May 7, 2008

Neural networks helping medicine

Some scientists of the Saint Raphael Institute of Milan have discovered an useful software for the analysis of proteins and very important in the diagnosis of some pathologies of the immune system. The new program, called FuzzyLab, concurs to obtain better results and to pull down of 80% the times of examination, to all advantage of doctors but also to patients. FuzzyLab is a software neuro-fuzzy, it means that is based on a combination of neural networks and fuzzy logic.

Before we begin, let's discuss what we hope you will learn through this article. Then we can begin to piece it together for you.


The neural networks are parallel to an artificial head in a place to continuously improving the own ability of learning. Fuzzy logic renders the processes of calculation more bendable. This typeface of logic is practical already from some year to many harvest of expansive consumption for example it allows the washing equipment of last generation to determine out water and detergent improperd on the plane of dirt in laundry. Fuzzylab carries out promptly all the automated job of interpretation of the analyses awaiting nowadays executed guidely and entrusted to the interpretation of the doctor.


"In a laboratory of great dimensions" - explains Stefania Del Rosso, the responsible of the Laboraf-consider laboratory of Saint Raphael Institute- "350 analyses are made every day in typical, whose recital strain to an practiced doctor at slightest two hours of job. In order to do the same mission the software employs only 20 notes. In this way the doctor can distinguish promptly between regular "and pathological outcomes" and commit then the saved time to the investigation of examinations more complex to understand ".


The vessel analysis considers proteins of the serum and, afar providing common indications about inflammatory states and promising hepatic torment, it can tell the company of anomalies in the immune system. Is the location for example of the monoclonal members - a group of even antibodies whose beginning is correlated to cruel lymphatic pathologies like the skin tumour, the continual lymphatic leukaemia and the lymphomas.


From this point forward, we will let you in on little secrets that will help you implement this subject into your life.


In order to learn to read the analyses, FuzzyLab has been subordinate to an hard teaching articulated in three moments. In the first part the software has been instructed to concede the mold of the regular analytical curves through a mathematical algorithm, on the improper of the facts complete from the doctor "guide". In the following part have been inserted in the mainframe the mathematical formulas for the identification of the pathological curves. In the third part (validation of the system) the guide interpretation and that one of FuzzyLab have been put to comparison in order to valuation of the slip margin, resulted poorer to 2%.


The system is awfully bendable and potentially in a place to increasing just the modernization with the experience. "The advantages of this attempt are important both for the doctor and for the patient", says Michelangelo Murone, junior executive of Laboraf. "The most clear one is the decrease of the times of coverage that allows the analyst to have more time displace in order to consider the pathological tracings. The following advantage is the improvement of the interpretative attribute, translated in a surer coverage, because the pathological curves are analyzed at slightest two times, before from the software and then from the doctor, but also more objective and standardized, because Fuzzylab reduces the probability of mistakes and deletes the hazard of interpretative differences due to the person cause "


About me: Italian scholar. Information technology and communication specialist. grade in knowledge of Communication. education management and sphere intelligence practiced.


If we have failed to answer all of your questions, be sure to check into other resources on this interesting topic.

0 comments: