Research Projects (Third party funds)


Machine learning methods for clinical data analysis

Machine learning has become a powerful tool to mine for patterns in biological and clinical data, disease diagnosis, assist drug discovery and predict treatment outcomes. In collaboration with a wide range of clinical partners, we have analyzed and processed heterogeneous clinical data (images, laboratory results, medical records, etc.) and implemented supervised and unsupervised machine learning methods to improve clinical decision making or biomarker discovery. 

As a recent example, we have worked on a large cohort of chronically infected Hepatitis B patients treated with different antiviral treatments to identify new biomarkers of virological relapse after treatment discontinuation. The data set available for this patient cohort consists of different time-resolved measurements of 47 cytokines during treatment. We developed a supervised machine learning method that allowed for the first time to reveal the predictive potential of cytokines for Hepatitis B virological relapse. We identified a small subset of five cytokines able to predict relapse with a high accuracy at any time during treatment. This facilitates clinical decisions on the management of Hepatitis B patients, potentiality avoiding over or under-treatment .

Simm members 

Haralampos Hatzikirou, Andreas Reppas (former member), Sahamoddin Khailaie, Michael Meyer-Hermann

hzi collaborators

Frank Pessler, Mark Brönstrup

other collaborators

Renata Stripecke, Markus Cornberg (Medizinische Hochschule Hannover), Michael Platten (DKFZ)


Theobald SJ, Khailaie S, Meyer-Hermann M, Volk V, Olbrich H, Danisch S, Gerasch L, Schneider A, Sinzger C, Schaudien D, Lienenklaus S, Riese P, Guzman CA, Figueiredo C, von Kaisenberg C, Spineli LM, Glaesener S, Meyer-Bahlburg A, Ganser A, Schmitt M, Mach M, Messerle M, Stripecke R. Multidimensional signatures of T cell maturation, PD-1 upregulation and B cell class-switching after HCMV latent infections and reactivations in humanized mice. Front Immunol 2018 Nov 22; 9: 2734.

Volk* V, Reppas* AI, Robert PA, Spineli LM, Sundarasetty BS, Theobald SJ, Schneider A, Gerasch L, Deves-Roth C, Klöß U, von Kaisenberg C, Figueiredo C, Hatzikirou° H, Meyer-Hermann° M, Stripecke° R. Multidimensional analysis integrating human T-cell signatures in lymphatic tissues with sex of humanized mice for prediction of responses after dendritic cell immunization. Front Immunol 8 (2017) 1709. [*shared first and °shared corresponding authors].





Funding agency


PrintSend per emailShare