Manta Voyager is TRL6 autonomous AI-powered medical decision support system designed for first responders operating in degraded conditions with limited medical expertise, connectivity, or time.
The solution primarily addresses Capability Gap 9, by transforming raw field data into structured medical decisions that responders can immediately act upon, while also supporting CG4, through standardized capture and sharing of patient data.
Manta Voyager was co-designed with operational partners representing distinct links of the emergency medical response chain, ensuring the solution directly addresses real field constraints.
These partners include SDIS 38 (fire and rescue services), the French Armed Forces Health Service (SSA), the Maritime Medical Consultation Center (CCMM), CNES, ESA, Airbus Defence and Space, and Thales Alenia Space, as well as innovation partners working in low-connectivity environments.
Each partner contributed insights from their position in the operational chain : from on-scene triage and medical reporting, to remote medical support and satellite communication constraints, ensuring the system addresses the full workflow behind actionable intelligence.
In many emergencies, responders must treat victims, perform triage and document the intervention simultaneously, often without a physician on site and with unstable communications.
This increases cognitive load, delays triage and leads to incomplete reports. Manta Voyager reduces this burden through three capabilities:
Voice-to-Report: responders describe the situation verbally while treating the victim; the system generates a structured Patient Care Report in under 90 seconds.
Automated triage support: doctrine-based prioritisation helps manage multiple victims.
Procedural medical guidance: step-by-step assistance supports responders in unfamiliar medical situations.
The device operates offline-first, performs local AI analysis, and supports voice, image, biomedical input.
Contact
The objective of Manta Voyager is to strengthen the operational capacity of first responders by providing autonomous medical decision support in environments where medical expertise, connectivity, or time are limited. Every day, firefighters and emergency responders across Europe and beyond must assess victims, perform triage and document medical interventions under intense operational pressure. In many situations, responders operate without a physician on site and must manage both patient care and reporting simultaneously, increasing cognitive load and the risk of information loss. Manta Voyager addresses these challenges by enabling responders to capture, analyse and structure patient information directly on scene using an autonomous edge device. Through voice interaction and multimodal inputs, the system converts raw field observations into structured medical reports, triage prioritisation and operational guidance within seconds. The ambition of the project is to turn fragmented field data into actionable intelligence, allowing responders to focus on the victim while maintaining the integrity of the medical response chain. Beyond daily emergency response, the solution is designed to support mass casualty incidents (NoVi), natural disasters linked to climate instability, and medical distress situations occurring during geopolitical crises or conflicts, where medical resources and communications may be severely degraded. This dual-use capability also addresses medical support needs for defence operations and extreme environments. The same technological foundations are being explored for future human space exploration missions, where medical autonomy is critical. By combining offline-first AI, edge computing and responder-centered design, Manta Voyager aims to strengthen the resilience of medical response systems across European and international emergency response ecosystems, spanning civilian, defence and exploration contexts.
The innovation of Manta Voyager lies first in its approach. The system was designed by analysing the full operational chain behind Capability Gap 9 – Actionable Intelligence, not only first responders but also actors involved in remote medical support, defence operations, maritime medicine, satellite communications and space assets. This ecosystem perspective ensures the solution addresses real operational constraints across the entire information chain. Second, it introduces an innovation in vision. Modern emergency response systems rely heavily on connectivity, cloud infrastructure and remote medical expertise. However, communications may fail, cloud infrastructures are not always sovereign, and physicians are rarely present on scene. Manta Voyager therefore removes these dependencies through edge AI, autonomous local processing and offline-first operation. Third, the solution integrates several technological innovations within a single architecture. The system combines two biomedical LLMs, medical decision trees and a retrieval-augmented generation (RAG) system to update operational doctrines. A vision-language model (VLM) enables contextual analysis of images and videos, while multilingual speech-to-text allows fully hands-free interaction. All capabilities are integrated into a dedicated responder interface co-designed with emergency services to minimise cognitive load. Structured reports can be transmitted using store-and-forward communication through various modalities. Developped with AI cluster in Grenoble (FR) and Roma (IT). Finally, the innovation is validated through extreme-environment field testing, including Arctic, Antarctic, desert and cave environments, ensuring the technology performs in the same conditions where responders may operate. To our knowledge, no existing system combines autonomous medical AI, multimodal data integration and resilient communications in a single portable device designed specifically for first responders.
Our solution, Manta by OneTreck, addresses CG9 – Actionable Intelligence, which concerns the ability to transform raw operational information into clear and usable decisions for responders in the field. During emergency interventions, critical medical information is generated across several steps of the response chain: victim assessment, triage, medical reporting, evacuation coordination and hospital preparation. In reality, these steps are often fragmented. Responders must treat victims while documenting the intervention, manage multiple casualties under time pressure, and transmit information through unreliable communications. These constraints create major operational challenges: high cognitive load, delayed triage, incomplete patient reports, and loss of critical medical information. When communications fail or bandwidth is limited, the medical chain can rapidly degrade, affecting coordination between field teams, remote medical support and receiving hospitals. Manta acts as an operational support tool that converts field observations into actionable intelligence in real time. For the responder, this means three concrete capabilities: • Voice-to-report: responders simply describe the situation while treating the victim. Manta automatically generates a structured Patient Care Report in less than 90 seconds. • Automated triage support: the system applies operational doctrines to help prioritise victims during multi-casualty situations and digitally tag patients to facilitate evacuation, handover and hospital care. • Procedural guidance: step-by-step assistance helps responders manage unfamiliar or complex medical situations. By structuring information directly on scene, Manta reduces documentation burden, supports faster triage and preserves the continuity of the medical response chain, even when communications are degraded. Manta allows responders to focus on victim care and maintain the integrity of the medical information chain.
Interdisciplinary team combining expertise in AI, edge computing, emergency medicine and operational deployment. The team includes engineers, emergency physicians and SDIS first responders, with collaboration from the French Armed Forces Health Service, ensuring the solution is designed, validated and aligned with real operational needs of emergency responders.
Interdisciplinary team combining expertise in AI, edge computing, emergency medicine and operational deployment. The team includes engineers, emergency physicians and SDIS first responders, with collaboration from the French Armed Forces Health Service, ensuring the solution is designed, validated and aligned with real operational needs of emergency responders.