University College Dublin
Nominated Award: Best Application of AI to achieve Social Good
Website of Company: https://aipremie.com/
Although it claims the lives of 50,000 expectant mothers and 500,000 babies globally every year and results in a further five million premature births, the medical condition of pre-eclampsia remains difficult to diagnose. Front-line clinical staff still rely on high blood pressure and the presence of protein in urine as possible indicators as they battle to do their very best for both the expectant mother and her baby, an approach that has not changed in the last 200 years. A new Irish breakthrough diagnostic test, AI_PREMie, that combines AI with patented biomarkers and clinical expertise, looks set to change everything, by not only accurately diagnosing pre-eclampsia but also predicting the patient’s future
outcome.
AI_PREMie is underpinned by a massive pan-institutional collaborative team of biomedical, medical, and computer scientists, health economists and statisticians from University College Dublin (UCD); front-line staff including obstetricians, haematologists, midwives, research, and clinical lab managers from three of Ireland’s largest and busiest maternity hospitals – the National Maternity, the Rotunda, and the Coombe Hospitals; as well as data engineers, data scientists and healthcare industry experts from SAS Institute and Microsoft. Fuelled by over 300 years of cumulative world-class, basic research, clinical, data science/engineering and industry acumen, together with valuable input from mothers and families who have been affected by pre-eclampsia; the collective mission of the 26-strong interdisciplinary AI_PREMie team is to get their cutting-edge and unique AI-powered solution for the devastating disease of pre-eclampsia to every person who needs it worldwide, as they really do believe it will save lives.
The idea for AI_PREMie was born a decade ago when project lead Prof Patricia Maguire spoke to colleague Prof Fionnuala Ní Áinle about her experiences as a consultant haematologist treating women with severe preeclampsia and the potential for blood-based biomarkers to serve as an early warning system. They went on uncover new diagnostics for pre-eclampsia but also found that they could unlock greater predictive insight by combining biomarker data with high-quality holistic information from the patient using data analytics and machine learning techniques. They both believe that augmented decision making in a hospital environment in real time is where AI is going to be truly transformative in medicine. Right now, the biomarker testing and data analysis is being carried out in UCD but in 2023 AI_PREMie will be deployed in the Rotunda hospital. Beyond this the team have their sights set on global deployment, aiming to reach even small clinics in remote areas with a solution that might involve a blood prick test in tandem with a mobile app that can quickly process the information in the cloud.
The goal is for AI_PREMie to be operationalised in hospitals across the world. “When I say one in ten pregnant women will develop preeclampsia and 500,000 babies die each year, this is probably an underestimation because it is likely that it is underreported in lower income countries.” says Prof. Patricia Maguire, “the dream of the whole team is to reach every person who needs this test across the world”.
Reason for Nomination
Societal Issue
Pre-eclampsia is a serious pregnancy complication typically characterised by the development of high blood pressure and protein in the urine, and it affects one in every 10 pregnancies. Every year it claims the lives of 50,000 mothers and 500,000 babies, making pre-eclampsia one of the world’s deadliest pregnancy complications. Furthermore, as pre-term delivery of the baby is the only cure, an additional 5 million babies are born sometimes very prematurely each year.
Symptoms can really vary between expectant mothers with pre-eclampsia and diagnosis can be a serious challenge as there is no test available to tell clinical staff that these 1 in 10 mothers have preeclampsia. Additionally, as the disease can often rapidly escalate, there is importantly, no test to help front-line staff to make the critical decision on when is the right time to deliver that baby. This is vitally important as preterm, every day inside in utero counts for that baby’s survival chances and long-term quality-of-life.
Pre-eclampsia can therefore be truly devastating, affecting the smallest, most vulnerable members of our society, their families, caregivers and indeed, their whole communities. Accurate risk stratification, where patients are assigned health risk statuses to help inform care, is urgently required to reduce the enormous competing risks for expectant mothers and babies.
It is difficult to sometimes know what to do: whether to deliver a baby because we fear for the mother’s safety or to keep baby in-utero for as long as possible.
— Dr Jennifer Donnelly, Consultant Obstetrician, Rotunda Hospital
Solution
Professor Maguire and a multi-disciplinary collaborative team have developed a prototype risk stratification tool called AI_PREMie. Built by combining AI with world-renowned clinical and scientific expertise, AI_PREMie will be able to augment front-line clinical decision making in real-time, hopefully enabling more accurate diagnosis and personalised treatment of expectant mothers and their babies that will save lives.
The team have used their unique knowledge of platelets, tiny cells that circulate in blood, to uncover new biomarkers to diagnose and predict the severity of pre-eclampsia. They then applied AI to combine quantitative results from the analysis of these patented biomarkers in the blood of expectant mothers together with high-quality, holistic patient information gleaned from multiple sources including demographics, medical history, clinical assessments, and other investigation data. Partitioning the data into training (80%) and validation (20%) sets, the performance of several machine learning models for risk stratification in preeclampsia was compared using receiver operator curve area under the curve (ROC-AUC) and KS Youden statistics. To date, the best performing/champion model to accurately separate high-risk from low-risk mothers was a Gradient Boosting model with an average ROC-AUC of 0.895 and a KS Youden of 0.647. Although based on a dataset of 250 women, this scoring algorithm is a robust indicator of the viability of the unique AI_PREMie approach and gives the team great confidence to move forward with their goal of leveraging their global clinical networks to recruit thousands of women with suspected pre-eclampsia to the project.
The end result will be an easily interpretable risk score that will be delivered and scaled as a cloud-based FaaS (Function-as-a-Service) direct to the front-line staff to augment clinical decision making in real-time. The algorithm will also continually learn and evolve as it is implemented into widespread clinical practice, getting better and better at predicting outcome in pre-eclampsia as it gleans more data worldwide.
Impact
“Accelerating the path to treatment helps to remove patients fear of uncertainty. It reduces psychological stress – all designed to improve outcomes”. Prof Fionnuala Ní Áinle, Consultant Haematologist, the Mater & Rotunda Hospitals.
Maternal health is a significant global and national challenge. The AI_PREMie team have combined their unique expertise to provide a prototype personalised treatment tool that will hopefully enable timely delivery decisions for expectant mothers, transforming their lives and their babies, as well as their families and extended communities.
Improving quality of life in this way aligns to the priorities of several Sustainable Development Goals (SDGs): SDG3 (Good Health and Well-Being), SDG5 (Gender Equality), and SDG10 (Reducing Inequality). This is not only a moral imperative but critical for maintaining international growth.
A 2015 World Health Organisation report acknowledges that many maternal and infant deaths are preventable and in theory could be avoided with effective and timely clinical interventions. The key is to ensure that high-risk pregnancies and complications are recognised early. Thus, new diagnostics are urgently required, and AI_PREMie will fill this gap.
Survivors of preeclampsia have a lifelong increased risk of developing other chronic diseases, such as heart and vascular disease. In fact, preeclampsia is associated with a fourfold increased risk of developing kidney failure within 10 years after pregnancy. This risk is increased even further by having more than one preeclamptic pregnancy, a low-birthweight offspring, or a preterm delivery. Therefore, any improvement in clinical decision-making will have an enormous preventative potential on the long-term health of the population and future healthcare resource requirements.
In Ireland, the costs associated with preeclampsia amount to €6.5-€9.1 million annually and in the US to $9.4 billion. Therefore, any new test has the potential to address the global market as an in vitro diagnostic (IVD) or point-of-care solution. The team believe that within 5 years of deployment, AI_PREMie will be operationalised as a cost-effective part of pregnancy screening programs and transforming the lives of expectant mothers and their babies.
“By providing a timely and accurate prognosis, AI_PREMie will be a game-changer for women with pre-eclampsia and should have a major impact on the health and mortality rates of pregnant women and their babies worldwide. As a clinician, I cannot wait to use it as part of our care.” – Prof Mary Higgins, Consultant Obstetrician, National Maternity Hospital, Dublin.
Additional Information:
Patricia Maguire is an interdisciplinary scientist and inventor who is passionate about the intersection of Artificial Intelligence with Biomedical Science. She is Professor of Biochemistry at University College Dublin and Director of the UCD Institute for Discovery. This institute recently launched the UCD AI Healthcare hub (AIHH), with the ambition to transform healthcare at the individual to the systemic
level.
Patricia’s own research is focused on platelets and extracellular vesicles in several inflammatory diseases including preeclampsia, multiple sclerosis, arterial and venous thrombosis, cancer-associated thrombosis and covid19. She has published widely including the journals Nature Communications, Proceedings of the National Academy of the Sciences, Proteomics, and Blood.
Through her unique expertise, she has developed a bespoke diagnostics platform PALADIN (PlAteLet bAsed DIagNostics) that combines the power of platelets in blood to sense their environment with advanced omics technologies and Artificial Intelligence to uncover secrets of health and disease.
Patricia has used PALADIN to uncover patented diagnostics that can diagnose preeclampsia in sick pregnant women; the multi award-winning AI_PREMie project. She also has a pipeline of potential new disruptive diagnostics from other projects in her lab. She collaborates with industry across multiple sectors including Bayer AG, Sanofi, Mallinckrodt pharmaceuticals, Microsoft, Google and SAS Institute.
Patricia is an advocate and mentor of women in STEM. In 2018, she won a UCD Values in Action award for her work in Equality, Diversity, and Inclusion across UCD and bringing the values of creativity, collegiality, and engagement to life.
She lives in Dublin, Ireland where she is married and is (a swim) Mum to three teenage girls.
Websites/Blogs:
https://aipremie.com/
Game-changer AI tool will save mothers and babies
https://blogs.sas.com/content/sascom/2022/07/05/delivering-the-future-how-biomarkers-and-analytics-in-maternity-care-save-lives-in-dublin/
https://www.ucd.ie/research/impact/casestudies/aipremiesavinglivesofmothersandbabiesusingai/
https://www.ucd.ie/discovery/aihealthcarehub/news
https://www.ucd.ie/conwaysphere/
Social Media:
Twitter: @maguirepatr @AIPREMie #AI_PREMie
LinkedIn: www.linkedin.com/in/patricia-maguire-UCD
UCD project 2021
Nominated Person: Eoin Kenny
Nominated Award: Best Application of AI in an Academic Research Body
University College Dublin (the company) is an educational institution; specifically, the nominee works across two research institutes in the School of Computer Science: the Insight SFI Centre for Data Analytics and the VistaMilk SFI Research Centre. Insight is one of the largest AI centres of its kind in Europe, with over 400 researchers, €100 million in funding, over 80 industry partners, and 8 research institutions. VistaMilk is a SFI and DAFM funded research centre directed at the one of Ireland’s largest indigenous sectors, the Agri-Food Diary sector. The purpose of the centre is to combine Agriculture and ICT expertise to innovate and enhance sustainability across the dairy supply chain.
The work was motivated by two considerations:
1) The first 15 years of AI-R&D funded by SFI in Ireland mainly benefited the US/EU multinational technology sector (e.g.. Microsoft, Intel, Ericsson, IBM, Siemens) through partnerships with Irish Universities. The indigenous Irish multinational, SME, and start-up sectors hardly benefited at all. Hence, a guiding motivation for this work was to benefit Ireland ’s indigenous sectors, specifically the Agri-Food sector, to provide advanced technology research and development support in AI to smaller firms (e.g., farmers and growers) and larger indigenous multinationals (e.g., food companies).
