Non-Invasive Anemia Screening Using AI

[GENERAL] Name and surname of main applicant

Pranav Nimmagadda

[GENERAL] Are you delivering this project as an individual or as an entity (whether formally incorporated or not)

Entity (Incorporated)

[GENERAL] Requested funds in ada

100000

[GENERAL] Please specify how many months you expect your project to last (from 2-12 months)

6

[GENERAL] Please indicate if your proposal has been auto-translated into English from another language

No

[GENERAL] Summarize your solution to the problem (200-character limit including spaces)

We will develop a scalable, AI-powered mobile app that non-invasively predicts hemoglobin levels using a conjunctiva image, providing instant, low-cost anemia screening for underserved populations

[GENERAL] Does your project have any dependencies on other organizations, technical or otherwise?

No

[GENERAL] If YES, please describe what the dependency is and why you believe it is essential for your project’s delivery. If NO, please write “No dependencies.”

No dependencies

[GENERAL] Will your project’s output/s be fully open source?

No

[GENERAL] Please provide here more information on the open source status of your project outputs

Smart Contracts to Manage data licensing, payment and rights will be open

[METADATA] Horizons

Healthcare

[SOLUTION] Please describe your proposed solution

Problem:

Traditional screening methods are invasive, costly and inaccessible to large populations, particularly in rural areas. Current solutions rely on hemoglobinometers, which require trained healthcare professionals, consumables (like strips and needles) and create biohazardous waste.

Our Approach:

We propose to solve this problem by using AI-driven technology to non-invasively predict hemoglobin levels via a mobile app. Our solution leverages a neural network model that analyzes images of the conjunctiva (the inner eyelid) to estimate hemoglobin levels. This innovative approach eliminates the need for blood samples and expensive equipment, making screening accessible and affordable for underserved populations.

Target Engagement:

Our project will primarily engage:

• Public health programs: Governments and NGOs involved in large-scale health initiatives.
• Healthcare workers: Field workers with minimal technical training, enabling mass population screening in areas lacking medical infrastructure.
• Vulnerable groups Children, pregnant and lactating women and others at high risk of anemia, particularly in rural or underserved regions.

Demonstrating Impact:

We will measure success through:

• Screening Coverage: The number of individuals screened in public health programs.
• Accuracy: Targeting 95% accuracy in detecting anemia, as tested on over 22,000 data points.
• Cost Efficiency: Reducing the cost per screening significantly compared to traditional methods.
• Validation Report

Solution:

Our solution is unique because it combines non-invasive, AI-based technology with mobile accessibility, allowing for real-time hemoglobin estimation with no healthcare expertise required. This makes it scalable for national health programs and a game-changer for public health interventions.

We aim to create a decentralized marketplace on Cardano to monetize anonymized health data collected through initiatives like the AI anemia screening app. This platform will enable healthcare providers, researchers and governments to access and purchase anonymized datasets for medical research, AI model training and healthcare analytics, supporting SDG3’s goal of ensuring healthy lives and promoting well-being for all.

How It Works:

• Anonymized Data Collection: Health data from anemia screenings (e.g., hemoglobin levels) is anonymized to protect privacy.
• Decentralized Marketplace: The Cardano-based marketplace allows stakeholders to buy and sell this data securely.
• Smart Contracts: Automated, transparent contracts manage transactions and ensure fair compensation for data contributors.

Blockchain Components:

• Smart Contracts: Manage data licensing, payment and rights.
• Tokenization: Use Cardano-native tokens for data purchases and reward contributors.
• Data Provenance: Blockchain tracks data authenticity and transaction history.

Value for SDG3:

• Accelerates Research: Provides critical data for research and AI model development.
• Empowers Communities: Data contributors receive fair compensation, incentivizing participation.
• Transparency and Trust: Blockchain ensures secure, auditable transactions, building trust.

Impact:

The marketplace will foster global health innovations, enabling better diagnostics and treatments, while driving SDG3 progress by improving healthcare access and outcomes.

Importance to Cardano:

This project aligns with Cardano’s mission to improve global health through decentralized, secure solutions. By integrating Cardano’s blockchain technology, we can ensure data privacy and transparency for health programs while enabling better decision-making through real-time, secure analytics on anemia prevalence and treatment outcomes.

