GPS for Nerves: How Nerve Trace Is Turning Fluorescent Imaging Into AI Guided Anatomical Navigation

Surgeons face a constant challenge: identifying and preserving nerves during complex procedures. This issue is not a reflection of a surgeon's skill or expertise, but rather a limitation by not having the adequate tools. Unlike blood vessels or organs, nerves:

• Do not have clear visual contrast

• Can vary significantly in location from patient to patient

• Are often buried and invisible to the naked eye and imaging technologies

Thus, even highly skilled surgeons must rely on experience and estimation, creating a persistent risk of accidental nerve damage.

Damaging nerves can lead to serious complications, including loss of function or chronic pain. Traditional imaging methods often fall short in providing clear, real-time nerve visualization. However, with Nerve Trace this limitation is fundamentally addressed.

For over a decade, the team behind Trace Biosciences has developed well established science and technology to create safe nerve-specific clinical fluorescent imaging agents to turn nerves from "hidden" structures into clearly identifiable anatomy in real time during fluorescence-guided surgery.

Beyond Imaging: Mapping The Future of Surgery Using Artificial Intelligence

Enabling real-time nerve visualization with Nerve Trace is only the beginning. Trace Biosciences is now advancing into artificial intelligence (AI) driven systems designed for anatomical mapping and guidance to further expand surgical insight and precision, redefining how surgeons and interventionalists navigate the intricate nerve framework in the body.

Critically, Nerve Trace does not just improve visualization, it creates ground truth data. By making nerves directly visible in real time, it enables the generation of high-quality labeled datasets that AI systems can learn from with unprecedented accuracy.

This foundation is being validated at a national level through a $31.3M ARPA-H program focused on mapping nerves and vasculature in 3D using dual-color imaging, where Nerve Trace agents play a central role. To learn more about the program see here. This initiative represents a clear signal that the future of surgery and medicine is moving toward comprehensive, data-driven anatomical mapping and guidance systems.

To understand how Nerve Trace enables the next generation of AI-guided procedures, it’s important to first examine the technologies shaping the field today. Below is an overview of the three primary categories of AI-driven nerve guidance systems.

1. Automated Nerve Segmentation for Digital Anatomical Maps

AI segmentation for anatomical mapping uses deep learning algorithms to automatically identify, outline, and label specific organs, tissues, or bones in medical images (CT, MRI, ultrasound) or intraoperative video to assist in diagnostics, surgical planning, and 3D printing.

The benefits:

• Enhanced preoperative planning with personalized surgical models (read here)

• Improved real-time surgical navigation (read here)

• Reduce operative time while increasing accuracy and safety (read here)

The limitations:

• Mono-modality systems suffer from tracking instabilities (read here)

• Automation bias and workflow disruption (read here)

• High costs and integration hurdles (read here)

Already, AI segmentation is advancing rapidly, but its performance is fundamentally limited by the quality of training data. Most current systems rely on indirect labeling from CT, MRI, or manual annotation.

Nerve Trace changes this paradigm. By directly illuminating nerves intraoperatively, it provides pixel-level ground truth labels for training AI models. This enables:

• Highly accurate nerve detection models

• Real-time segmentation with improved robustness

• Continuous learning systems that improve with every procedure

With fluorescent imaging data as the training backbone, AI systems can automatically segment nerves from background tissue with far greater reliability than traditional approaches. Surgeons can then use these maps to plan incisions, guide dissection, and avoid nerve damage with a level of precision not previously possible.

2. 3D Reconstruction of Patient-Specific Nerve Networks

3D reconstruction of patient-specific nerve networks utilize preoperative noninvasive imaging (MRI/CT) data and its segmented structures to convert them into personalized 3D digital anatomic models.

The benefits:

• Improved surgical planning and accuracy (read here)

• Mimic exact geometry of a damaged area (read here)

• Enhance communication and education (read here)

The limitations:

• Deep learning models are required and are limited (read here)

• Unable to reflect dynamic processes such as tissue deformation (read here)

• High expense of software, equipment, and expertise (read here)

Today, these models are primarily used for preoperative planning and coregistered intraoperative guidance. However, when powered by Nerve Trace-derived datasets, they become far more powerful. Particularly in addressing one of the most fundamental challenges in surgery: the fact that anatomy is not static.

During procedures, tissues are constantly shifting, deforming, and being displaced due to patient positioning, insufflation, retraction, and surgical manipulation. This makes it difficult for conventional models, built on preoperative imaging, to remain accurate in real time.

Instead of approximating nerve locations, these models can be built from directly observed nerve anatomy and continuously updated, enabling:

• True patient-specific nerve mapping

• High-fidelity reconstruction of nerve networks with accurate mapping to deformed and dynamically moving tissues

• Real-time registration and alignment between preoperative models and intraoperative anatomy

• Integration into AR/VR and robotic systems with persistent anatomical accuracy despite tissue shift

This evolution transforms 3D models from static approximations into adaptive, living anatomical maps grounded in real biological signal and capable of maintaining accuracy even as the surgical field changes in real time.

