Speakers
The Association of Clinical Biochemistry & Laboratory Medicine
Artificial intelligence (AI) has potential to revolutionize pathology diagnosis; accuracy is dependent on the training dataset. Sensitivity to input changes reduces real-world performance. Generalizable AI requires diverse data obtained from multiple sources, scanners, and staining platforms.
To address this problem, NPIC have created a unique multi-scanner facility, the NPIC AI FORGE (Facilitating Opportunities for Robust Generalisable data Emulation). This facility comprises 15 scanners from 8 vendors as follows:
Leica Biosystems (3 x GT450s, 1x AT2), Roche (DP200, and DP600), Hamamatsu (S360MD, S60, S60v2, and S20), 3DHistech (P250 Flash), Olympus (VS200), Objective Imaging (Glissando), OptraScan OS-Ultra 240, Akoya PhenotImager HT.
The NPIC AI FORGE has broad capabilities including high-capacity WSI brightfield scanning of 1x3 inch slides; 3x2 inch slides; multiple magnification capture; z-stacking across 3 instruments, and WSI fluorescence scanning. The NPIC AI FORGE can generate over 3500 WSI images per day, equating to over 5Tb of data. The NPIC AI FORGE is supported by a research image management system, and secure research data repository with 15 petabytes of backed up storage, and disaster recovery mitigations, ensuring the safety, security and longevity of all digital pathology data generated. Our PACS system presents as an easy to access online portal for rapid QC, viewing, annotation, project creation and sharing, and download.
The ability to compare multiple scanning platforms provides an opportunity to map differences between image data sets, to accurately measure the impact this has on AI specificity. We are able to accurately QC each system to specific calibration settings, ensuring robust and reliable data for AI training. Furthermore, unlike other areas of healthcare, slides can be re-scanned multiple times, which offers an opportunity develop high-quality cross-platform AI. To our knowledge, this is the only such digital pathology facility in the world.
Matt is the Research Operations Manager at the National Pathology Imaging Cooperative (NPIC), at Leeds Teaching Hospitals NHS Trust.
Matt is an accomplished digital pathology scientist and subject matter expert in molecular pathology, immunohistochemistry, and digital image analysis. He has over a decade of experience in senior leadership roles within the pathology domain. Prior to NPIC, he was the Scientific Lead for Tissue Hybridisation & Digital Pathology at the Precision Medicine Centre of Excellence, Queen’s University Belfast.
His currently directs the NPIC AI FORGE and Biomarker sections of NPIC at St James University Hospital, Leeds. His team lead on the utilisation of digital pathology solutions, high throughput digital slide scanners, and automated staining platforms. Dr Humphries is PI and Co-I of multimillion pound research programmes, funded by the likes of NIHR, MRC, and CRUK. As Study Director, he had led projects and collaborations with large internationally recognised pharma-organisations for example; Leica, Roche, and Philips.