[IMPACT] Please define the positive impact your project will have on the wider Cardano community

Value to the Cardano Community:

Our project will bring significant value to the Cardano community by:

• Enhancing Global Health: By providing a non-invasive, AI-powered anemia screening tool, we contribute to global public health, positioning Cardano as a driver of impactful, real-world healthcare solutions.
• Data Transparency and Security: Leveraging Cardano’s blockchain, we will securely store health data and ensure privacy, demonstrating Cardano’s potential as a trusted infrastructure for sensitive, large-scale health data management.
• Real-World Use Case: Our project showcases a high-impact use case for Cardano in public health, strengthening Cardano's reputation as a blockchain platform that goes beyond finance, into sectors like healthcare and social good.

Measuring Impact:

We will measure the success of our project through both quantitative and qualitative metrics:

• Quantitative Metrics:
• Number of screenings conducted: We aim to screen over 1 million individuals in the first year of deployment.
• Accuracy of hemoglobin predictions: Targeting an accuracy rate of 95% based on hemoglobin levels compared to clinical tests.
• Cost reduction: Significant reduction in the cost of anemia screening per individual, as compared to traditional invasive methods.
• Qualitative Metrics:
• Feedback from public health officials and healthcare workers: We will gather insights on ease of use, effectiveness in rural areas and scalability for national programs.
• User experience: Collecting testimonials from individuals screened by the app to measure satisfaction and trust in the technology.

Sharing Outputs:

We will share our project’s outputs and opportunities through multiple channels:

• Open Data Reports: Publicly available reports on the number of screenings, accuracy rates and real-time analytics via secure blockchain infrastructure.
• Collaborations with Health Programs: Engage with public health departments, NGOs and governments to demonstrate how Cardano can be used to tackle health crises through scalable technology.
• Community Engagement: Regular updates and results will be shared with the Cardano community through Catalyst channels.
• Open-Source Integration: The AI model and technical infrastructure will be available to the Cardano developer community, fostering innovation and adaptation for other healthcare applications.

[CAPABILITY & FEASIBILITY] What is your capability to deliver your project with high levels of trust and accountability? How do you intend to validate if your approach is feasible?

Team Expertise:

We have a highly experienced and multidisciplinary team equipped to deliver this project with precision and accountability:

• Dr. N.S.D. Prasad Rao, Founder of SDG3 Health, brings 38 years of experience in healthcare and public health. He has led large-scale projects that have significantly impacted patient outcomes across India.
• Dr. Pranav, Co-founder and CEO of SDG3 Health, has a background in both medicine and business management. His experience in leading strategy and operations for reputed hospitals and health programs in India and the UAE positions him well to execute this project at scale.

This team’s deep healthcare knowledge, combined with expertise in technology-driven solutions, makes us uniquely qualified to deliver the proposed AI-powered anemia screening app.

Feasibility Validation:

Our approach has already been validated through:

• Prototype Development: A functional prototype of the anemia screening app has been developed, tested on 22,000 data points and has shown an average percentage error of under 7%, with a mean absolute error of 0.69 g/dl.
• Real-World Testing: The app has been tested in collaboration with public health programs such as Anemia Mukt Bharat and other health initiatives in Telangana, providing practical proof of its effectiveness in large-scale screenings.
• AI Model Validation: Using a state-of-the-art deep neural network, we have demonstrated high levels of accuracy in hemoglobin prediction. Further refinement, guided by user feedback and continuous data input, will ensure sustained accuracy.

Processes to Ensure Trust and Accountability:

• Transparent Fund Management:
• Regular Financial Reports: We will provide transparent financial reporting at each milestone of the project. This includes detailed breakdowns of how funds are allocated and spent on AI development, mobile app updates, data collection and deployment.
• Milestone-Based Funding: Funds will be managed in phases, with clear milestones defined for each stage of the project (e.g., development, testing, deployment). This will ensure that funds are disbursed in a controlled manner, only after key deliverables are met.
• Governance and Oversight:
• Community Engagement: We will actively engage with the Catalyst community, providing regular updates on our progress, challenges and achievements through Catalyst channels. Feedback from the community will guide project decisions and ensure we remain accountable.
• Use of Blockchain for Transparency: By integrating Cardano’s blockchain, we ensure that all data related to the project, including results from anemia screenings and public health reports, are stored securely and transparently, accessible for auditing by stakeholders.
• Risk Mitigation:
• Quality Assurance (QA): Rigorous testing and quality assurance processes will be implemented, particularly around the mobile app’s performance and image quality validation modules.
• Continuous Monitoring and Adjustments: We will monitor the app’s performance in real-time and adapt the AI model based on the results and user feedback, ensuring the solution stays relevant and reliable for users.

