Chest X-ray image segmentation method based on a dilated multiscale U-Net; Молодежь и современные информационные технологии

Bibliographic Details
Parent link:	Молодежь и современные информационные технологии.— 2026.— С. 284-287
Main Author:	Wang Yuqian
Other Authors:	Li Qilun (научный руководитель), Aksyonov S. V.
Summary:	This paper proposes a dilated multiscale U-Net for chest X ray image segmentation. By fusing multi scale contextual information and channel features via dilated convolutions and an improved inception module, the method enhances lesion detection and localization. Experiments on the ChestX ray14 subset demonstrate superior accuracy and efficiency Текстовый файл
Language:	English
Published:	2026
Subjects:	электронный ресурс труды учёных ТПУ image segmentation deep learning U-Ne multi-scale features dilated convolution chest radiography сегментация изображений глубокое обучение многомасштабные функции рентгенография
Online Access:	https://earchive.tpu.ru/handle/11683/139098
Format:	Electronic Book Chapter
KOHA link:	https://koha.lib.tpu.ru/cgi-bin/koha/opac-detail.pl?biblionumber=687387

Internet