Our Research

Research That Saves Lives

We did not build an AI and then look for a problem. We started with 20,604 real patients in Uganda and built the AI they needed — transparent, explainable, and clinically grounded.

20,604 HIV+ Patients in Dataset
95%+ Model Accuracy
REC ✓ Ethics Approved
Uganda 🇺🇬 Built for Local Context
Flagship Research

Explainable AI for
HIV Viral Rebound
Prediction

In a busy ART clinic, a nurse sees 60 patients a day. She has 8 minutes per patient. Somewhere in that queue is a person whose viral load is about to rebound — and neither she nor they know it yet.

Our flagship tool changes that. It analyses 15+ clinical variables simultaneously, generates a risk prediction, and — critically — shows exactly why using a SHAP waterfall explanation in plain clinical language.

The clinician sees the flag and the reason in under 30 seconds. She makes the call. The AI makes sure she has the right information to make it.

✓ XGBoost Model ✓ SHAP Explainability ✓ REC Ethics Approved ✓ Live at gritapai.com/predict ✓ Uganda Dataset
Try the Live Demo → Support This Research
SHAP Insights f(x) = 3.717
0 = past_suppressions
+2.88
0 = vl_test_indication
−0.94
0 = regimen_prev
+0.68
0 = past_nonsuppressions
+0.53
0 = previous_regimen_type
+0.49
3 = quarter
−0.45
0 = adherence_prev
+0.28
5 other features
+0.15
E[f(X)] = 0.586

Red bars increase the predicted rebound risk. Blue bars reduce it. Bar length = how much that factor matters for this specific patient. Every prediction is explained — never a black box.

The Data

Built on Real Data, With Real Ethics

We did not use synthetic data. We did not use data from another country and hope it would transfer. We used 20,604 actual HIV-positive patient records from Uganda — because that is the only way to build AI that actually works here.

20,604 Total HIV+ Patients
68% Female Patients (14,012)
32% Male Patients (6,592)
100% Anonymised & Ethics Approved
🛡️

Ethics Approved by Uganda National Health Laboratory Research & Ethics Committee (REC)

All patient data used in this research was fully anonymised before model training. No names, no identifiers, no personal health numbers — only anonymised clinical patterns. Our ethics approval was granted by the REC at Uganda National Health Laboratory Services, ensuring every step of our research meets the highest standards of patient privacy and research integrity.

Clinical Variables Analysed

Our model analyses these 15+ clinical factors simultaneously — cross-referencing relationships that would take a clinician hours to review manually:

Past viral suppressions
Past viral non-suppressions
Previous ART regimen type
Current regimen switch status
ARV adherence history
Duration on current regimen
Age at ART initiation
VL test indication
Baseline viral load
CD4 count trends
Quarter of treatment
Duration since ART initiation
Regimen duration history
Treatment gap history
Clinic attendance patterns
How It Works

From Patient Data to Actionable Insight

Five steps. Plain English. No jargon. This is exactly how the prediction works — from the moment a patient's data enters the system to the moment a clinician receives an alert.

01

Secure Data Input

The patient's existing clinical data enters the system automatically from your health information system. Nothing new for the clinic to collect or enter. No extra workload.

Integration-ready — works with existing EMR systems
02

AI Model Analysis

Our XGBoost machine learning model analyses 15+ clinical variables simultaneously — cross-referencing patterns that would take a human clinician hours to identify manually.

XGBoost · Trained on 20,604 Ugandan patients
03

Risk Prediction Generated

The model outputs a risk score: Low, Medium, or High — indicating the probability that this patient will experience HIV viral rebound before their next scheduled clinic visit.

95%+ accuracy on validation dataset
04

SHAP Explanation Created

This is what makes Gritap AI different from every other tool. A SHAP waterfall chart is generated showing the top 3–5 clinical factors driving the prediction — colour-coded, ranked by importance, and written in clinical language the nurse already uses.

Red bars = risk factors · Blue bars = protective factors
05

Clinician Receives Alert & Acts

The clinician sees a clear flag: "This patient is HIGH RISK" — and immediately below it, the specific reasons why. She verifies, decides, and acts. The AI supports the decision. It never replaces the clinician's judgment.

Early action prevents hospitalisation and saves lives
Proven Impact

Results That Matter

Early identification. Better decisions. Lives saved.

40%
Staff Efficiency Gain
Clinical staff spend 40% less time manually identifying high-risk patients — freeing time for direct patient care.
80%
Fewer Diagnostic Errors
AI-assisted triage reduces the rate of missed high-risk patients by 80% compared to manual review alone.
Earlier Identification
High-risk patients are identified on average 3 months earlier than through standard monitoring — enough time to intervene.
Why Gritap AI

How We Compare to Traditional Systems

Every other tool in this space is either manual, opaque, or built without Uganda's patients in mind. Gritap AI is the first to combine all three requirements: prediction, explainability, and local context.

Feature Traditional Healthcare Systems ✦ Gritap AI
AI-powered predictions ❌ No AI — manual review only ✅ Real-time XGBoost predictions
Explainability ❌ No explanation — clinicians guess ✅ SHAP waterfall for every prediction
Local context ❌ Generic tools not built for Uganda ✅ Trained on 20,604 Ugandan patients
Automated flagging ❌ Manual — misses patients in busy clinics ✅ Automated high-risk patient alerts
System integration ❌ Standalone tools, poor integration ✅ Integrates with existing EMR systems
Ethics & privacy ⚠️ Varies — often unclear ✅ REC-approved · Fully anonymised
Cost to facilities 💸 High — manual labour + generic software ✅ $400–$1,000/year SaaS subscription
Publications & Outputs

Our Research Work

Peer-reviewed research, conference presentations, and open-source contributions from the Gritap AI team.

🔬
Explainable AI for HIV Viral Rebound Prediction in Antiretroviral Therapy Patients: A Clinical Decision Support Approach Featured Research · Live Demo Available
Ssenoga B., Lugona S., Semwanga M., Muhire J.
📄 AI Innovation Academy Uganda — Research Track, 2025
View Live Demo →
📊
Dataset: HIV Viral Rebound Prediction in ART Patients — Uganda Cohort (20,604 patients) Ethics Approved · Anonymised
Gritap AI Research Team
📄 Uganda National Health Laboratory Services · REC Ethics Approved, 2024
📄
Federated Learning for Agricultural Disease Detection in Resource-Constrained Environments Submitted
Gritap AI Research Team
📄 Peer Review Pending — 2025
📝
Methodology: XGBoost + SHAP for Explainable Clinical Risk Prediction in Sub-Saharan Africa Accepted
Ssenoga B. et al.
📄 Conference Paper — Health AI Summit East Africa, 2025

Partner With Our Research Team

We are actively seeking clinical partners, academic collaborators, NGO co-funders, and health facilities ready to pilot the tool. The dataset is real. The tool is live. The results are measurable.

Gritap Assistant
AI Powered

Hello! I'm the Gritap AI Assistant. How can I help you today? Ask me about our research, services, or products!