Centre for Doctoral Training in AI-enabled Healthcare



Find out about the latest news and upcoming events at CDT in AI enabled Healthcare Systems


Future of AI in Healthcare CDT Seminar Series
During year one the CDT will be hosting a series of seminars around the theme of 'Future of AI in Healthcare'. See below for further details. These events are open to the public. 

Upcoming Events

March 12: 13.00-14.15
JARVIS: Prioritizing non-coding regions based on human genomic constraint and primary sequence context with deep learning
Speaker: Dimitrios Vitsios, Associate Principal Scientist, AstraZenecar 
Elucidating functionality in  non-coding regions is a key challenge in human genomics. It has been shown that intolerance to variation of coding and proximal non-coding sequence is a strong predictor of human disease relevance. Here, we integrate intolerance to variation, functional genomic annotation (such as methylation and chromatin accessibility) and primary genomic sequence to build “Junk Annotation” Residual Variation Intolerance Score (JARVIS): a comprehensive deep learning model to prioritize non-coding regions. JARVIS outperforms comparable human lineage-specific scores in inferring pathogenicity of non-coding variants. Furthermore, despite not incorporating information on evolutionary conservation, JARVIS performs comparably or outperforms other conservation-based scores in classifying pathogenic single-nucleotide and structural variants. In constructing JARVIS, we introduce a new intolerance metric: the genome-wide Residual Variation Intolerance Score (gwRVIS), which uses a sliding-window approach applied to Whole Genome Sequencing data from 62,784 individuals and is among the most important features in JARVIS. Both JARVIS and gwRVIS capture previously inaccessible human-lineage constraint information to help prioritize genetic variants found in the human non-coding regulatory sequence and will enhance our understanding of the non-coding genome.
This event will take place on Zoom. To RSVP please email: aihealthcdt@ucl.ac.uk

Previous events
04/10/2019Julien Fauqueur, (BenevolentAI)Relation extraction from biomedical literature for drug discovery
22/11/2019Dr Hugh Harvey (HardianHealth)Bench to Bedside: Understanding Regulatory Pathways for Delivering Health Technology to Patients
06/12/2019Jessica Rose Morley (Oxford Internet Institute)AI for healthcare: how to get it right?
24/01/2020Benjamin Irving (Sensyne Health, Oxford University)Machine learning for patient stratification and outcome prediction from ‘real world evidence’
28/02/2020Paul Clarke (Health Data Insight CIC)Simulated datasets for health data access and research
20/03/2020John Reid (Blue Prism)Bayesian Active Learning with Gaussian processes
03/04/2020Ciarán M. Lee (Babylon Health)Causal Inference in Healthcare
06/08/2020Mustafa Ghafouri (Data Scientist, Hitachi) and Adrian Conduit (Director, Hitachi)Practical challenges in operationalising data science models to improve flow, coordination and care in hospitals
19/10/2020Caterina La Porta and Stefano ZapperiEstimating individual susceptibility to Sars-CoV-2 in human subpopulations using artificial neural networks
15/01/2021Giuseppe Sollazzo , NHSXIntroduction to NHSX
29/01/2021Kenton O'Hara, Microsoft Facilitating proactive care planning: using Machine Learning for Predicting Hypotension in the Post Anaesthetic Care Unit