According to the national transplant registry data, transplantation serves hundreds of thousands of patients each year, and requires collaboration to create an optimal patient experience. A lag in the system, such as having outdated transplant data on file, can be detrimental to the process and, ultimately, the patient. Imagine what could happen if a clinician were to make decisions about medication dosages based on inaccurate data. It could lead to organ rejection or other serious health complications, such as severe infections or damage to the transplanted organ.
By leveraging artificial intelligence (AI), labs have the potential to significantly improve organ transplantation by delivering results faster and enhancing analysis at every step. AI can help healthcare teams determine risk signals earlier, interpret immune and molecular data faster and make more informed decisions. In clinical settings, AI can support clinicians in determining accurate dosing of immunosuppressants, setting appropriate timelines for follow-up testing and flagging abnormal results. This could enable earlier detection of organ rejection and faster intervention. In fact, studies are already showing the positive effects of adopting next-generation technologies like AI. A report from the National Library of Medicine found that AI and machine learning (ML)-based kidney allocation models had a higher accuracy in predicting graft survival and waitlist mortality than traditional risk scores, which resulted in more efficient use of donated kidneys.
AI’s role in transplant diagnostics is still evolving, and its integration in the transplant industry is slower compared to other specialties due to higher stakes and less standardized workflows. There are also strict regulatory requirements in the clinical lab, and existing regulations such as the Clinical Laboratory Improvement Amendments (CLIA) may need to evolve to ensure that AI systems are appropriately validated and managed. While the benefits of using AI are clear, adoption can’t be rushed, particularly when patient lives are on the line. Every model, workflow, step and output must be validated, documented and safely implemented by lab managers and clinicians before it’s fully integrated.
Why AI adoption in transplant diagnostics is methodical
AI is best used in settings where consistency is mainstream. In transplant labs, workflows are often less standardized and data is more complex, creating a significant operational challenge.
The complexity and scale of transplant lab data make it uniquely challenging to integrate the technology. Decisions in the transplant lab rely on a wide range of data, such as donor characteristics and pre-transplant laboratory measurements. These decisions are also used in post-transplant outcomes, including graft function, rejection episodes, and response to immunosuppressive therapies. AI is well-positioned to sort through all the data in the transplant lab, but in practice, it is much more challenging.
AI tools used in the lab must perform reliably and consistently across various sites and patient types, especially when new instruments or workflows are producing critical data. Tools could work well in one center or patient population but produce drastically different results in another. To start the integration of this technology, lab leaders should begin with focused, well-defined use cases before expanding into broader rollouts.
The role of AI in the lab today should be to support clinician teams in their decision-making, such as notifying them if a patient is at a higher risk of rejection, rather than making the final call. In this step, the clinician’s role remains crucial to the transplant process.
Ultimately, AI should support and inform decisions rather than replace clinical reasoning. AI does not replace quality control, staff competency, documentation or validation practices; those expectations are still prevalent and require clinician oversight, especially as we aim to remove bias.
How AI can enable more personalized transplant care
The various potential risks for a transplant patient can change rapidly based on immune status, infections and medication levels and compliance. Given the array of factors that must be considered in transplant care, being able to analyze various potential risks all at once and leveraging the data to determine the best treatment paths for each patient can transform their quality of life.
While AI’s role in the lab is evolving, the human leukocyte antigen (HLA) lab or transplant physician must still determine an appropriate response to data or AI outputs, such as additional review, repeat testing or escalation. For labs using AI, it’s critical to establish clear guardrails, including where human review is mandatory and where AI should not be used, to ensure safe, appropriate application.
AI has the potential to enhance patient care in HLA typing because it can help match immune markers between donors and recipients. By doing this, clinicians can reduce the risk of rejection and can determine antibody profiles, immune risk factors and trends after transplant to ensure each patient is receiving care that is specific to them.
By leveraging AI, clinicians can act on each patient’s full data history by compiling information across labs, clinical context and historical trends. Decisions are better tailored to the individual and labs can rapidly analyze HLA and other compatibility data at scale.
AI can sort through match pathways to support faster, more consistent reviews and discussions. It can also find inconsistent results that warrant repeat testing or manual review. By leveraging AI in the clinical lab, clinicians can support more patient-fit plans, determine which patients need closer monitoring and engage in consistent clinician oversight of medication regimens.
AI’s use in the clinical lab is here, but implemented gradually
Any AI adoption plan should designate a single, accountable owner who oversees the full lifecycle of the technology and should include specific instructions on when to stop using it.
Expectations around AI governance are evolving, but having the right foundation is crucial. Once the foundation is set, AI can start to unlock more personalized, patient-first transplant care without compromising safety. Therefore, establishing a cross-functional review group with regular oversight will be increasingly important to include in any governance plan. Lab teams should also deploy strict access controls, including least-privilege permissions, role-based access and audit trails. Lastly, with the changing regulatory landscape, ensuring that consent and privacy rules are clear and documented wherever local requirements apply will be crucial to the successful integration of AI in the transplant lab.
The strongest AI adoption plans follow a simple pattern: start small, validate and scale only when results hold up across sites, shifts and real-world variation. The goal is safer, more personalized care, which is made possible by clinicians receiving earlier signals, better-tailored monitoring and more confident decisions for each patient. Think of AI as a second set of eyes, not an autopilot. It can scan far more data than any single team can, but the clinicians still own the call. AI adoption in the lab can significantly improve patient outcomes by flagging discrepancies earlier, comparing similar patient records to determine treatment plans and sorting through vast amounts of data faster. These use cases are still emerging, and as adoption grows, we move closer to a future where more personalized transplant care is a reality.
Photo: eternalcreative, Getty Images
Tina Liedtky has over 20 years of leadership experience in the medical device and diagnostics industry. As President, Transplant Diagnostics, Tina is responsible for the strategic direction and growth of Thermo Fisher Scientific’s transplant diagnostics portfolio across the patient care continuum. Previously, she served as President, Clinical Diagnostics, managing a broad portfolio of businesses within Thermo Fisher’s Specialty Diagnostics Group. Prior to joining Thermo Fisher, Tina led US and global commercial teams for the Diabetes Care and Rapid Diagnostics divisions of Abbott Laboratories, as well as sales and marketing leadership roles at Medtronic, Covidien, and Boston Scientific.
Tina serves on the Board of Directors for Neurovalens, a non-invasive neuromodulation company based in Belfast, Ireland, and is actively involved with Miracle Babies, a nonprofit organization focused on supporting families with babies in the NICU. She holds a bachelor’s degree in cognitive neuroscience from Princeton University and an MBA from Harvard Business School.
This post appears through the MedCity Influencers program. Anyone can publish their perspective on business and innovation in healthcare on MedCity News through MedCity Influencers. Click here to find out how.