3. Population-Scale Anatomical Mapping and Variation Modeling

As AI continues to reshape healthcare, one of the most promising frontiers is population-scale anatomical mapping by using AI to analyze and model human anatomy across large and diverse patient populations.

The benefits:

• Enhanced disease modeling and analysis (read here)

• Improved surgical precision and reduced complications (read here)

• Targeted bioelectronic therapies (read here)

  
  

The limitations:

• Lack of functional and dynamic data (read here)

• Study population bias as it relies on older cadaveric specimen (read here)

• Significant anatomical variability (read here)

Nerves in particular, are highly variable between patients.

Small differences in nerve location, branching patterns, or tissue depth can significantly impact surgical outcomes, resulting in nerve injury, chronic pain, loss of function, and long-term disability for millions of patients worldwide. Because of this issue, training AI systems on large datasets of nerves can assist in creating predictive models that recognize both common and rare nerve patterns across diverse patient populations.

Researchers are beginning to address this challenge through approaches such as immunostaining for body-wide mapping of sympathetic nerves (e.g., a 2025 study published in Cell) and initiatives like the NIH SPARC program to create the first 3D human vagus nerve anatomical map. However, these methods are largely limited to static, ex vivo datasets and often lack real-time, functional, or intraoperative relevance.

Nerve Trace fundamentally expands this paradigm by enabling not only anatomical mapping, but also functional insight through myelin-specific contrast. Because signal intensity and distribution are directly tied to myelin integrity, Nerve Trace can:

• Differentiate healthy, intact nerves from damaged, degenerated, or non-functioning nerves

• Provide contrast that reflects functional status, not just anatomical presence

• Enable accurate, representative population level nerve anatomy datasets

• Capture subtle changes in nerve structure that are invisible to conventional imaging

This allows the creation of datasets that go beyond “where nerves are” to include “how nerves are functioning.”

When aggregated across surgical procedures, Nerve Trace can power AI systems that:

• Predict nerve location before incision

• Adapt guidance in real time

• Account for dynamic and individual anatomical variations

• Enable autonomous surgical robotics for partial or total assistance

This represents a shift from static atlases to dynamic, functional, and predictive models of human neuroanatomy, unlocking a deeper understanding of both nerve structure and nerve health in clinical decision-making.

Beyond Surgery: Enabling a New Era of AI-Guided, Non-Invasive and Precision Nerve Therapies

While improving surgical safety is the immediate application, the long-term impact of Nerve Trace combined with AI extends far beyond the operating room.

By establishing a high-resolution, ground truth map of the human nervous system, this platform enables entirely new categories of intervention:

• Incisionless and minimally invasive procedures guided by AI-driven anatomical maps

• Targeted nerve ablation therapies for chronic pain and disease modulation

• Precision nerve stimulation and bioelectronic medicine

• Robotic and image-guided nerve blocks with millimeter-level accuracy

• Closed-loop therapeutic systems that sense, map, and modulate neural pathways in real time

These fields are rapidly expanding as core treatment modalities across medicine. Advances in device technology and growing clinical validation are driving adoption of neuromodulation and noninvasive procedures in areas like chronic pain, neurological disorders, inflammation, and organ dysfunction. Increasingly, these approaches offer targeted, lower-risk alternatives to open surgery, pointing toward a future where many conditions are treated through precise, image-guided, and incisionless interventions.

However, today many nerve-targeted or noninvasive therapies, such as ablation, nerve blocks, and neuromodulation, are limited by incomplete visualization and anatomical uncertainty. Clinicians often rely on indirect landmarks, electrical feedback, or imaging modalities that lack specificity for nerves. This creates variability in outcomes, risk of irreversible nerve damage, and limited broader adoption of these powerful treatments.

In this paradigm, AI systems trained on Nerve Trace data act as a navigation and decision making layer for the nervous system, enabling clinicians, and eventually autonomous systems, to interact with nerves safely and precisely without relying on open surgery.

This is the foundation for a new field: AI-guided neural intervention, where treatment is planned, guided, and executed based on a continuously improving understanding of human nerve anatomy.

Overview - Enhancing Surgical Safety with AI and Nerve Trace Imaging

Nerve Trace fluorescent imaging agents have been developed and validated over more than a decade to safely highlight nerves during surgery. Their proven safety and effectiveness provide a foundation not just for better visualization—but for building the ground truth data infrastructure that will power the future of AI-guided medicine.

By combining Nerve Trace imaging with AI-driven anatomical mapping:

• Surgeons gain real-time, actionable insight into nerve anatomy and function

• AI systems are trained on high-quality, in vivo ground truth datasets

• Robotic and navigation platforms become more precise and reliable, even in dynamic surgical environments

• Emerging fields like neuromodulation, nerve ablation, and minimally invasive interventions can be guided with far greater accuracy

With continued advances, and major initiatives like the $30.3M ARPA-H program accelerating this direction, the vision is clear:

A world where nerve anatomy is no longer hidden or estimated, but fully mapped, understood, and navigated in real time. Where surgery evolves from reactive to predictive. And where the nervous system becomes not just something to avoid, but something we can precisely see, understand, and treat.