The buyer can access the feature that scans the crop product and takes certain parameters as inputs to decide if the crop is contaminated or not. Here we aim to empower the buyer with this tool to purchase the best quality uncontaminated products. 

  1.  Multilingual and Inclusive interface – 

The Application is available in multiple languages that are suitable for the people of Kenya and the design of the application will be inclusive to create impact on more people.

Following show the AI chatbot and couple of images of the app

AI based Chatbot

App Demo

Role of AI 

  1. AI based Chatbot – We have planned to integrate the app with AI based chatbot that will help the farmer by providing guidance related to Aflatoxin Intervention techniques, good farming practices, storage suggestions and many more. 
  2. Inclusive approach – Considering the inclusivity of the solution we have developed a chatbot that responds with both text and voice media, thus easing the use of app for the  farmer.
  3. Aflatoxin Detection – Deep Learning based Convolutional Neural Networks are deployed to detect whether the crop is contaminated or not. These neural networks achieved an accuracy of 94% for predicting contamination in corn products.

Sustainable Business Model 

Sustainability is a pivotal concept in today’s business landscape. Businesses nowadays focus on the aspects like usage of green energy and reducing carbon footprint. In this section we explore the business model of our solution

  1. Key Partners
  • Government of Kenya: For aflatoxin dashboards, distribution, suggestions on how to reduce aflatoxin.
  • Maize buyers and traders
  • Farmers
  1. Key Resources
  • Pretrained ML models 
  • API keys
  • Flutter based mobile app
  • Farmer contributed data and user feedback
  1. Key Activities
  • App development for farmers, government officials, buyers. 
  • Model training and improvement
  • Data Collection
  • Educating farmers about aflatoxin.
  1. Value Propositions
  • Providing suggestions in the pre flowering stage to prevent aflatoxin when the plant is grown. 
  • Detecting aspergillus flavus and aspergillus parasiticus and then finding out if aflatoxin is present in the crop.
  • Trust & Traceability: Enabling transparent, verifiable supply chains that benefit all stakeholders
  1. Customer Segment
  • Farmers → Primary users of risk prediction,chatbot and get tailor made suggestions.
  • Buyers/Traders → Pay for verified maize sourcing
  • Government Officials → Use dashboard for intervention and hotspot detection.
  1. Customer relationships-

We aim to build Long-term trustworthy relations with the customers by providing cutting edge tools and ensuring significant reduction in Aflatoxin contamination.

  1. Revenue Streams
  • Subscription Model for Buyers: 

Buyers pay a monthly fee to access visual verification and batch-level aflatoxin risk data. With this even the buyer gets a badge of verification that produce with less aflatoxin is sold here encouraging more people to buy his produce

  • Government grants in rural areas.
  • Freemium app for farmers:Farmers can get access to consultancy services,detailed visual insights on what is wrong with their crop,access to verified buyers.
  • Affiliate Marketing-By marketing certain fertilizers or products approved by NGOs and related organisations revenue can be generated.
  1. Channels – 
  • Direct sales
  • Partnering up with NGOs and organisations like AgriAssure.
  1. Cost Structure-
  • Overhead costs for cloud and API calling,
  • Marketing and distribution cost

Externalities: The ripple effect of our solution

Our solution brings a range of unintended yet impactful consequences beneficial and others require thoughtful mitigation.
Positive Externalities
1. Enhances public health by reducing aflatoxin-related diseases.

2. Strengthens food safety from farm to fork.

3. Promotes sustainable farming practices.

4. Enables the government to take data-backed decisions

5. Increases consumer trust through transparent traceability

6. Contributes to multiple UN Sustainable Development Goals.

7. Educates farmers and builds long-term digital literacy in rural communities.


Negative Externalities


1. Leakage of sensitive farmer and location data.

2. Risk of model bias leading to false negatives or false positives

3. Limited accessibility for farmers lacking smartphones or reliable internet.

4. Dependence on consistent API data and device uptime for accuracy.

Sustainable Development Goals (SDGs) 

Aflatosafe directly contributes to several United Nations Sustainable Development goals as mentioned below – 

SDG 2: Zero Hunger

Our platform by reducing crop loss and enhancing food safety, we ensure more secure food supplies.

SDG 3: Good health and Well-being

Our platform with early detection of aflatoxin before crop and detection while choosing crop  helps prevent diseases caused by aflatoxin.

SDG 9: Industry, Innovation and Infrastructure

Our platform integrates modern AI solutions to traditional farming, creating agri tech systems.

SDG 12: Responsible Consumption and Production:

Our platform promotes sustainable crop monitoring before and after harvest, and is planning to reduce post-harvest by repurposing contaminated produce.

SDG 13: Climate Action

To minimise crop loss, our platform helps in better soil monitoring and stress prediction, farmers adapt to faster changing climatic conditions.

