data analytics

learn_bAIome offers workshops and trainings in biomedical AI/data science with tailored formats that take into account background, programming skills and intensity to provide unique, focused, and effective courses. These courses are free and open to students, clinicians, and researchers across academic institutions in Hamburg.

Prerequisites: Intermediate level computational background and basic knowledge of machine learning

Description: This 3-day international workshop is organised by University of Hamburg’s European University Alliance for Global Health (EUGLOH), Hub of Computing and Data Science (HCDS) and Center for Biomedical AI at UKE (bAIome) to foster international exchange and cooperation among students and researchers working in machine learning relating to biomedical questions. The vision is to create a supportive network and inspire international collaborations.

The workshop will explore various aspects of machine learning using biomedical data with the hands-on practical projects providing the main focus, allowing participants to work in a team environment to understand how machine learning is applied to specific biomedical challenges.

For further details and registration check out the EUGLOH website

learn_bAIome offers workshops and trainings in biomedical AI/data science with tailored formats that take into account background, programming skills and intensity to provide unique, focused, and effective courses. These courses are free and open to students, clinicians, and researchers across academic institutions in Hamburg.

This workshop is open to students, researchers, and clinicians wanting to learn how machine learning is applied for biomedical datasets, the different classes of machine learning algorithms that may be used, as well as the best practices in selecting and evaluating algorithms, and their limitations.  The aim of the course is to provide concepts and tools to navigate the use of machine learning in the biomedical landscape. The course will use biological datasets and there will be hands-on components as well as discussions. Participants should already have taken an introduction to machine learning and be familiar with Python programming. The workshop will be in presence and therefore each participant should bring their own laptop (no ipads).

Topics

• Taxonomy of machine learning algorithms
• Linear regression, logistic regression and related methods
• Decision trees
• Support Vector Machines
• Bias & Variance, curse of dimensionality
• Representation learning
• Neural networks and deep learning: MLPs, transformers, CNNs
• Applications to RNAseq and imaging data

Whether you have a background in computer science, or biomedical-related fields, this Mini Symposium is designed to be very inclusive for students, researchers, and clinicians working and/or interested in biomedical AI. Our invited speakers will provide engaging presentations that introduce core ideas in AI and demonstrate how they are being applied to solve challenges in biomedicine. There will be opportunities for short lightning talks, Q&A, and networking, all in a fun and relaxed atmosphere with drinks and snacks provided. Please register for this event here.

Invited speakers Fabian Kern (UHH) and Fatemeh Haedaghi (UKE) will give talks on new developments in their research. See titles and abstracts here.

Lightning talks: Interested in giving a short talk about your current research? Let us know in your registration. There will be a number of short talks from selected topics to give exciting impulses of cutting-edge research as well as plenty of time and encouragement of audience participation.

Topics of interest:

• Application of AI in diagnostics (omics, pathology, radiology)
• Predictive modeling for patient outcomes
• Drug discovery and personalized medicine
• Foundation models in biomedicine
• Federated learning and data privacy in biomedicine
• Explainable AI and model interpretability

Abstract: Modern software development has embraced the concept of "code reuse," which is the practice of relying on third-party code to avoid "reinventing the wheel" (and rightly so). While this practice saves developers time and effort, it also creates liabilities: the resulting app may behave in ways that the app developer does not anticipate. This can cause very serious issues for privacy compliance: while an app developer did not write all of the code in their app, they are nonetheless responsible for it. In this talk, I will present research that my group has conducted to automatically examine the privacy behaviors of mobile apps vis-à-vis their compliance with privacy regulations. Using analysis tools that we developed and commercialized (as AppCensus, Inc.), we have performed dynamic analysis on hundreds of thousands of the most popular Android apps to examine what data they access, with whom they share it, and how these practices comport with various privacy regulations, app privacy policies, and platform policies. We find that while potential violations abound, many of the issues appear to be due to the (mis)use of third-party SDKs (i.e., supply chain problems). I will provide an account of the most common types of privacy and security issues that we observe and how app developers can better identify these issues prior to releasing their apps.

Bio: Serge Egelman

We are excited to announce a guest talk by Alexandra Diehl from the University of Zurich, who will present her latest research on the visualization and communication of weather forecasts. Alexandra is a senior researcher and lecturer in the Multimedia and Visualization group at UZH's Department of Computer Science. Her work focuses on efficient analysis, decision-making, and communication of high-impact weather events (HIWE) and their associated risks.

Talk Title: Visualization Research for Weather Forecast Communication and Analysis

Abstract: This talk will cover Alexandra's recent contributions to developing efficient visualization tools for analyzing and communicating weather forecasts and characterizing HIWEs. Additionally, she will discuss her current efforts in citizen data analysis and the challenges of effectively communicating severe weather events through participatory citizen science.

Speaker Bio: Alexandra Diehl holds a Ph.D. in Computer Science from the University of Buenos Aires and has extensive experience in data visualization, visual analytics, and geographic information systems. She has been a postdoctoral researcher in the Data Visualization and Analysis Group at the University of Konstanz and currently works with Prof. Dr. Renato Pajarola's group at UZH.

We look forward to seeing many of you at this insightful talk.