RESULT ANALYSIS OF COVID19 DISEASE DIAGNOSIS USING HOPFIELD AND HEMMING NEURAL NETWORK MODELS
Основное содержимое статьи
Аннотация
Currently, COVID-19 is the most dangerous and deadly disease for the human body caused by the new coronavirus. In December 2019, the coronavirus spread rapidly around the world, probably from Wuhan, China, and caused a large number of deaths. Early detection of COVID-19 through proper diagnosis, especially in cases without obvious symptoms, can reduce patient mortality. Chest X-rays are mainly used to diagnose this condition. This study proposed an imaging technique for detecting COVID-19 from chest X-rays. A classification model was developed by training the features obtained using a neural network (Hopfield and Hemming). During the test phase, aggregated data were reviewed to evaluate model performance. Cross-validation analysis showed that Hopfield neural network with 80 percent recognition rate and Hemming neural network with 90 percent recognition rate showed better classification performance.
Информация о статье
Библиографические ссылки
Dj.Tu., Gonsales R. Printsipy raspoznovaniya obrazov. Perevod s angliyskogo I.B.Gurevicha. M.:Mir.1978 g. 412 s.
Berkinblit M.B. Neyronnye seti. Uchebnoye posobiye. M.: MIROS i VZMSh RAO, 1993. -96 s.
Neyronnye seti: istoriya razvitiya teorii. Kn. 5: Ucheb. posobiye dlya vuzov. / Pod obshey red. A.I. Galushkina, Ya.Z. Sypkina. - M.: IPRJR, 2001. - 840 s.
Kruglov V.V., Borisov V.V. Iskusstvennye neyronnye seti. Teoriya i praktika.- 2-e izd.-M.:Goryachaya liniya-Telekom. 2002 g. 382 s.
Kallan R. Osnovnye kontseptsii neyronnyx setey.: Per. S angl.-M.: Izdatelskiy dom “Vilyams”, 2001 g. 287 s.