Team BODMAS stands as a shining example of the new generation of student innovators driving India’s AI entrepreneurship revolution. Representing Kamla Nehru Institute of Technology, this team brought a transformative idea to life at HackWithUttarPradesh, India’s first Agentic AI Founders Hackathon, hosted at Chandigarh University Lucknow Campus and organized by the Technology and Business Incubator (TBI), Chandigarh University Mohali.
Their innovation — the Advanced Ambient AI Medical Scribe — impressed judges and mentors alike for its ability to blend deep AI innovation with human-centered healthcare design. Where most health tech projects focus on patient outcomes, BODMAS looked at a problem affecting the very core of healthcare delivery — clinician burnout and inefficiency.
What are they building
The Advanced Ambient AI Medical Scribe is an intelligent, always-listening digital assistant that transforms the way doctors document their interactions with patients. Instead of manually typing notes or relying on basic transcription tools, clinicians can now focus entirely on patient care while the AI automatically listens, understands, and documents conversations in real time.
Built using next-generation LLMs, Whisper-based ASR models, and emotion-aware computing, this AI scribe doesn’t just transcribe speech — it comprehends medical context, differentiates multiple speakers, and updates patient records directly through FHIR/HL7 integration.
Key Highlights:
Multi-party Conversation Capture: Accurately distinguishes between doctors, patients, and caregivers — even in noisy clinic environments.
Noise-Resilient Speech Recognition: Fine-tuned for medical terms, accents, and real-world settings.
Seamless EHR Integration: Directly syncs with Electronic Health Records (EHRs) using FHIR/HL7 APIs, reducing manual data entry.
AI Clinical Decision Support: Suggests diagnostic insights and reminders during consultations.
Emotion & Empathy Recognition: Detects tone and emotion to make AI communication more human-like and compassionate.
With context-aware documentation, real-time coding, and after-visit summaries, this solution ensures doctors spend less time on paperwork and more time healing lives.
Tech Stack
Frontend: React.js, Tailwind CSS, ShadCN UI
Backend: Node.js, FastAPI, PostgreSQL
AI/ML: Whisper/Faster-Whisper (ASR), Neural Diarization, spaCy, LangChain, RAG, VectorDB
Integration: FHIR/HL7 APIs, SMART on FHIR
Cloud & Deployment: AWS, Docker, Kubernetes, PyTorch Lightning, MLflow
This sophisticated stack enables low-latency real-time processing, scalability, and HIPAA-compliant data security, ensuring clinical-grade reliability.
Why are they building or to solve what?
In India and worldwide, doctors spend over 35% of their time on documentation, leading to fatigue and burnout. The BODMAS team saw how this administrative overload impacts both clinician well-being and patient experience.
Their goal: to restore time back to care.
Through AI-driven automation, the Medical Scribe reduces documentation time by up to 2 hours per day, minimizes manual entry errors, and ensures compliance with medical coding standards (ICD-10/CPT). By integrating affective computing, it even recognizes emotional tone — a subtle yet crucial step toward empathetic AI systems in healthcare.
Mentors at HackWithUttarPradesh applauded the project’s real-world relevance and technical depth, comparing it favorably to winning projects like OncoBind AI for its blend of innovation, accuracy, and societal impact.
Scope
The Advanced Ambient AI Medical Scribe isn’t just a hackathon prototype — it’s a vision for the future of healthcare automation. By combining Agentic AI reasoning with ambient intelligence, the system is evolving into a platform that could soon assist hospitals, telemedicine providers, and digital health startups.
Future Roadmap:
Integrating multilingual speech support for Indian healthcare diversity.
Adding predictive insights for early diagnosis and treatment recommendations.
Deploying on secure, federated AI networks for large-scale healthcare adoption.
Get Event Updates
For more information, check out our Privacy Policy and Terms of Service. You can unsubscribe at any time.
© 2025 CUBE. All rights reserved | Cookie preferences