Proven Track Record:

Our team has successfully executed similar projects in healthcare. We have a strong track record of delivering projects with measurable health outcomes and have worked with large healthcare systems in both the public and private sectors. Our leadership in launching this AI-based anemia app is backed by years of industry expertise and familiarity with the complexities of public health deployment.

[PROJECT MILESTONES] What are the key milestones you need to achieve in order to complete your project successfully?

Milestone 1: Refinement of AI Model & Image Quality Validation Module (1 month)

A: Milestone Outputs:

• Refinement of the AI model for hemoglobin prediction to improve accuracy.
• Development of the image quality validation module to ensure high-quality image input (addressing issues like motion blur, brightness and clarity).

B: Acceptance Criteria:

• AI model reaches 95% prediction accuracy with a maximum error rate of less than 7%.
• Image validation module successfully filters out poor-quality images in real-time with 90% precision.

C: Evidence of Milestone Completion:

• Technical report on AI model performance improvements and validation metrics.
• Video demo showing how the image quality validation module works within the app.
• Signed-off test results from internal QA team validating improvements in model performance and image quality checks.

Milestone 2: Mobile App Development for Data Collection (1 month)

A: Milestone Outputs:

• Completion of mobile app development with features for capturing conjunctiva images, user data entry (age, gender, pregnancy status) and cloud data submission.
• Integration of the AI model and image validation module into the app.

B: Acceptance Criteria:

• Mobile app successfully captures and uploads test data for at least 100 subjects.
• Seamless integration of AI model and validation module into the app with real-time predictions.

C: Evidence of Milestone Completion:

• Submission of a working mobile app for testing by the Catalyst community.
• Video walkthrough of the app showcasing its functionality.
• Documentation of app performance metrics (e.g., upload speed, prediction time) and user testing results.

Milestone 3: Pilot Testing and Data Collection (2 month)

A: Milestone Outputs:

• Pilot deployment of the app in a public health program or NGO partnership (e.g., Anemia Mukt Bharat).
• Collection of data from at least 1,000 subjects across different hemoglobin levels to further refine the model.

B: Acceptance Criteria:

• Collection of quality conjunctiva images and accurate hemoglobin data from at least 1,000 individuals.
• AI model delivers prediction accuracy of 95% based on real-world pilot data.

C: Evidence of Milestone Completion:

• Report on pilot results, including the number of subjects screened, accuracy levels and performance insights.
• Signed collaboration agreement with a public health entity or NGO for pilot deployment.
• Feedback from field health workers on the app’s ease of use and effectiveness.

Milestone 4: Full-Scale Deployment & Public Health Program Integration (1 month)

A: Milestone Outputs:

• Full-scale deployment of the app across multiple regions (target of 100,000 screenings).
• Integration of the app with existing public health data systems (e.g., Anganwadi Information System).

B: Acceptance Criteria:

• Successful deployment in at least 3 regions, with a minimum of 100,000 individuals screened.
• Real-time data syncing with public health systems, ensuring accurate reporting of hemoglobin levels.

C: Evidence of Milestone Completion:

• Deployment report including the number of users, regions covered and screenings performed.
• Confirmation of integration with government health systems through API or data-sharing documentation.
• Testimonials from public health officials or program leads.

Milestone 5: Project Close-out Report and Video (1 month)

A: Milestone Outputs:

• Completion of a comprehensive project close-out report summarizing all phases of development, testing and deployment.
• Video detailing the project outcomes, lessons learned and the app's impact on public health.

B: Acceptance Criteria:

• Submission of final project close-out report and video to the Catalyst community.
• Report includes financial breakdowns, project achievements and next steps for sustainability or further scaling.

C: Evidence of Milestone Completion:

• Publicly available close-out report uploaded to Catalyst.
• Final project video shared with the Cardano community through relevant channels.

[RESOURCES] Who is in the project team and what are their roles?

