The project, EuResist for the development of an integrated system for the clinical management of the use of drugs to prevent the replication of the AIDS virus, given to carriers and patients with the disease
Researchers at the IBM research laboratory in Haifa are leading a pan-European EuResist project to develop an integrated system for clinical management of the use of drugs to prevent the replication of the AIDS virus, given to carriers and patients with the disease.
The computerized system, which they are developing, will make it possible to predict in advance the specific response of each patient to treatment, in order to choose the most correct combination of drugs for each genetic variation of the virus. Successful implementation of the system will improve the effectiveness of AIDS treatment.
Infection with the AIDS virus is no longer a death sentence - as more and more treatments and drugs are approved by the health authorities after they have proven their success in preventing the multiplication and replication of the virus in the body. But the virus is developing resistance to a growing number of drugs. The treatment of an increasing number of patients fails, after the virus develops resistance to the drug that is supposed to prevent its replication processes. In order to deal with the disease, each patient receives a "cocktail" of drugs whose composition changes as the virus they carry develops resistance to one of the drugs. The doctors treating AIDS today need a system that will help them determine which of the cocktails have a higher chance of success, and adapt these cocktails to each patient, according to the genetic mutation of the virus that he developed.
Boaz Carmeli, director of life sciences at the IBM research laboratory in Haifa, noted that "monitoring the history of the treatments and the development of the virus itself become essential in the treatment process. Access to the accumulated knowledge from huge amounts of data collected during the treatment of many AIDS patients will allow the attending physician to base his treatment decisions both on the patient's history - and on extensive information about the virus, mutations of the virus and the stage of the disease in which the patient is."
The EuResist project managed by the European community uses an innovative approach and tools developed at IBM in order to predict in advance the effectiveness of antiviral drugs and treatments against well-defined types of the AIDS virus, with a European-wide analysis of the HIV database across the continent. The main databases integrated into this comprehensive database have so far been managed separately in Italy, Germany and Sweden. The unified database is expected to be the largest of its kind in the entire world, and it already encompasses over 17,000 patients.
The project is carried out as part of the sixth framework program for research and development of the European community, with the participation of the IBM research laboratory in Haifa, the University of Siena in Italy, the Karolinska Institute in Sweden, the Max Planck Institute and the University of Cologne in Germany, RMKI from Hungary, Kingston University in the United Kingdom, and the European Pharmaceutical Companies Association (EFPIA).
IBM's research laboratory in Haifa has a long history of leading and promoting standards and innovations in various industrial fields. The work in the laboratories as part of the new project began with the definition of a uniform architecture for clinical medical documents, which will allow communication and transfer of information between medical institutions treating AIDS patients in different countries, using a uniform format in the XML environment. The unification of the data is a necessary condition for searching and locating information, in the consolidated record collection. As the use of the new format expands to more institutions, more doctors and research institutes will be able to access the data, and use it for treatment tasks and new drug development.
According to Boaz Carmeli, "a uniform standard for medical records in the treatment of AIDS will allow different medical bodies to share data and information - and will pave the way for international cooperation in the fight against the disease."
The European project also includes the development of several calculation engines, which will allow predicting the behavior of the virus and its response to various treatments. The various forecasting engines include models for assessing the development of the disease, theoretical models, dedicated systems for analyzing data warehouses and data mining, systems for analyzing disease cases, and machine learning systems that improve the answers provided by the various forecasting tools.
The prediction engines developed in the IBM research laboratory in Haifa are based on machine learning technologies, in which the laboratory personnel have developed long-standing expertise.
Shay Payne, director of the machine learning group at IBM's research laboratory, stated that "these systems combine the prediction obtained from a large number of different engines - and present the doctor with one integrated result."
"The challenge is to understand and predict - automatically - which of the engines is more suitable for each individual patient, in different development scenarios." Already today, a machine learning system knows how to produce more accurate results than any prediction engine operating on its own, Shay Payne added.
The results of the project so far indicate a particularly high rate of success, of 75%. The theoretical prediction provided by the system is currently being tested and compared to the data obtained in actual treatments.
One response
Important important important! can relieve a lot of people who are dealing with quite a few…