As healthcare pushes toward faster, more accessible decision-making, diagnostics remains one of its most critical bottlenecks. While much of the industry conversation is dominated by AI, Ashissh Raichura, Founder and CEO of Scanbo, argues that the real challenge lies deeper — in the quality, consistency, and accessibility of data itself. In this conversation, he unpacks why point-of-care diagnostics must evolve beyond fragmented devices into reliable, multi-parameter platforms, and why building robust hardware and data pipelines is a prerequisite before AI can truly deliver value.
Q. Scanbo is working at the intersection of hardware, diagnostics, and software. What problem were you trying to solve when you started?
If you look at how diagnostics actually works on the ground, especially outside large hospitals, the system is still heavily dependent on central labs. Even for basic screening, there’s a delay between testing, reporting, and decision-making.
That delay is not just inconvenient — it affects outcomes. In primary care or semi-urban settings, the first interaction often determines whether a condition is escalated or ignored.
We started Scanbo with a simple question:
Can the first level of diagnostic insight happen instantly, at the point of care, without depending on a lab ecosystem?
That required rethinking diagnostics not as a service, but as a product that can travel to the patient.
Q. Point-of-care diagnostics is not a new category. Where do you see the gap today?
The category exists, but most solutions are still single-parameter, fragmented, or workflow-heavy.
A doctor or a health worker doesn’t want to manage five different devices for five parameters. Nor do they want to deal with calibration issues, inconsistent readings, or complex operating protocols.
The real gap is not access to devices — it is reliability and usability at scale.
If a device cannot deliver consistent results across different environments, operators, and patient conditions, it doesn’t matter how advanced the technology is.
That’s where most point-of-care systems break down today.
Q. You are building a multi-parameter platform instead of a single-use device. Why take that route?
Because that’s how clinical decision-making actually works.
A single data point rarely tells you anything meaningful. Whether it’s glucose, cardiac signals, or vitals — interpretation happens in context.
We are building Scanbo as a consolidated diagnostic layer, where multiple parameters can be assessed through one platform, rather than isolated devices.
This reduces friction at three levels:
l Operationally — fewer devices to manage
l Economically — better utilisation per device
l Clinically — more complete inputs for decision-making
It’s a harder problem to solve, but a more relevant one.
Q. Non-invasive diagnostics, especially for glucose, has seen many attempts globally. What makes this technically challenging?
The biggest challenge is signal integrity.
In invasive testing, you are directly measuring from a biological sample. In non-invasive systems, you are interpreting indirect signals — optical, electrical, or physiological.
Those signals are highly sensitive to:
l Skin variations
l Environmental conditions
l Device positioning
l Patient movement
So the challenge is not just measurement — it’s filtering noise and ensuring repeatability.
That’s why many solutions show promise in controlled environments but struggle in real-world deployment.
Q. There is a lot of conversation around AI in diagnostics. How do you see its role in what you are building?
AI is useful, but it is often misunderstood in this space.
Most conversations assume that AI will compensate for everything. In reality, AI is only as good as the data it receives.
If your input data is inconsistent, poorly calibrated, or noisy, AI will amplify those errors rather than fix them.
Our approach has been to focus first on:
l Device-level accuracy
l Calibration stability
l Structured data capture
Only once that layer is reliable does AI become meaningful.
So for us, AI is not the starting point — it’s the second layer built on top of dependable hardware and data pipelines.
Q. How do you think about scalability, especially in a country like India with diverse healthcare settings?
Scalability in healthcare is often misunderstood as distribution.
In reality, it depends on whether your system can function consistently across:
l Urban clinics
l Tier 2 and Tier 3 setups
l Remote or resource-constrained environments
For that, three things matter:
Ease of use — minimal training required
Device robustness — stable across conditions
Workflow simplicity — integrates into existing practice
If a solution requires behavioural change from doctors or staff, scaling becomes difficult.
We’ve designed Scanbo to fit into existing workflows rather than asking the system to adapt to us.
Q. Where do you see the biggest use cases emerging for point-of-care diagnostics over the next few years?
The strongest use cases are where time-to-decision matters the most.
That includes:
l Primary care screening
l Chronic disease monitoring
l First-level triage in outpatient settings
l Remote or home-based care
In all these scenarios, the question is not just “What is the diagnosis?” but
“How quickly can we move to the next step?”
That’s where decentralised diagnostics becomes critical.
Q. What has been the most difficult part of building Scanbo so far?
Hardware-led healthcare is inherently complex.
Unlike software, you are dealing with:
l Regulatory pathways
l Clinical validation
l Manufacturing precision
l Long feedback cycles
You can’t iterate overnight. Every improvement has to be validated thoroughly.
The challenge is balancing speed of innovation with reliability, because in healthcare, errors have real consequences.
Q. Looking ahead, how do you see diagnostics evolving as a category?
Diagnostics will move closer to the patient — that shift is already underway.
But the real change will be in how diagnostics is used.
Instead of being a separate step, it will become part of a continuous feedback loop:
l Screening
l Monitoring
l Intervention
The boundary between diagnostics and care will start to blur.
And in that system, the companies that solve for accuracy, accessibility, and usability together will define the next phase.