Treating HIV with Smart Algorithms
Medical, Health Informatics and Computational Biology Accomplishment | 2008
Where the work was done: IBM Haifa Research Lab
What we accomplished: The EuResist Network GEIE (formed by MPG, Informa, KI, UniSiena and UniKoeln), together with IBM, developed a Web server that predicts the effectiveness of anti-HIV therapies on the basis of viral genotype and clinical information. The EuResist prediction system helps clinicians choose the best drugs and drug combinations for any given HIV genetic variant. It is trained on data from more than 65,000 individuals with AIDs treated in ten different European countries. The system has performed better than nine out of ten expert physicians in a study of the ability to predict the outcome of a given treatment (HIV Medicine Journal). This work is an ongoing effort where the recommendation system is updated as more data are accumulated and new drugs become available.
- Computerworld Award for this Work
- Wikipedia page
- Movie (English, French) and Customer case studies.
- 12+ publications since 2008
- Key publications:
- M. Zazzi, F. Incardona, M. Rosen-Zvi, M. Prosperi, T. Lengauer, A. Altmann, A. Sonnerborg, T. Lavee, E. Schülter, and R. Kaiser (2012). Predicting Response to Antiretroviral Treatment by Machine Learning: The EuResist Project. Intervirology, 55 pp.123-127
- M. Zazzi, R. Kaiser, A. Sonnerborg, D. Struck, A. Altmann, M. Prosperi, M. Rosen-Zvi, A. Petroczi, Y. Peres, E. Schulter, C. Boucher, F. Brun-Vezinet, R. Harigan, L. Morris, M. Obermeier, C. F. Perno, R. Shafer, A. Vandamme, K. van Laethem, A. Wensing, T. Lengauer, F. Incardona (2010). Prediction of Response to Antiretroviral Therapy by Human Experts and by the EuResist Data-Driven Expert System (the EVE Study). HIV Medicine, 2010
- M. CF Prosperi, A. Altmann, M. Rosen-Zvi, E. Aharoni, G. Borgulya, F. Bazso, A. Sonnerborg, Y. Peres, E. Schuflter, D. Struck, G. Ulivi, F. Incardona, A.-M. Vandamme, J. Vercauteren and M. Zazzi for the EuResist and Virolab study groups, (2009) Investigation of Expert Rule Bases, Logistic Regression, and Non-Linear Machine Learning Techniques for Predicting Response to Antiretroviral Treatment, Antiviral Therapy, 14: 433-442
- Altmann, M. Rosen-Zvi, M. Prosperi, E. Aharoni, H. Neuvirth, E. Schulter, J. Buch, D. Struck, Y. Peres, F. Incardona, A. Sonnerborg, R. Kaiser, M. Zazzi, T. Lengauer(2008) Comparison of Classifier Fusion Methods for Predicting Response to Anti HIV-1 Therapy, PLoS ONE 3(10)
M. Rosen-Zvi, A. Altmann, M. Prosperi, E. Aharoni, H. Neuvirth, A. Sonnerborg, E. Shulter, D. Struck, Y. Peres, F. Incardona, R. Kaiser, M. Zazzi, T. Lengauer (2008), Selecting anti-HIV therapies based on a variety of genomic and clinical factors, ISMB conference/ bioinformatics journal 2008.
- New work
- Tal El-Hay, Omer Weissbrod, Elad Eban, Maurizio, Francesca Incardona (2014), Structured Proportional Jump Processes. UAI2014
- Gregson J. et al, Global epidemiology of drug resistance following failure of WHO recommended first line regimens for adult HIV-1 infection - an international collaborative study, Lancet Infectious Diseases, Jan 2016.
Image credit: euresist