Strengthening Frontline Capacity to Reduce Maternal & Neonatal Mortality
How ARMMAN deployed a WhatsApp-based AI assistant for Auxiliary Nurse-Midwives, delivering real-time protocol-aligned clinical guidance across UttarPradesh and Telangana. 123
Watch the story in 60 seconds (AI-generated video; visuals are illustrative)
Case at a Glance
The Story Subtitle 60 Second
ARMMAN is a nonprofit working to reduce preventable maternal and child mortality through a blended tech-plus-touch model integrated with government systems. Since 2008, they have reached 70 million+ women and children across 28 States.
High-risk pregnancies (20–30% of all pregnancies) cause 70–80% of maternal and neonatal deaths. Auxiliary Nurse-Midwives are overburdened, undertrained, and lack real-time clinical guidance — while digital fatigue from 6–8 government apps leaves no room to learn.
A multilingual, multimodal WhatsApp AI assistant for ANMs — providing protocol-aligned clinical guidance, doubt resolution, and bite-sized learning, supported by a human-in-the-loop escalation pathway for safety and accuracy.
A woman dies due to childbirth-related complications every 30 minutes in India. Over 19,000 women die annually — largely from preventable causes. High-risk pregnancies, which constitute 20–30% of all pregnancies, account for 70–80% of perinatal mortality and morbidity.
Early identification and effective management of these cases depends heavily on Auxiliary Nurse-Midwives (ANMs). But ANMs are overworked, inadequately trained, and often lack the skills and confidence required to detect and manage complications early.
Deployed across pilot districts, the AI clinical decision support tool has demonstrated significant improvements in ANM performance, referral quality, and early detection of high-risk pregnancies.
- WhatsApp-native access to instant, protocol-aligned clinical guidance — no app download required
- Multilingual and multimodal support — text and voice interactions via OpenAI Whisper + Sarvam TTS
- Human-in-the-loop escalation pathway to maintain safety, trust, and clinical reliability
- Bite-sized continuous learning embedded into day-to-day ANM workflows — not separate training events
- Designed to integrate with government antenatal care protocols for contextual accuracy
ARMMAN's AI clinical decision support tool empowers ANMs with real-time, evidence-based guidance — reducing cognitive burden, improving referral accuracy, and enabling earlier detection of high-risk pregnancies.
Technology Stack
| Tool | Used For | What It Enabled | Category |
|---|---|---|---|
| Turn.io ChatbotConversational platform | ANM-facing WhatsApp interface | AI-driven conversational support for learning and on-the-job clinical guidance via WhatsApp | Commercial |
| ChatGPT-4.oOpenAI | Clinical guidance & learning content | Natural language understanding, response generation, and structured text formatting grounded in medical protocols | Commercial |
| OpenAI WhisperSpeech recognition | Voice-based interactions | Speech-to-text transcription enabling ANMs to ask clinical questions by audio, not just text | Commercial |
| SarvamIndian language AI | Multilingual audio responses | Text-to-speech generation for multilingual audio responses in regional languages | Commercial |
Key Project Learnings
ANMs lacked timely guidance while managing high-risk pregnancies. Embedding protocol-aligned support within WhatsApp enabled real-time assistance without disrupting existing workflows — zero new apps required.
Pure automation was insufficient for complex clinical scenarios. A human-in-the-loop escalation model ensured safety while allowing the solution to scale — maintaining trust at every step.
Classroom trainings were infrequent and workers faced digital fatigue. Bite-sized learning integrated into routine interactions enabled continuous capacity building without adding burden.
Potential Application Across Sector
| Sector / Domain | How this solution adapts |
|---|---|
| Frontline health services | Any setting where frontline workers need real-time, protocol-based guidance at the point of service — ASHA workers, CHOs, community paramedics. |
| Education & teacher support | Teachers in low-resource schools facing large classes and minimal coaching infrastructure can receive on-demand, curriculum-aligned pedagogical guidance. |
| Agricultural extension | Agricultural extension workers needing real-time crop advisory, pest identification, or weather-informed guidance during field visits. |
See it in Action
A walkthrough of the WhatsApp-based clinical decision support tool and its integration with ANM workflows.
A frontline health worker describes how the AI assistant changed her ability to manage high-risk pregnancies on the spot.
A health worker shares how bite-sized learning through WhatsApp reduced her dependency on refresher trainings.
Related Case Studies
View all case studies
A courtroom operating system automated transcription and documentation, significantly speeding case processing and improving judicial efficiency.
AI platform scaled public consultations nationwide, enabling governments to incorporate citizen input into stronger, more inclusive policies.
Hybrid tech stack, AI/GIS, enables scalable, verifiable agroforestry monitoring and impact.