The call will be open from 11 October to 10 November 2022.
The start of the doctorate will be on January 1st, 2023. Those who graduate by December 31st, 2022 can also apply.
Title: Artificial Intelligence methods and techniques for the understanding of texts and risk estimation, Predictive models for healthcare claims management
The student will be able to enter the context of the DISIT lab for the development of the doctoral thesis and its training in the AI, Explainable AI and NLP (natural language processing) fields, in close collaboration with the Careggi University Hospital in the context of the ecosystem of innovation of the PNRR called “The Tuscany Health Ecosystem”. The PhD student will have to deal with the study and development of AI solutions for understanding the text, for example with BERT techniques, Bidirectional Encoder Representations from Transformers, explainable BERT, for understanding the text as well as the development of AI risk prediction models. The application domain is that of the medico-legal area, which specializes in medico-legal terminology and the assessment of the risk of litigation. Integrations are envisaged with ontologies and knowledge bases in the medical field, and with graph neural network techniques, GNN, as well as transfer learning, generative techniques, etc. The objectives can be multiple, the prediction of the risk of litigation, the assessment of complexity, the assessment of influence of the various factors (type of error and damage, discipline, competence of consultants and lawyers, type of patient, etc.), classification of disputes, production of summaries, prediction of evolution and economic impact, automatic anonymization, etc.
The University Hospital of Careggi is a national center and the primary hospital structure for Tuscany. DISIT Lab is one of the most active labs in Tuscany on AI / XAI and NLP issues, it is involved in the CBDAI (Big Data AI Tuscany regional center), CINI Big data node, etc., also in relation to the platforms Https://www.
For DISIT Lab publications see: http://www.disit.org/5487
For the CV of the primary Tutor: https://scholar.google.
For more info: paolo.nesi@unifi.it