2) The work was also motivated by the enormous challenges of climate change and sustainability impacting the farming sector; specifically, those facing dairy farmers and the Agri-Food sector as a whole. The aim of this research was to produce an AI solution that would be a working system designed to be widely-used by dairy farmers in Ireland. The aforementioned system should help farmers better manage their fa rms from a sustainably perspective by providing accurate grass-growth prediction for feed budgeting for their herds. Accurate forage prediction and budgeting underlies better farm management as it can help reduce fertiliser use (reducing nitrate pollution) and the use of carbon-intensive feed supplements (replacing them with on-farm grass that has much lower transport costs).
PastureBase Ireland (PBI, www.pbi.ie) is a decision support system for dairy-farm pasture management, delivered by Teagasc to several thousand dairy farmers. The long-term aim of this work is to develop an accurate predictive system combining first-principles mechanical models (developed by Teagasc) paired with AI models that use historical data (developed by UCD) to solve the grass-growth prediction problem. This AI system also tackles the accompanying problem of explaining the model’s predictions (ie, explainable AI), as well as trying to handle the difficulties that arise in predicting in the context of climate change.
Reason for Nomination:
We nominate the work of Eoin Kenny (PhD student at UCD) to recognise the innovative and bleeding-edge technological breakthroughs he has made in the last three years in predictive AI modelling, explainable AI and precision agriculture. These discoveries have been recognised internationally by the AI community at major AI conferences and through a Best Paper Award. It would be nice for him to be re cognised in his home-country too.
His contributions have technological and scientific impacts, but also have significant societal and economic impact for the future of the Irish Agri-Food sector, specifically in the context of sustainability and climate change.
TECHNOLOGICAL & SCIENTIFIC IMPACTS
Kenny’s AI work involved solving several key problems using state-of-the-art techniques that have been recognised inter nationally as ground-breaking and highly significant (see Additional Information for details).
Three major AI solutions were developed to solve problems in crop-growth prediction for the dairy sector: (i) the accurate prediction of grass growth from historical data, (ii) the automated explanation of the model predictions to farmer end-users, (iii) the prediction of grass growth under climate disruption, when historical data becomes less useful. Consider each in turn:
PREDICTIVE AI FOR GRASS GROWTH The Teagasc PastureBase Ireland system holds a large dataset of on-farm reports of grass growth entered by end-users (farmers) throughout the year for many years, along with environmental conditions. The problem was to build a predictive AI model using this historical data. However, though extensive, this dataset of almost 100,000 records was very noisy (e.g., including miss-entries and partial records). The AI model Kenny developed, called PBI-CBR, achieved the required accuracy predicting grass growth (a week a head) for individual farms using a case-based reasoning (CBR) or k-NN model. The model’s main novelty was a Bayesian case-exclusion method that cleaned the noisy dataset; it removes records that fall significantly outside the distribution of a gold-standard research dataset (compiled by Teagasc’s in Fermoy). This AI data-cleaning method won a best paper award at ICCBR-19 (Osnabrück, Germany).
EXPLAINABLE AI On top of this innovative predictive model, Kenny also developed an eXplainable AI (XAI) module to explain the model’s predictions. This module used example-based explanations, finding very similar nearest-neighbouring cases to a target case for a given week; specifically, the system found explanatory cases for the same-farm as the target-case telling the farmer that “The model predicts grass growth of 10kg/ha/DM next week because it looks like a week two-years ago on your farm with very similar climate and environmental conditions. Example-based explanation in XAI was well-known, but Kenny’s innovation has been to extend this explanation strategy to Deep Learning models too (by abstracting feature-weights from the neural network). He has also innovated further developing new example-bas ed strategies involving counterfactuals and semi-factuals. This XAI work has been published at the A* leading international AI conferences in the last three years (see IJCAI-19, IJCAI-20, IJCAI-21, AAAI-21) giving Kenny an unprecedented and uniquely successful research profile ahead of any current AI researcher in Ireland.
