The Problem

Delayed responses to disease outbreaks in Nigeria cost lives. Public health systems rely too much on manual reports and outdated tools. Our team set out to build a tech-based early warning system using AI and everyday tools like WhatsApp, SMS, and USSD.

The Solution

We designed a multi-channel wireframe to detect outbreak trends, which included:

• SMS/USSD: For reporting symptoms by low-tech users.

• AI Bot: For monitoring WhatsApp health group chats.

• Dashboard: For visualizing real-time symptom trends for NCDC officials

My Role  (Data Analyst) 

As the data analyst, I translated symptom reports and user interview findings into structured datasets. I analyzed patterns to define which metrics would flag potential outbreaks and designed mock trend reports and dashboards to show how NCDC could monitor symptom spikes. I also worked closely with my team to ensure our wireframes and slides clearly communicated our insights. 

• I used Excel to simulate and clean SMS/WhatsApp-style health reports by standardizing symptoms and preparing datasets for dashboards. Integrated AI (ChatGPT) to generate realistic messy data and suggest Excel formulas (TRIM, PROPER, VLOOKUP) for faster cleaning.

• I used SQL [SELECT, WHERE, GROUP BY, SUM()]to analyze symptom data, highlighting common illnesses and geographic hotspots to simulate faster outbreak detection. 

• I used Python libraries, Pandas and Matplotlib, to clean and visualize a sample outbreak dataset. This streamline the analysis workflow and helped identify the most common symptoms reported.

• I explored how AI tools like ChatGPT and Gemini could automate repetitive cleaning and reporting tasks to make workflows more efficient.

Tools Used & Skills

Tools: Excel, SQL (DB-Fiddle), Python (Google Colab), Google Sheets, Canva, Google Slides, ChatGPT & Gemini.

Skills: Data Cleaning, Data Visualization, Trend Identification, User Behavior Analysis, Data Storytelling, Collaboration.

Key Outcome

Status: Completed June 2025

• We delivered a low-cost, AI-driven wireframe that simulated how health authorities could be alerted when symptom patterns rise in a location. This project serves as a scalable case study for early disease detection in underserved areas.

Problem Statement

In underserved Nigerian communities, limited access to clean water leads to health risks. This project uses structured frameworks and data tools to assess and improve water safety. 

Solution

This solution uses logic models and data tools to evaluate water quality and propose targeted interventions. 

Frameworks Used

• EGAD: Explain, Gather, Analyze, Deploy

• MECE Logic Tree

• Flowcharts & Pseudocode

My Role  

• Data Analyst

Tools Used & Skills

Tools: 

• Excel  

• SQL (DB-Fiddle)  

• Python (Google Colab)  

• draw.io (for flowchart and MECE diagram design)

• Power BI or Tableau (for dashboards) 

Skills: 

• Excel formulas & logic  

• Python automation  

• Problem framing  

• Data cleaning & decision modeling  

• Data visualization

Key Outcome

Status: Currently in development - dashboard and reporting tools in progress

• Water Safety Checker (Excel & Python)

• MECE Tree for clean water solutions

• Flowchart & pseudocode logic