Diagnostics & Medical Devices

LESSON 01

Diagnostics & Medical Devices

Types of Diagnostics: Labs, Imaging, Point-of-Care, and Digital Health

The word 'diagnostic' covers four fundamentally different businesses with different regulatory paths, cost structures, and buyer relationships.

12 min read

Diagnostics is not a single market. It is a collection of distinct categories that share the goal of generating clinical information but differ in how they are built, cleared, sold, and paid for. A founder who treats laboratory diagnostics, imaging, point-of-care testing, and digital health tools as interchangeable will build the wrong regulatory strategy, target the wrong buyer, and price the product incorrectly for the channel it actually runs through.

Laboratory diagnostics — commonly called in vitro diagnostics, or IVDs — are tests performed on biological samples such as blood, urine, or tissue outside the human body. The dominant commercial form is a test run in a central laboratory, typically a hospital lab, a reference laboratory like Quest or LabCorp, or a specialty lab focused on a clinical area like oncology or infectious disease. Central labs invest heavily in automation, throughput, and assay panels, which means a new test entering this channel must justify its cost against existing workflows and deliver results that change clinical decisions or reduce downstream costs.

Imaging diagnostics generate clinical information through visual representation of anatomy or physiology. Modalities include X-ray, computed tomography (CT), magnetic resonance imaging (MRI), ultrasound, and nuclear medicine approaches like PET scanning. Each modality has its own physics, its own capital cost structure, and its own clinical use cases — ultrasound is portable and low-cost; MRI is high-resolution but slow and expensive. The device in imaging is typically a capital purchase by a health system, while the clinical service is billed separately as a procedure. Founders entering imaging are almost always building software, hardware components, or AI-assisted reading tools layered on top of existing installed infrastructure.

Point-of-care testing, abbreviated POC, moves diagnostic testing to the location of patient care — a clinic, an emergency department, a pharmacy, a home. The defining characteristic is turnaround time: POC tests return results in minutes rather than hours or days. Speed changes clinical workflow, which is why POC commands a premium in many settings. The tradeoff is analytical sensitivity — POC tests are frequently less accurate than their central lab equivalents, and understanding where that tradeoff is acceptable versus dangerous is one of the most consequential clinical decisions a diagnostics founder makes.

Digital health diagnostics use software, sensors, or algorithms to generate clinical information without a physical assay or imaging modality. Continuous glucose monitors, wearable cardiac monitors, and AI-based screening tools that analyze retinal photographs are all examples. The regulatory boundary that matters here is Software as a Medical Device, or SaMD — a classification that the FDA has developed specific guidance around and that carries its own risk stratification logic. A digital health tool that generates a diagnosis or informs a treatment decision is regulated differently from one that provides general wellness information, and the line between those categories is less obvious than founders expect.

The buyer for each category is structurally different, and this shapes everything about go-to-market strategy. Central lab tests are evaluated by laboratory directors and medical directors who care about analytical performance, workflow integration, and total cost of testing. Imaging decisions are made by radiology departments and capital committee processes that move on multi-year procurement cycles. POC products are often purchased by nursing leadership or decentralized clinic operators who care about simplicity, CLIA waiver status, and training burden. Digital health tools increasingly require hospital IT security review, EHR integration, and clinical champion sponsorship before procurement. Knowing which buyer owns the decision changes who you hire to sell, what evidence they require, and how long your sales cycle will be.

The site of care determines the buyer, the regulatory pathway, the reimbursement code, and the acceptable performance threshold — and none of those choices are independent of each other.

This lesson is coming soon.

TERMS

An IVD is a test performed on a biological sample collected from the human body, analyzed outside the body to generate clinical information. The category includes assays run on analyzers in central laboratories, rapid antigen tests, and molecular diagnostics. FDA regulates IVDs as medical devices, and the specific classification determines the evidence required for market authorization.

POC testing refers to diagnostic testing performed at or near the site of patient care rather than in a central laboratory. The defining advantage is fast turnaround time, which can accelerate clinical decisions in emergency, primary care, and resource-limited settings. The tradeoff is typically lower analytical sensitivity or specificity compared to laboratory-grade instruments.

SaMD is software intended to be used for a medical purpose without being part of a hardware medical device. FDA and the International Medical Device Regulators Forum have developed risk-based frameworks that classify SaMD by the severity of the condition it addresses and the significance of the information it provides to clinical decisions. A SaMD that drives a diagnosis or treatment selection receives more regulatory scrutiny than one used for general informational purposes.

CLIA — the Clinical Laboratory Improvement Amendments — establishes standards for laboratory testing in the United States, and a CLIA waiver designates a test as simple enough to be performed safely outside a certified laboratory. Waived tests can be run in physician offices, pharmacies, and patient homes without a laboratory license. For POC test manufacturers, obtaining a CLIA waiver dramatically expands the addressable market and reduces the operational burden on customers.

Analytical sensitivity measures a test's ability to detect a target analyte at low concentrations — how little of the thing it can reliably find. Analytical specificity measures the test's ability to detect only the intended target without interference from similar substances. These are different from clinical sensitivity and specificity, which measure performance in a patient population, and conflating them is a common error in early-stage diagnostics validation.

A reference laboratory is a high-volume commercial laboratory that receives samples from hospitals, clinics, and physician offices for testing that cannot be performed locally. Quest Diagnostics and LabCorp are the largest examples in the United States. For test developers, reference labs represent both a distribution partner and a competitive threat, because they have the infrastructure to develop or license competing tests.

In imaging, modality refers to the specific technology used to generate the image — X-ray, CT, MRI, ultrasound, PET, and so on. Each modality has different physics, different clinical applications, different capital costs, and different reimbursement structures. A device or software company entering the imaging market must specify which modality it targets, because regulatory clearance, clinical workflow, and buyer relationships are modality-specific.

BEFORE YOUR NEXT MEETING

Which category of diagnostics are we actually building — IVD, imaging, POC, or SaMD — and have we validated that classification with a regulatory consultant before investing in clinical studies?

Who is the actual economic buyer for our product in the channel we are targeting, and have we spoken to that specific person about what evidence they require before a purchase decision?

If our test is POC, have we modeled the performance tradeoff between speed and analytical accuracy, and have we identified the clinical scenarios where that tradeoff becomes patient safety relevant?

Does our product qualify for a CLIA waiver, and if not, have we mapped the certified laboratory network we would need to reach our target sites of care?

If a reference laboratory decided to build a competing version of our test, what would we have that they could not easily replicate?

REALITY CHECK

SOURCES

LESSON 01 OF 04