COMBINED HYBRID MODELLING Finally, Kenny has combined AI models of grass-growth prediction (using a modified neural network model) with traditional mechanical models to gain the best of both modelling worlds to handle the vagaries of climate change. Teagasc traditionally uses a mechanical model (called MoSt that formulaically relates climate variables to soil and other chemical factors affecting grass growth) . Data analyses revealed that the AI model worked better during climate disruptive events (e.g., the 2018 drought and forage crisis), but the mechanical model works better during more-normal periods. So, Kenny built an AI model that learns which model works best for a given time-period, selecting the best model to use for different prediction periods. Therefore, this hybrid model is better positioned to deal with future climate change.
SOCIETAL & ECONOMIC IMPACTS
These AI modelling efforts positively impact the everyday lives of people immediately. It offers solutions to the problem of grass growth prediction for a major indigenous sector, offering this sector a technological basis for a more sustainable agriculture, one that better manages a sustainable resource (grass), while minimising the potentially polluting effects of farming activities. The path to use of these innovations is also clear, as they can be integrated into the existing PastureBase Ireland platform to provide direct predictive capabilities to several 1000 dairy farmers. This state-of-the-art AI work in AI promises to improve the international competitiveness of this Irish Agri-Food sector.
Additional Information:
The nominee, Eoin Kenny, has a uniquely successful publication/award profile amongst the 100s of AI researchers in Ireland over the last few years, with over 15 publications for his PhD
work (see https://scholar.google.com/citations?user=AzMTFY4AAAAJ&hl=en ).
These publications are in some of the most prestigious venues for AI research, including.
1) Kenny, Eoin M., and Mark T. Keane. “Twin-systems to explain artificial neural networks using case-based reasoning: Comparative tests of feature-weigh ting methods in ANN-CBR twins for XAI.” IJCAI-19, Macao, 10-16 August 2019. 2019. A* Core conference, with 80%+ Rejection Rate. Ranked 9th in Google-Scholar Global Impact Rankings
2) Kenny, E. M., Ruelle, E., Geoghegan, A., Shalloo, L., O’Leary, M., O’Donovan, M., & Keane, M. T. (2019, September). Predicting grass growth for sustainable dairy farming: A CBR system using bayesian case-exclusion and post-hoc, personalized explanation-by-example (XAI). In International Conference on Case-Based Reasoning (pp. 172-187). Springer, Cham. B Core conference, with 60%+ Rejection Rate. Best Paper Award
3) Kenny, E. M., Ruelle, E., Geoghegan, A., Shalloo, L., O’Leary, M., O’Donovan, M., & Keane, M. T. (2 020). Bayesian Case-Exclusion and Explainable AI (XAI) for Sustainable Farming. IJCAI-20, Montreal, 2020 (Best Paper Track) A* Core conference, with 80%+ Rejection Rate Ranked 9th in Google-Scholar Global Impact Rankings
4) Kenny, E. M., & Keane, M. T. (2021 , May). On Generating Plausible Counterfactual and Semi-Factual Explanation s for Deep Learning. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 35, No. 13, pp. 11575-11585). A* Core conference, with 80%+ Rejection Rate Ranked 4th in Google-Scholar Global Impact Rankings
5) Kenny, E. M., Ford, C., Quinn, M., & Keane, M. T. (2021). Explaining black-box classifiers using post-hoc explanations-by-example: The effect of explanations and error-rates in XA I user studies. Artificial Intelligence, 294, 103459. Journal Impact Factor 9.088+
6) Kenny, E. M., & Keane, M. T. (2 021, Dec). Explaining Deep Learning Using Examples: Optimal Feature Weighting Methods for Twin Systems Using Post-Hoc, Explanation-by-Example in XAI. Knowledge Based Systems Journal Impact Factor 8.038 Ranked 13th in Google-Scholar Global Impact Rankings