Advanced searches left 3/3

Lyme Disease - Springer Nature

Summarized by Plex Scholar
Last Updated: 15 November 2022

* If you want to update the article please login/register

Effect of Small Dataset Quality on Deep Neural Network Performance for Lyme Disease Classification

Yuri Stirenko, Sergii Oholtsov, Illia Gordienko, Sergii Ograny, Sergii Oholtsov, Illia Gordienko, Yuri Stirenko, Sergii Oholtsov, Obteo, Sergii is a student at the University of Napier, Oholtsov The classification task is considered for a series of problems where images of Lyme disease symptoms were collected and organized in the open-access database with the various subsets of the whole database: clean images only, dirty images only, and clean and dirty images in the same proportions were considered. Two new data preprocessing techniques were used: normalization without data conversion and normalization with data increments. The weights were obtained after training on the ImageNet database, and two versions of the latest NASNet DNN architectures of various sizes were used, namely NASNetLarge and NASNetMobile. But the results of the combined testing of the new unknown subset of Lyme disease images with the same number of clean and dirty photos produced very different results.

Source link: https://doi.org/10.1007/978-981-19-3590-9_44

* Please keep in mind that all text is summarized by machine, we do not bear any responsibility, and you should always check original source before taking any actions

* Please keep in mind that all text is summarized by machine, we do not bear any responsibility, and you should always check original source before taking any actions