Oemah Sehat

Stunting is a serious public health problem in Indonesia, affecting around 27.7% of children under five in 2019. Stunting can cause impaired physical growth and cognitive development in children, as well as increasing the risk of chronic diseases in the future. Therefore, efforts are needed to prevent and treat stunting from an early age
machine learning-based Android application with educational and counseling features for preventing and treating stunting in toddlers. This application can be used by parents or caregivers of toddlers to measure and monitor the nutritional status of toddlers, get information and advice about toddler nutrition and health, and interact with professional health workers if necessary

This application uses machine learning to predict the risk of stunting in toddlers, not just measuring the nutritional status of toddlers using the z-score method. Machine learning can help recognize patterns and risk factors for stunting from more comprehensive anthropometric, health and environmental data on toddlers.
This application provides educational and counseling features for preventing and treating stunting in toddlers, not only providing information about the nutritional status of toddlers. Education and counseling features can help improve users' nutritional and health knowledge and behavior as well as provide psychosocial support for users

Mothers of toddlers, Human Development Cadres, Village Heads who number in the millions in Indonesia

TBA

The solution we offer is an Android-based health service, where almost everyone now has a smartphone. Coupled with machine learning in the application which makes the service even better. Including education and counseling features in the application

Teknologi kami akan sesuai dengan Tujuan Pembangunan Berkelanjutan ke-3 dalam Agenda Pembangunan Berkelanjutan tahun 2030 adalah “menjamin kehidupan yang sehat dan meningkatkan kesejahteraan bagi semua orang di segala usia”

• Designing an Android-based machine learning application interface with education and counseling features for the prevention and treatment of stunting in toddlers using the user-centered design method, namely a method that involves users in the design process to ensure that the application suits the user's needs and preferences.
• Develop an Android-based machine learning application with educational and counseling features for preventing and treating stunting in toddlers using the Java programming language and the TensorFlow Lite framework, a framework that allows developing machine learning applications on mobile devices.
• Collect anthropometric, health and environmental data on toddlers from 100 parents or caregivers of toddlers who are willing to be research respondents. Anthropometric data includes height, weight, head circumference and upper arm circumference of toddlers. Health data includes history of illness, immunizations and exclusive breastfeeding of toddlers. Environmental data includes socio-economic status, sanitation, and access to family health services. This data will be collected through the Android application that has been built.
• Choose the most suitable machine learning algorithm to predict the risk of stunting in toddlers from the data that has been collected. The machine learning algorithms that will be tried are decision trees, k-nearest neighbor, support vector machines, and neural networks. Algorithm selection will be based on the criteria of accuracy, precision, sensitivity, specificity and f-measure.
• Provide educational and counseling features that are relevant and useful for users. The educational feature will provide information about the definition, causes, impacts and prevention of stunting in toddlers. The counseling feature will provide advice and recommendations regarding toddler nutrition and health according to the results of stunting risk predictions. The counseling feature will also facilitate communication between users and health professionals if necessary.
• Evaluate the impact of using the application on reducing the prevalence of stunting in children under five. The evaluation will be carried out by measuring changes in the nutritional status of toddlers after using the application for three months. Changes in the nutritional status of toddlers will be measured using the z-score indicators of height for age (HAZ), weight for age (BAZ), weight for height (WHZ), and body mass index for age (BMIZ). Apart from that, the evaluation will also measure the level of user satisfaction and engagement with the application using a questionnaire

The data we use is public data obtained from the local health department

In Indonesia, stunting is one of the government's priority programs which is targeted to be completed in the next 5-10 years. By eliminating stunting, the vision of a golden Indonesia 2045 will be increasingly possible to achieve. Therefore, in the next 5-10 years we think this will be a new technological breakthrough that will accelerate the goal there

We consist of 2 main teams, namely technical and non-technical, totaling 5 people

We have been developing various technology solutions for the last 5 years, some of which are AI-based. And this is one of our plans in the health sector

TBA

Developing a Machine Learning Based Android Application Design with Education and Counseling Features for Preventing and Handling Stunting in Toddlers is a complex project and involves various aspects, including technical development, design, education and counseling. Below is a starter plan you can use as a starting point. Please keep in mind that this is only a general overview and the actual project will require more in-depth analysis and planning.

Project Model
1. Project Description
Building an Android application that uses Machine Learning to detect stunting in toddlers.
Provides educational features based on educational content about child nutrition and growth.
Provide access to counselors or health resources to help parents and caregivers of toddlers affected by stunting.
2. Project Objectives
Prevent and detect stunting in toddlers early.
Provide education and resources to parents and caregivers to ensure healthy growth and development of toddlers.
Operational Plan
1. Needs Analysis
Identify technical needs, such as programming language (perhaps Kotlin for Android), Machine Learning platform, and backend infrastructure.
Determine educational and counseling content needs, as well as health resources that will be involved.
2. Application Development
Determine the development methodology (e.g. Agile or Waterfall).
Divide tasks and development schedule.
3. Building a Machine Learning Model
Identify a dataset to train a stunting detection model.
Choose a suitable Machine Learning algorithm (for example, Convolutional Neural Network or Decision Trees).
Train and evaluate models using that data.
4. User Interface Design
Design the user interface (UI/UX) with the needs of users, including parents, caregivers, and counselors in mind.
Implement UI/UX design to the application.
5. Development of Education and Counseling Features
Develop educational content based on nutrition and toddler growth.
Determine how to access counselors or health resources, perhaps via chat or video call.
6. Integration of Machine Learning into Applications
Integrate Machine Learning models into Android applications.
7. Testing and Evaluation
Perform functional and non-functional testing (such as performance and security tests).
Evaluate Machine Learning models using separate test datasets.
8. Launch and Marketing
Plan the application launch strategy.
Do marketing to promote the app and get users.

1. Source of Income:
Freemium Model: The app can be downloaded and used for free with limited access. Premium features, such as live counseling, are available via monthly/annual subscription.
2. Marketing Strategy:
Online marketing campaigns via social media, blogs and websites.
Partnerships with educational institutions and child health organizations.
Referral program to increase the number of users.
3. Monetization Plan:
Subscribe to premium for full access to additional features.
Advertisements or promotions related to children's health.

1. Initial Development Costs:
Android App Development: $15,000
Machine Learning Model Development: $20,000
UI/UX Design: $5,000
2. Operational Costs:
Team Salary and Compensation: $40,000
Marketing Costs: $7,000
Server and Infrastructure Cost: $8,000
Testing and Evaluation Cost: $5,000
Administrative Fees: $3,000
3. Education and Counseling Costs:
Educational Content Cost: $4,000
Counseling/Consulting Fee: $10,000
4. Data Management and Security Fees:
Security and Compliance Fee: $3,000
5. Miscellaneous Expenses and Reserves:
Unexpected Expenses: $5,000
Total Estimated Operational Costs (over the next year):
$75,000

1. Initial Development Costs:
Android App Development: $15,000
Machine Learning Model Development: $20,000
UI/UX Design: $5,000
2. Operational Costs:
Team Salary and Compensation: $40,000
Marketing Costs: $7,000
Server and Infrastructure Cost: $8,000
Testing and Evaluation Cost: $5,000
Administrative Fees: $3,000
3. Education and Counseling Costs:
Educational Content Cost: $4,000
Counseling/Consulting Fee: $10,000
4. Data Management and Security Fees:
Security and Compliance Fee: $3,000
5. Miscellaneous Expenses and Reserves:
Unexpected Expenses: $5,000
Total Estimated Operational Costs (over the next year):
$75,000

I think it's more about initial funding, because the product being developed is 100% digital so other facilities are not a top priority although other things such as mentorship in the business sector are certainly needed