The 1st Workshop on

AI in Predictive HEAlth: architectures for prevention (AIPHEA)

21-26 Jun - from hh:mm to hh:mm, Room xyz

Workshop Aims

A revolutionary approach in healthcare is the shifting of its focus from treating diseases to proactively preventing them. Machine Learning models and Neural networks can play an important role in this process, unlocking new ways to interpret complex and various medical data generated as healthcare continues to digitalize. Our workshop, "Neural Network Frontiers in Predictive Medicine: Architectures and Applications for Prevention and Early Diagnosis", is designed to be a meeting point for the researchers, clinicians, and engineers who are shaping this new landscape. We will explore the latest machine learning models and neural network architectures and their tangible impact on early diagnosis and prevention.

The sessions will cover a broad range of applications, models used to interpret visual data, such as medical scans, and architectures designed to analyse sequential data from patient histories. We will also explore multi-modal methods for integrating diverse data sources, such as genomic and clinical information, to create a more comprehensive view of patient health. This workshop will balance two key aspects: we will showcase practical solutions to real-world clinical challenges while also creating a space for discussing the next generation of theoretical models. We aim to encourage a conversation that bridges the gap between groundbreaking theory and the development of robust, trustworthy tools for predictive medicine.

Possible topics of interest include (not limited to):

● Identify State-of-the-Art Architectures: Recognise and describe the latest neural network architectures being successfully applied to various data types (imaging, sequential, and graph-based) for predictive health.

● Understand Application Challenges: Grasp the primary challenges associated with deploying neural networks in clinical settings, including data heterogeneity, model interpretability (XAI), and regulatory hurdles.

● Applications in Bioinformatics and Computational Biology: Explore the use of ML models for analyzing complex biological data, including genomic, proteomic, and metabolomic datasets.

● Biomarker Discovery: Investigate methods for identifying novel biomarkers from diverse biomedical data using machine learning techniques.

● Novel Datasets, Anonymization, and Synthetic Data: Discuss the presentation of new biomedical datasets, challenges in anonymization, and the generation of synthetic data for training robust models.

● Evaluate Model Suitability: Critically assess which types of neural models are best suited for specific clinical problems in early diagnosis and disease prevention.

● Foster Cross-Disciplinary Collaboration: Engage in meaningful dialogue with experts from both the machine learning and medical fields to spark new ideas and collaborations.

● Pinpoint Future Research Directions: Identify promising open problems and emerging trends in the field that will shape the future of neural networks in preventative medicine.

Important Dates