SDG 17: Partnership for Goals:

To create a scalable impact our platform builds bridges between farmers, government and NGOs. 

Inclusivity

Tech only works if people can actually use it. Here’s how we’re making sure this doesn’t stay locked in a lab:

  • Voice-First Design:
    Most of our farmers prefer speaking over typing. That’s why we built a voice-based chatbot, so even low-literacy users can get answers quickly.
  • Offline Support:
    The Internet in rural Kenya isn’t always reliable. Our app works offline and syncs when it can, so farmers aren’t left hanging.
  • Simple to use:
    We guide users through one step at a time—no clutter, no confusion. It’s designed for real people, not tech bros.
  • Explains “Why,” Not Just “What”:
    If the app tells you your aflatoxin risk is high, it also tells you why—and what you can do about it. That builds trust and helps farmers learn as they go.
                  

Case Study

Imagine you’re a maize farmer just outside Nairobi. It’s mid-season, the rains have been all over the place, and it’s been about two weeks since you planted your first seeds.

Amidst this there is a tinge of worry “what if my crop has aflatoxin?”, “what if I am unable to sell my produce?”,then you open the Aflatosafe app.

The home screen gets you in our preferred language and can sign up as a farmer or a buyer. From the farmer’s side, they need to put up 7 parameters and 8 parameters will be filled through APIs to check the health of the soil. Within seconds this app warns if the soil is contaminated with an AI summary.

From the buyer’s side, the buyer just scans the crop and within seconds this app warns if the crop is contaminated. In both cases the government is notified with the region having the aflatoxin so that it further plans accordingly for the safety measures to take care of. 

Further if we scroll to the community feed, there are other farmers who can communicate with each other.

Our platform has a chatbot, in which one can receive the necessary information for the crop growth. 

Ahead and future vision

The journey of Aflatosafe is just a beginning.

This platform is designed to develop as a scalable, adaptable platform that meets the active challenges of agriculture and food safety. We envision the following expansions and innovations in the upcoming phases:

  1. Geographic Expansion
    We aim to extend our platform beyond Kenya into high-risk regions enabling aflatoxin management spread in all regions.
  1. Multilingual Support

To ensure that no farmer is left out due to language barriers, our platform plans to extend our language system to include more regional and tribal languages.

  1. Extension of partnership

We’re opening doors for deeper collaboration with governments, NGOs, researchers, and agri-startups. These partnerships will help us scale faster, reach further, and solve problems more holistically.

  1. IoT Integration

We are working on an affordable IoT device that helps to know the percentage of land in risk, inserting the device into land.

  1. Beyond Aflatoxin

The problem is not only Aflatoxin. Tomorrow’s risks may occur due to fungal infections, pests, or climate-driven stressors. AgriAssure is developing into broader crop disease early-warning system, powered by deep learning.

  1. Giving New Life to Waste

What if a contaminated crop did not mean a total loss?We’re exploring ways to reuse infected produce for biofuel, compost, or other eco-safe applications. Our goal is to create a zero waste agriculture ecosystem.

  1. Farmer Credit Scoring

Our platform aims to grade every farmer so that he has a profile that becomes their key to unlock fair credit, insurance and better market opportunities by saving all the actions of the farmer from scanning their crops to reporting contamination and ensuring safe harvests contribute to a digital reputation.

Acknowledgments

We, the Aflatosafe team, would like to express our heartfelt gratitude to everyone who guided, encouraged, and supported us throughout this journey. We are deeply thankful to the Technical University of Denmark (DTU) for giving us the opportunity to work on a socially impactful project that bridges the gap between technology and real world problems. This experience has truly expanded our perspective.

Our sincere thanks go to our mentors and guides, whose tireless support, sharp insights, and thoughtful feedback were helpful at every step. Your encouragement helped us innovate our thinking, stay grounded in impact, and push boundaries when needed.

A special thanks to our partner organisations and domain experts who offered valuable inputs on the aspects of scoping, ideation and execution. Your field insights helped us align our solution with community needs.

And finally, to every team member who poured in hours of effort, late-night brainstorming, and tireless energy thank you for your collaboration, trust, and shared passion to build something that matters.

Conclusion

In conclusion, our platform Aflatosafe represents more than a technological innovation. It is a step towards safer, smarter and more inclusive farming techniques. We have used AI, IoT and community driven design that addresses real challenges faced by farmers, government, agencies and consumers. These learning sessions not only strengthened our technical background but also made our deep understanding of how technology can be used for social good. As we look ahead, we are excited about the potential to scale this solution, expand its reach to new regions and continue building tools that empower communities to ensure food security.

Team BlueVerse-YA-IN-1

By

Aryan Karmore

Aditya Bhalerao

Raveena Basarimarad