Representing Patients and Predicting Clinical Outcomes with Large Language Models

Speaker: Alexander Loeser, BHT

Medical professionals are faced with a large amount of textual patient information every day. Clinical decision support systems (CDSS) aim to help clinicians in the process of decision-making based on such data. We specifically look at a sub-task of CDSS, namely the prediction of clinical diagnosis from patient admission notes. When clinicians approach the task of diagnosis prediction, they usually take similar patients into account (from their own experience, clinic databases or by talking to their colleagues) who presented with typical or atypical signs of a disease. They then compare the patient at hand with these previous encounters and determine the patient’s risk of having the same condition.   I will review our work on this set of tasks over the last three years and will briefly introduce Large Language Model architectures, data sets and lessons learned from papers published at EACL’21, COLING 22, IJCNP 22 and LREC 22. 

Lecture is open to the public and starts at 4:00 pm. Location: ECDF, conference room, 1st floor (Wilhelmstraße 67, 10117 Berlin).

The event will take take place in hybrid mode. Registration by e-mail is requested.

About the speaker:

Prof. Dr.-Ing. habil. Alexander Löser conducts research and teaching in the areas of Data Science and Text-based Information Systems at Berlin University of Applied Sciences  since September 2013. He is also speaker  of the Data Science Research Center,  founded the international Master’s program “Data Science” at the University and is expert for LLMs at the BMBF Platform “Lernende Systeme”.  

Alexander’s research interests lie at the intersection of natural language processing and machine learning. He published over 90 refereed scientific papers in prestigious international conferences and journals, including ACM CIKM, ACM TheWebConference, TACL, EACL, COLING, Elsevier Information Systems or ISWC. Alexander serves on the editorial boards and program committees of top international journals or conferences in the areas of data science and computational linguistics. In 2019, he received the best paper award at IEEE BigComp 2019.

Alexander has a well-established track record of innovation and technology transfer. Previous stations are HP Labs Bristol, IBM Almaden Research Center, the research division of the SAP AG and the Technische Universität Berlin. He gave more than 120 invited talks and holds 3 granted patents. His research has been incorporated into commercial products, such as IBM Lotus Notes and SAP HANA.  He also worked as an independent consultant with eBay, Zalando, MunichRe, SpringerNature, Babbel, Fresenius, Krohne, Babbel and as an independent expert with the European Union, among others. Over the time he helped these organizations to create six data platforms with more than 50 data products.