10085226NISTA: a novel non-invasive, data-enabled ML tool enabling earlier and more accurate diagnosis of pancreatic cancer, stratifying patients into more personalised treatment pathways to improve outcomes.ClosedCollaborative R&DInnovate UK~50% of the global population will develop cancer in their lifetime \[WHO,2022\]. Pancreatic cancer is the 11th most common cancer in the world with ~495,773/year new cases and almost as many deaths (~466,003/year)\[Sung,2021;CRUK,2023\]. Additionally, pancreatic cancer has one of the worst survival rates, with one--year survival at ~24%, compared to 70% for the 20 common cancers combined. The overall 5-year survival for pancreatic cancer has changed little over the past decade(~9%)\[Neoptolemos,2018\]. This is due to pancreatic cancer often being diagnosed late(80%:diagnosed StageIII/IV), in part through lack of diagnostic sensitivity, reducing options for treatment. Early diagnosis of pancreatic cancer is key to improving chances of survival. Furthermore, understanding the subtype of cancer (classical/basal-like) is key to personalising the treatment pathway and improving health outcomes. As the global population rises and ages, pancreatic cancer is predicted to be the second leading cause of cancer-related mortality in Western countries \[Neoptolemos,2018\]. Current-state-of-the-art in pancreatic cancer diagnostics is limited to imaging the cancer through approaches such as pancreatic protocol computerised tomography(CT) or endoscopic ultrasound(EUS). However, studies have found these have limited sensitivity when tumours are in early development. Additionally, compared to other cancers, precision medicine approaches that translate the increased knowledge of the molecular pathology of cancer into the clinic are in their infancy for pancreatic cancer \[Froeling,2021;Siegel,2022\]. In this 24-month industrial research project, EosDx (UK-SME-Lead) will collaborate with Ulster University to develop and optimise NISTA, a unique data-enabled machine-learning tool to more accurately diagnose and stratify pancreatic cancer patients according to _proteomic structural biomarkers_. The consortium possesses leading expertise in ML-model development, structural biomarkers, bio-crystallography, clinical testing, and pancreatic cell function, and has previously developed game-changing technologies within the diagnostic sector. NISTA improves significantly over the current-state-of-the art as it allows low-cost, pre-symptomatic, rapid, safe, non-invasive and highly accurate diagnosis and subtype stratification in one scan. This project provides a step-change within pancreatic cancer diagnostics, expanding the sector from within the UK and driving the growth of a completely new industry sub-segment. EosDx aligns with the NHS's long-term plan to improve national screening programmes, giving people faster access to diagnostic tests (<2-hours for NISTA), investing in cutting-edge technologies, and ensuring more patients benefit from precise, highly-focused treatments \[NHS,2023\].