Project Team and Roles

Dr. N.S.D. Prasad Rao – Founder and Project Lead

• Role: Overall project leader, responsible for strategic direction, partnerships and stakeholder management. He will ensure the project aligns with public health objectives and oversee collaboration with government entities and public health programs.

• Experience: Dr. Rao has 38 years of experience in healthcare management, with a focus on leading large-scale health initiatives. He has extensive experience in working with public health organizations, ensuring the project is well-positioned to integrate into existing government health programs.

Dr. Pranav – Co-Founder and Chief Operating Officer (COO)

• Role: Responsible for day-to-day operations, managing project execution, timelines and resources. Dr. Pranav will lead the app's deployment, coordinate field testing and work closely with public health workers to ensure successful implementation.

• Experience: Dr. Pranav is a medical doctor and a healthcare entrepreneur with extensive operational experience. He has led large hospital networks and health programs in India and the UAE, making him adept at scaling health initiatives.

Rakesh J – AI Development, Data Analysis and Model Validation

• Role:
• The AI Specialist will be responsible for refining the AI model for hemoglobin prediction, improving the accuracy of the model and ensuring the smooth integration with the mobile app. This role includes developing and validating the image quality module for the app.
• Responsible for analyzing the data collected from the pilot testing and fine-tuning the AI model based on the results. The data scientist will ensure the quality and reliability of data collected and work closely with the AI specialist to continuously improve the model.

Akash – Full Stack and Cardano Developer

• Role: Responsible for the design and development of the mobile application, ensuring seamless data collection and integration with the AI model. This includes user interface (UI) design, backend development for cloud storage and real-time image processing within the app.

[BUDGET & COSTS] Please provide a cost breakdown of the proposed work and resources

Cost Breakdown (100,000 ADA Request)

• AI Model Refinement & Development (25,000 ADA)
• AI Specialist: Refine the AI model and integrate image quality validation.  
• Cloud Computing: GPU resources for model training.

• Mobile App Development (40,000 ADA)
• App Developer: Build the app, integrate the AI model and develop UI.
• Backend & Infrastructure: Cloud storage, API, security tools.

• Data Collection & Pilot Testing (15,000 ADA)
• Data Scientist: Analyze pilot data to refine model accuracy.
• Public Health Consultant: Coordinate pilot programs and regulatory approvals.

• Publicity & Community Engagement (15,000 ADA)
• Marketing & Outreach: Promote the app to public health programs.
• Content Creation: Reports, videos and community updates.

• Final Reporting & Close-Out (5,000 ADA)
• Report & Video: Produce a final project report and close-out video.

[VALUE FOR MONEY] How does the cost of the project represent value for money for the Cardano ecosystem?

Value for Money for the Cardano Ecosystem

Our project offers exceptional value to the Cardano ecosystem by delivering an innovative, real-world health solution that leverages blockchain for transparency, security and scalability. Here's how the costs reflect value for money:

• Impactful, Real-World Use Case:
• By addressing a global health challenge (anemia), our project positions Cardano as a leader in health tech, showcasing its potential to deliver life-changing solutions.
• The project's scalable nature, through non-invasive, AI-powered screenings, enables it to reach millions, creating a massive societal impact while also expanding Cardano's use in healthcare.

• Proportional Costs:
• The AI Specialist and App Developer costs are aligned with industry standards for experienced professionals in the healthcare AI and mobile development sectors. We based these costs on typical freelance rates for AI development ($80-$100/hour) and mobile app development ($60-$90/hour), ensuring we are paying competitive but fair rates.
• Cloud computing costs for AI model training and app hosting are based on competitive pricing from services like AWS and Google Cloud, optimized to ensure cost-efficiency for large-scale deployments.

• Low Overhead for High Impact:
• Compared to traditional medical devices (like hemoglobinometers), our app provides a low-cost solution with minimal recurring expenses (no consumables like test strips). This makes it ideal for public health programs, providing substantial cost savings long-term.
• By using blockchain for secure data storage and reporting, the project ensures transparent and immutable records for public health programs, increasing trust and reducing the administrative costs of health data management.

• Sustainable Growth:
• The costs of publicity, outreach and engagement are critical to expanding adoption and driving large-scale deployment. The value created by integrating public health programs and securing partnerships will ensure the project’s long-term sustainability and impact.