General
publicado em 15/06/2011 às 18h00:00
   Dê o seu voto:

New algorithm can give more precision to the diagnosis of breast cancer

Brazilian engineers developed mathematical algorithm that promises to give new impetus to the diagnosis of disease

 
font size
A-
A+

Some pieces of information on this page may have been automatically translated. Makernews is not responsible for the irregularities resulting from these translations. When in doubt,      consult the original text.



A mathematical algorithm of image processing developed by two engineers from the Brazilian Center for Physics Research (CBPF), siblings and Marcelo Márcio Portes de Albuquerque, promises to give new impetus to the diagnosis of breast cancer. The new approach, described in an article published by researchers in 2004, was tested at the end of last year by three researchers at the Indian Institute of Technology Kanpur and the results were encouraging as compared to the commonly used techniques. The accuracy rate of positive results rose considerably. Since false positive tests when the outcome appears mistakenly as positive plummeted. "The reliability index of the positive results of the current jumped 80.2% to a rate very close to absolute certainty: 96.5%," said Márcio e Marcelo.

"Since the false positives fell from 8.1% to almost absolute zero, only 0.4%," complement.

The procedure, which filters the relevant information that appear as white spots on mammography X-rays used to diagnose the disease helps to identify the tiny stones called microcalcifications resulting from the accumulation of calcium. "In 20% of cases, irregular contours on the microcalcifications can be caused by tumors," say the engineers. Described by the two researchers, the technique is to use for processing images from X-rays, the so-called non-additive entropy, the theory formulated by another researcher CBPF: physicist Constantino Tsallis, which is the Scientist of Our State and FAPERJ coordinator of the National Institute of Science and Technology Complex Systems (INCCT-SC), co-financed by CNPq and FAPERJ.

The theory, also called Tsallis statistics, is to understand the behavior of complex systems like the human body, economics and linguistics, among others. The physical didactically explains how his theory, illustrating with the air mass present in a room. "To move, the air molecules can take thousands of these behaviors, how to go forward, backward, left, right, up, down, and so on," teaches Tsallis. According to the researcher, these molecules are called simple systems, and the right to use all possibilities of movement will always be exercised. Already in complex systems, like the human brain and its thousands of neurons, or even the stock market, income distribution, population and language, the number of possibilities is restricted and therefore lower. "They are very selective and always restrict their behavior. I try to identify these opportunities, which is not always easy," he adds.

"We have a broad debate about innovation," Marcelo argues

Constantino Tsallis appears to be quite excited about using his theory to image processing and, more specifically, to improve the diagnosis of breast cancer. He explains the importance of the new technique. "When there is suspicion of breast cancer, the first examination that the gynecologist asks is this the white dots. He looks and, if required, ask for a specific examination of microcalcifications," explains Tsallis.

"But as we saw with the technique developed by engineers and Marcelo Márcio, the reliability index of examinations has increased considerably, going from 80.2% certainty in the case of positive results for 96.5%," he adds.

Engineers and Marcelo Márcio now want to repeat the experiment carried out by Indians along the Brazilian physicians. For this, they hope to draw support from a national and multinational companies interested in joining this endeavor. Researchers from India used as bases algorithms with specific images for characterization techniques. "We needed, through an association with companies in the health sector, to invest in the characterization of the algorithm for a large image database. This technology could be effectively transferred to the companies that could exploit it commercially in existing equipment," said Marcio. "We're trying to use existing incentives for technological innovation to come in contact with companies interested in innovating and helping us in this endeavor," he adds.

The fact illustrates the need of federal, state and municipal companies mobilize to steadily strengthen the innovation environment in the country "Both universities need to approach businesses and vice versa, by hiring researchers or the real incentive to research and development, "says Marcelo. "Currently, all spheres of government are encouraging the creation of areas for research and development companies in order to strengthen the partnership with research institutions in Brazil," he concludes.

A physicist of international renown

Born in Greece in 1943, Constantino Tsallis was four years old when, led by the family emigrated to Brazil. After a brief stay here, the family moved to Mendoza in Argentina, where he completed his primary and secondary studies. Graduate of the Institute of Physics in Bariloche, in 1965, went to Europe the following year, after obtaining a scholarship from the French government. Over nearly a decade of residence in France, regularly taught Physics at the University of Paris and the Ecole de Physique et Chimie de Paris. In 1974 he received the degree of Docteur d'Etat en Sciences physiques from the University of Paris-Orsay for research in the Theory of Phase Transitions. In 1975 he emigrated to Brazil around the country to which he has sought and obtained Brazilian citizenship in 1984. In 1988 he made his most famous theory, the non-additive entropy, which tries to explain the behavior of complex systems, which bears his name.

Because of its scientific production, has been cited by name (his name in the title or abstract of the publication) in more than one thousand four hundred articles of other researchers in over 60 countries that made the approach in Brazil, this indicator, the famous health physician Carlos Chagas, the discoverer of the disease which bears his name. The numbers that reflect its activities are superlatives. He has published over 300 works which received more than ten thousand citations in the literature internationally, and delivered more than 800 conferences and events and institutions of 40 countries.

In May this year, Constantino Tsallis received an invitation from the Vatican Academy of Sciences to lecture in October. It is the first time in history that a Brazilian physicist is invited to perform at the institution. Tsallis minister to the conference "From statistical physics to quantum mechanics non-linear." Besides him, the event will still have the physical presence of Carlo Rubia and Gerardus' t Hooft, both Nobel Prize winners in Physics.

Japan

Source: FAPERJ
   Palavras-chave:   Cancer diagnosis algorithm    FAPERJ   
  • Share this pageShare this page
  • Share this pageCorrect
  • ShareShare
  • AlertAlert
Reduced link: 
  • You are recommending this story: New algorithm can give more precision to the diagnosis of breast cancer
  • Fill in the following form to send your recommendation to your friend:

  • You are suggesting a correction for this story: New algorithm can give more precision to the diagnosis of breast cancer


Receba notícias do iSaúde no seu e-mail de acordo com os assuntos de seu interesse.
Seu nome:
Seu email:
Desejo receber um alerta com estes assuntos:
Cancer diagnosis algorithm    FAPERJ   
Comments:
Comment
Leave your comment
Close
(Required fields are marked with an *)

(Your email address will never be published or shared.)

Enter the letters and numbers below and click in the button "send"

  • Twitter iSaúde
advertising
Informe Saúde printed version

Recommend the portal
Close [X]
  • You are recommending this story: http://www.isaude.net
  • Fill in the following form to send your recommendation to your friend:

RSS news from the portal  iSaúde.net
Get the newsletter of the portal  iSaúde.net
Recommend the portal iSaúde.net
News from  iSaúde.net in your blog or website.
Get news on the subject of your interest.
© 2000-2011 www.isaude.net Todos os direitos reservados.