An industry perspective on how AI, automation and connected diagnostics will reshape patient care

Clinical diagnostics is entering a new phase, shaped by artificial intelligence, automation, connected data and testing models that bring care closer to the patient. In this CLINICAL24 interview, Duncan Porter, Vice President, General Manager, Immunodiagnostics, Plasma Proteins and Drug Testing, Siemens Healthineers and CLINICAL24 Ambassador, draws on more than 25 years of experience across clinical diagnostics and life sciences to explore where innovation is already delivering value, why adoption at scale remains challenging, and how laboratory leaders can turn emerging technologies into meaningful improvements for patients and healthcare systems.

As a CLINICAL24 Ambassador, Porter also considers where laboratories should focus their efforts now to build more productive, accessible and insight-driven diagnostic pathways.

What developments have exceeded your expectations, and where do barriers remain?

DP: What has probably exceeded my expectations is the speed at which diagnostics is moving beyond the traditional laboratory model.

We are seeing much greater convergence between diagnostics, digital technology, AI, remote sampling, and decentralized testing. In particular, the progress in areas such as liquid biopsy, molecular diagnostics, digital pathology, and increasingly sophisticated home or remote sample collection has been remarkable. The ability to generate clinically meaningful information from smaller sample volumes and potentially closer to the patient is opening up completely different models of care.

AI has also progressed faster than many of us anticipated. We have moved quickly from discussing what AI might do to identifying practical applications across laboratory operations, image interpretation, clinical decision support, and workflow optimization. Both in clinical and research labs, as well as in the manufacturing and development of products and solutions which serve these labs.

Where I think progress has been slower is adoption at scale. Healthcare is understandably conservative when introducing technologies that influence clinical decisions. Innovation must demonstrate not only analytical performance, but clinical utility, health-economic value, integration into existing workflows, and ultimately better outcomes for patients.

Interoperability remains another significant challenge. We have tremendous amounts of diagnostic data, but it often sits in fragmented systems. The opportunity is not simply to generate more data; it is to connect that data and turn it into actionable clinical insight.

So, for me, the next phase of diagnostics is increasingly about implementation rather than invention.

Where will AI deliver the greatest value in diagnostics over the next three to five years?

DP: I see three major opportunities. The first is laboratory productivity and automation. Laboratories are under enormous pressure from increasing testing volumes, workforce shortages and financial constraints. AI can help optimize workflows, predict demand, manage inventory, identify instrument or process issues before they become failures, and reduce the amount of manual intervention required.

The second is interpretation of increasingly complex diagnostic information. Whether that is digital pathology, imaging, genomics or large diagnostic datasets, AI can help clinicians recognize patterns that would be extremely difficult for an individual to identify consistently. This will become more compelling for patients with complex comorbidities or chronic conditions, tracking disease progression and treatment optimization.

The third, and potentially the most transformative, is bringing multiple sources of information together around the patient. Imagine combining laboratory results, imaging, genomics, longitudinal medical history, and potentially wearable data to provide clinicians with much more contextualized diagnostic insight.

One area I find particularly exciting is the emergence of the patient digital twin. We are beginning to see genuine clinical evidence, particularly in cardiology, where a digital representation of an individual patient's heart can be used to simulate an intervention before treating the actual patient.

Imagine extending that concept further. If we can combine laboratory diagnostics, imaging, genomics, pathology, longitudinal clinical history and potentially wearable data, we start to build a much richer digital representation of the individual.

The really interesting question then changes from ‘what is happening to this patient?’ to ‘what is likely to happen next, and what happens if we intervene?’

I think that is potentially one of the most profound applications of AI in precision medicine over the next decade.

We shouldn't overstate where we are today. Most digital twins remain organ-specific or disease-specific rather than a complete digital representation of a person. The barriers are data interoperability, validation, regulation and demonstrating that predictions translate into better clinical outcomes.

But there are important challenges.

