In this work we find that vital signs embed racial information that can be learned by AI algorithms, posing a significant risk to equitable clinical decision-making. Mitigating measures are challenging, considering the fundamental role of vital signs.

I am among the World’s top 2% scientists based on Scopus data, published by Elsevier (doi.org/10.17632/btchx…)

Thanks for the invite Healthy Lifespan Institute Enjoyed talking about #GenerativeAI for #clinical data. The details of the talk can be found in our paper venetosmani.com/publications/g… Information School

Our GAN-based generative AI can synthesise authentic, high-dimensional, time series clinical data, outperforming state-of-the-art methods, while avoiding mode collapse (paper: venetosmani.com/publications/g…)

Sharing our recent article that looks at the changing nature of clinical variables that predict #delirium over time. sciencedirect.com/science/articl…
Venet Osmani

Thank you for the invite National Research School in ICT! Looking forward to sharing my research experience in AI for health University of Prishtina

Stimulating discussions, interesting ideas and a bit of fun marked the recent away day of the Information School's Health Informatics Research Group!

Only few days remaining

Funded PhD scholarship to work on the cutting-edge of Machine Learning, Medicine and Social Science, at Healthy Lifespan Flagship Institute (Healthy Lifespan Institute) and Information School at The University of Sheffield, a world top-100 university
findaphd.com/phds/project/m…

If you are interested in one of these three positions in my DSH Lab at Fondazione Bruno Kessler - FBK , please contact me through DM or at [email protected] !

My team is hiring! We have two amazing openings for staff ML scientist positions for exciting work at Tempus (plenty of EHRs, ECGs and other multi-modal data to work with). Apply online if interested and please share with your network as well. DM if you have any question.

Best paper award at NeurIPS 2022 workshop on SyntheticData4ML ! We devise a method to generate synthetic clinical data for underrepresented groups to mitigate health data poverty. Paper: arxiv.org/abs/2210.13958 with Giuseppe Jurman

You couldn't make it to #NeurIPS2022 this year?

Nothing to worry - I curated a summary for you below focussing on key papers, presentations and workshops in the buzzing space of ML in Biology and Healthcare 👇

Machine learning methods predict outcomes in Asian, African and American COVID-19 patients using a model derived solely on European patient cohort.
Map below shows the prevalence of COVID-19 for each of the 37 countries, 17 European and 20 non-European. doi.org/10.1371/journa…

We devise a conditional GAN (CA-GAN) architecture that outperforms wrt to state of the art GAN and SMOTE in generating synthetic data to mitigate health data poverty for underrepresented groups Giuseppe Jurman

We also investigate the trade-off between the length of observation, prediction window and performance of the model in predicting delirium in intensive care units. Collaboration between Mayo Clinic Cleveland Clinic MIT IMES and Fondazione Bruno Kessler - FBK

Predicting COVID-19 outcomes using a model derived from a multi-centre cohort of European elderly patients, and externally validated on an Asian, African and American patient cohort Fondazione Bruno Kessler - FBK TrentinoSalute4.0 Behrooz Mamandipoor

Enjoyed talking about AI in medicine at the 28th congress of the Italian Society of Palliative Care (Società Italiana di Cure Palliative - SICP). Many thanks for the invite!