The Transplant Dilemma
How can we expand the donor pool of available kidneys, without compromising outcomes?
Improved utilisation of kidneys from older donors could enhance transplant rates, providing better access, particularly for older recipients. Remarkably, only half of these kidneys are currently accepted when offered, whilst over 5,000 patients remain on the wait list.
Professor Gavin Pettigrew
Clinician Scientist Fellow and Professor in Experimental and Clinical Transplantation at the University of Cambridge
A National Digital Pathology Service
A quality assessment tool for the future
In ancient times, the Pythia, or Oracle of Delphi, was renowned for her prophecies—a legend cautioning against the perils of uncertain predictions. In modern medicine, the stakes are equally high. Today’s clinicians are confronted with the challenge of identifying safe transplants in the face of organ scarcity, and increasing need. Yet their tools have not kept pace. Diagnostic pathology has remained the same for over a century. Current methods grossly under-sample the tissue by up to 800-fold, limiting their reliability and the confidence doctors can have in their results. These methods are also time-consuming and laborious; and are required at unsociable hours; increasing the the demand on an already overextended histopathology workforce.
Trusted automation offers services an alternative route to high quality 24hr quality assessment, by enabling clinicians to conduct biopsy assessments with remote oversight. This approach not only makes specialist pathology expertise more widely accessible, but brings assessments methods into the 21st century and stands to satisfy an urgent need within UK transplant and globally.
More on the PITHIA trial ↗
Our team
World class expertise
Professor Pettigrew leads multiple NIHR-funded multicentre studies, most notably the PITHIA trial, which pioneered the first National Digital Pathology Service. Additionally, he serves as the chair of the NHS Blood and Transplant (NHSBT) Research, Innovation, Novel Technologies Advisory Group (RINTAG).
Professor Pietro Lio is a key member of the Cambridge Centre for AI in Medicine. His research is dedicated to advancing neural network modelling, in health applications.
Dr. Samoshkin brings a wealth of experience from his tenure at world-class biomedical laboratories, including the NICHD and NCI at NIH, USA, and UNC Chapel Hill. He is responsible for external partnerships that drive innovation.
Dr Tobi Ayorinde was a NIHR Academic Clinical Fellow, and continues his clinical academic training. He plays a prominent role in driving progress and the success of the interdisciplinary effort, bridging the gap between clinical, AI, and pathology teams.
Clinician Scientist Fellow and Professor in Experimental and Clinical Transplantation at the University of Cambridge
Translational Technology Manager University of Cambridge
NIHR Academic Clinical Fellow | NHS Clinical Entrepreneur
Professor of Computer Science and Technology at the Cambridge Centre for AI in Medicine
The PITHIA app
A new gold standard
The PITHIA app aids clinicians to identify key anomalies which indicate kidney disease. At the heart is the integration of Human-in-the-Loop Artificial Intelligence (HIL-AI), trained on over 15-years of real world transplant activity. Leveraging SAS’s dynamic analytics platform, our goal is to create an intuitive, data-driven application.
We understand that the nuances of clinical judgment need to be reflected in the engineering choices we make. Therefore, the PITHIA app is carefully designed to mirror the eye of an overseeing pathologist. The user interface offers real-time interpretable analysis and the flexibility for clinicians to engage with the assessment process as deeply as they feel necessary.
- The app uses established criteria for renal transplant quality assessment to ensure clinical validity.
- Future iterations will explore how AI can not only complement, but also elevate human expertise, fine-tuning the predictive accuracy of assessment standards
- Doctors can combine assessments from multiple slides from the same biopsy, organ or patient
- A consolidated pathological report is generated, combining both the software’s analytics and the clinician’s expertise.
Faster
Empowering doctors
- Parallel analysis of multiple histological sections reduces assessment time. Our efficiency allows for more comprehensive tissue evaluation without prolonging the kidney’s cold ischemia time, thus preserving transplant outcomes
More reliable
Precision
- We enable multi-slide and multi-biopsy protocols to surpass human level accuracy and, importantly, reliability.
- This mitigates the significant inter-observer variability traditionally observed among pathologists
More accurate
Confidence
- Our app aims to reduce kidney discard rates and help more patients off dialysis.