Cancer Diagnosis via Machine Learning Modeling of Genetic Data
>>> Basic concepts of genetic modifications characteristic to cancer cells and how they can be used to try and predict tumor grade, patient survival, reaction to certain treatment.
>>> Intro to special machine learning techniques, that are focused not only on accuracy, but on revealing insights on which genes are involved in cancer progression – aka model interpretability.
>>> How correlated features hinder model interpretability?
>>> Challenges of introducing AI models in cancer treatment.
LECTOR: Laura Tolosi
>>> Laura Tolosi is a Data Scientist with more than 10 years of experience in various domains.
>>> She has a PhD in Computational Bioinformatics from the Max-Planck-Institute in Saarbruecken. The doctoral thesis is concerned with applying machine learning to genetic data in order to help improve cancer treatment.
>>> She worked closely with cancer clinics in Germany, and her models have revealed insights into the progression of Neuroblastoma tumors.
>>> In Sofia, she became an expert in the field of Natural Language Processing as a lead data scientist in Ontotext. She worked on European projects such as Pheme, developing algorithms for rumor detection in Social Media.
>>> She conducted a semantic and sentiment analysis on Twitter data preceding the Brexit referendum and concluded that the „Leave“ campaign has a louder voice on Social Media.
>>> Currently, she is interested in Reinforcement Learning and is working on a personal project.
WITH THE SUPPORT OF: HyperScience & Skyscanner & Factset
HyperScience brings AI to the office. Through ML solutions we help various enterprises and government institutions to automate the hectic back office work.
FactSet (NYSE:FDS | NASDAQ:FDS) delivers superior analytics, service, content, and technology to help more than 88,000 users see and seize opportunity sooner.
MANDATORY FREE REGISTRATION:
The event is free of charge but with mandatory pre-registration: here