IBM Research - Almaden

Assets

In 2010 number of staff was 400 researchers among whom there are also invited scientists.

At this time in laboratory there is a specialized scanning tunnel microscope which is used for movement and positioning of separate atoms.

Directions of researches

The main directions of researches in the center in 2010 were:

• fundamental sciences,
• nanotechnologies,
• computer sciences in areas of content management, programming, creation service - the focused information architectures.

The scientific and technological center develops new technologies which are critical for business of IBM and which allow to understand the main directions of development of science and technology more deeply. One of the priority directions are researches of chemical materials of semiconductors. Also researches of non-volatile storage cells on the basis of magnetic tunnelling junctions and organic electronics are conducted. In the research center scientists also carry out researches in the scientific areas directed to expansion of knowledge for the further growth of the main IT industries and creation of base for new IT paradigms.

History

2016: Data analysis from cash scanners for identification of sources of infections

The infectious diseases extending through a digestive tract such as salmonellosis, intestinal infections and intestinal flu, are a serious problem of health sector. Big Data and analytics can help to identify potential sources of an infection at the outbreaks of diseases of food origin.

According to the Centers for control and prevention of diseases of the USA (Centers for Disease Control and Prevention, CDC), for 2016 annually cases of infection are registered at every sixth American. During the outbreak of diseases of digestive tract fast identification of the infected products is vital. It helps to minimize risk of development of epidemic, to reduce losses and negative impact on society in general.

On September 8, 2016 it was announced that scientific divisions of IBM Research – Almaden in the American city of San Jose (State of California) found out that data processing from cash scanners of barcodes in grocery stores can accelerate detection of a source of an infection taking into account geography of identification of cases of diseases of a digestive tract. According to data of a research, scientists can limit in only several hours number of potentially infected products to 12 names, having only about ten reports on results of physical examination.

Within the research the staff of IBM Research created information processing methodology for hundreds of categories of foodstuff. She allows to consider spatio-temporal data, including a geographical location and approximate time of consumption. Having analyzed each product and term of its storage, scientists also revealed the place where it was eaten, and accumulation probability in it separate pathogenic microorganisms. This information was imposed on the card of the centers of an infection. Then a system created the rating of all products on the basis of the probability of infection from which doctors could select 12 most suspicious names of products, test them on presence of an infection and inform the population in case of threat.

Standard check takes weeks and months that can significantly affect economic and social effects of a disease outbreak. Usually process includes poll of citizens for tracking of a source of an infection. In 2011 60 days were required to reveal in Europe a source of outbreak of the infection caused by colibacillus – the imported fenugreek seeds. By the time of completion of investigation all sprouts of these seeds were eaten. According to the European supervisory authority on security of European Food Safety Authority foodstuff, about four thousand people in 16 countries got sick and more than 50 people died before the infection source was found.

"At the outbreak of the infectious diseases extending through a digestive tract, it is extremely important to doctors to reveal quickly a source of infection and to warn the population about threat, – Kang Hu, the researcher in the field of IBM Research health care – Almaden told. – Traditional methods, such as polls of the population and research, are still necessary, however processing of Big Data from cash scanners in grocery stores can reduce considerably in few hours the list of suspicious names of products for further laboratory testings. Our work shows that Big Data and analytics allow to reduce significantly time of check and influence of a human factor. Such method will be of great importance for health sector".

The method described in a research was applied at flash if earlier - infections in Norway. Already for a moment, when only 17 cases of infection were registered, doctors began to use new methodology for data processing from cash scanners in grocery stores. On the basis of data on more than 2,600 food products they selected ten most probable causative agents of an infection. The further laboratory analysis helped to find an infection source, to define a specific batch and the product serial number – to them there was a sausage.

Research 'From Farm to Fork: How Spatial-Temporal Data can Accelerate Foodborne Illness Investigation in a Global Food Supply Chain' was published in the Association for Computing Machinery's Sigspatial Journal magazine.

Relational DBMS

In the history of the research center there are a lot of significant opening, one of them — creation of a new concept of structuring information known as — relational databases.

1986: Opening of the center

The center was open in 1986.