We need high-quality and representative datasets, appropriate clinical validation, transparency around how algorithms are developed, clear regulatory frameworks and, critically, clinician confidence.

I do not see AI replacing laboratory professionals or clinicians. I see it augmenting their expertise, removing lower-value activities and allowing highly trained professionals to spend more time on interpretation, complex cases and patient care.

Ultimately, the test of AI will not be how sophisticated the algorithm is. It will be whether it improves clinical outcomes, productivity or access to healthcare.

How can diagnostic innovation help healthcare systems do more with fewer resources?

DP: I think we have to challenge the assumption that innovation always means adding another test or another piece of technology. Some of the most valuable innovation will be about redesigning the diagnostic pathway itself.

Automation is clearly part of that. Laboratories need platforms that consolidate testing, reduce manual intervention, improve uptime and enable staff to manage larger workloads without compromising quality.

But there is also an enormous opportunity to move appropriate parts of the diagnostic pathway closer to the patient.

Advances in capillary sampling, remote collection and decentralized testing potentially allow patients to provide samples without travelling to a hospital or phlebotomy center. That can improve access for people in rural communities, patients with chronic diseases requiring frequent monitoring, or people for whom taking time away from work or caring responsibilities is difficult.

The laboratory remains central because quality, standardization, and clinical governance are essential. But the front end of the diagnostic pathway can become much more distributed.

The other major opportunity is better use of data. If we can connect diagnostic information more effectively, we can potentially identify disease earlier, avoid unnecessary testing and intervene before patients require more expensive care.

So when healthcare systems talk about doing more with fewer resources, I would frame diagnostics not simply as a cost to be reduced but as an investment that can make the broader healthcare pathway more efficient.

A relatively small investment in earlier and better diagnosis can potentially prevent considerably greater downstream costs.

Where should laboratory leaders focus their innovation efforts today?

DP: I would prioritize four areas. First, automation and workflow productivity. With workforce shortages unlikely to disappear, laboratories should invest in technologies that simplify processes, consolidate testing and allow highly skilled people to focus on activities where their expertise genuinely adds value.

Second, connectivity and data. Laboratories already generate extraordinary amounts of information. The next opportunity is to connect that information across laboratory systems, electronic health records and other diagnostic modalities, and translate it into clinically actionable insight.

Third, decentralized access to diagnostics. Remote sampling, capillary collection and appropriate point-of-care testing can make diagnostics more accessible while also reducing pressure on traditional healthcare infrastructure. The important point is ensuring those approaches remain connected to high-quality laboratory testing and clinical pathways.

And fourth, partnerships. I increasingly believe that some of the most important innovations in diagnostics will not come from one organization working alone. Diagnostic companies, laboratories, technology companies, pharmaceutical companies, healthcare systems, and increasingly, AI companies all bring different capabilities.

For laboratory leaders, therefore, I would not ask simply, “What technology should we buy?”

I would ask, “What problem in our healthcare pathway are we trying to solve, and who do we need to partner with to solve it?”

That change in mindset is important because innovation should never be technology looking for a problem. It should start with an unmet clinical or operational need and work backwards from there.

Turning diagnostic innovation into better patient care

Taken together, these developments signal an especially exciting period for diagnostics.

Historically, diagnostics has sometimes been viewed as a relatively small component of healthcare expenditure. Yet almost every major clinical decision depends in some way on diagnostic information.

The opportunity now is to move from simply providing a test result toward providing actionable insight, earlier, more efficiently and increasingly closer to the patient.

If we combine advances in laboratory science with AI, automation, connected data and new models of sample collection, diagnostics has the potential not only to improve laboratory productivity but to fundamentally improve how healthcare is delivered.

Duncan Porter spoke with Sonia Nicholas, Managing Editor and Clinical Lead at SelectScience, as part of the CLINICAL24 campaign exploring the people, technologies, and innovations shaping modern clinical laboratories.