Our Connected Care System

The gap we are starting from

Important clinical information often emerges in care before it becomes durable, measurable, or easy to build on. That creates a gap not only for continuity in treatment, but also for personalization and for learning from care over time.

The result is a fragmented learning loop and fragmented care. The same system is being asked to support continuity in care, track meaningful progress, personalize what happens next, reduce clinician burden, and produce evidence about what works in the real world, but those responsibilities rarely live in one connected loop, and are instead split across different tools, workflows, and moments in care. What is missing is infrastructure that can turn everyday care into something more measurable, more actionable, and more learnable over time.

What this system is, and what needs validation

Feelpath is designed to get more value out of existing care, by making sessions easier to learn from and easier to carry forward.

We are building a connected system for longitudinal, session-based measurement, personalized between-session support, clinician workflow tools, and research-ready infrastructure.

Feelpath is built around three connected layers. One layer is the transcript-derived measurement layer: whether conversation-based markers can capture patterns that are interpretable, clinically meaningful, and worth validating against stronger reference points. A second layer is the client-facing adjunctive layer: whether tools like wheels, annotation, and session-based review actually help clients reflect, label emotions more clearly, and carry the work forward between sessions. A third layer is the therapist workflow layer: whether these tools can fit into ordinary practice in a way that supports follow-up, recall, and continuity without creating too much friction.

That is why the next phase is not simply to prove the product. The task now is more specific: to study whether the measurement layer is valid, whether the client tools are helpful, and whether the whole system fits real care well enough to deserve broader use. That is the transition Feelpath is trying to make now, from a working integrated system to a validated approach.

What we’ve already done

Here is a visual overview of our progression timeline. It shows what is already in place, what has already been used, and where the work goes next.

We have already built a real product system around therapy sessions: session and transcript infrastructure, transcript-derived analysis, client-facing emotional learning tools, therapist-facing review tools, and the first layer of research work around them. What exists today is not just session capture, but a set of connected tools for seeing more from the session, carrying the work forward between sessions, and making that work more visible to both clinicians and clients.

Parts of this system have also already been used across more than 300 peer group sessions, giving us an early setting to see how these tools support reflection, expression, and follow-through in real conversations over time. On the client side, that already includes tools like emotion wheels, emotion annotation, and session-based review.

On the therapist side, it already includes emotional review, emotion annotation, emotion analytics, and other session insight views. The next phase is to study this well, support real use in practice, and move from a working integrated system to a validated approach.

What needs to happen next

The immediate next step is not a large formal validation trial. It is a first feasibility study in routine care. We need enough real sessions on the platform to learn whether the tools are usable, helpful, and workable for clients and therapists in practice.

That is the main bottleneck right now. We already have substantial interest, but recruitment, participant compensation, design support, and research support are still practical constraints. Without enough supported real-world use, the learning loop slows down before we can gather the feedback and data needed to improve the system.

• run real therapy sessions on the platform with compensated participants
• work with therapist design partners in routine care
• collect therapist and client feedback on usefulness and workflow fit
• gather usage and engagement data from real sessions and between-session use
• track early pre/post signals where appropriate
• use those learnings to refine the product and prepare later ALI and ELI studies