Aplicação de beamformer eigenspace com técnicas adaptativas e filtros para formação de imagens de ultrassom de elastografia com ondas planas

• Ano de defesa: 2026
• Autor(a) principal: Neves, Larissa Comar
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Tecnológica Federal do Paraná; Curitiba; Brasil; Programa de Pós-Graduação em Engenharia Elétrica e Informática Industrial; UTFPR
• Palavras-chave em Português: Diagnóstico por ultrassom; Ultrassonografia; Reconstrução de imagens; Processamento de imagens - Técnicas digitais; Filtros adaptativos; Diagnóstico por imagem; Diagnostic ultrasonic imaging; Ultrasonic imaging; Image reconstruction; Image processing - Digital techniques; Adaptative filters; Diagnostic imaging; CNPQ::ENGENHARIAS::ENGENHARIA BIOMEDICA::BIOENGENHARIA::PROCESSAMENTO DE SINAIS BIOLOGICOS; Engenharia Elétrica
• Link de acesso: http://repositorio.utfpr.edu.br/jspui/handle/1/40749

Ultrasound devices stand out for their practicality and safety in assisting medical diagnosis, and new approaches are increasingly being sought to improve image quality. Among the applications of ultrasound imaging systems is the study of tissue elasticity (elastography), which can help detect and diagnose diseases such as cancer. To study tissues, it is necessary to use high-frequency and plane-wave techniques, which enable higher frame rates and are valuable tools for elastography. Therefore, new adaptive beamformers for Plane Wave Imaging have emerged, enabling improved contrast and resolution. The use of the Minimum Variance-based Eigenspace Beamformer (EBMV) with the Generalized Sidelobe Canceler (GSC) reduces interference and noise, yielding images with higher contrast and resolution than those from traditional Delay-and-Sum (DAS) and Minimum Variance (MV) methods. In addition, image-processing filters are applied to reduce speckle noise in ultrasound images. In this work, the EBMV-GSC beamformer was combined with Wiener, Lee, and Kuan filters, and their enhanced versions, to improve plane-wave ultrasound image processing for elastography applications. In B-mode results, the Wiener-filter-based methods showed the best performance in contrast and image quality, especially the EBGSC-We method, which improved contrast by up to 74 % compared to the DAS method and 60 % compared to the EBGSC method. On the other hand, methods based on Lee and Kuan filters showed greater reductions in geometric distortions and sidelobes. In the elastography results, the EBGSC-W and EBGSC-We techniques yielded velocity estimates closer to the expected theoretical values than the EBGSC techniques combined with Lee and Kuan filters, and showed greater stability in shear wave tracking. The EBGSC-We technique stood out by achieving a percentage error of approximately 1.2 %, resulting in a reduction of more than 80 % compared to the DAS method. Furthermore, in the strain map analysis, the EBGSC-W and EBGSC-We techniques showed gains of up to 19% and 16 % in contrast (CR), 24 % and 18 % in CNR, and 10 % and 8 % in SNR compared to the DAS method, respectively, indicating better preservation of image contrast and signal-to-noise ratio in ultrasound images. Overall, the results indicate that the combination of the EBMV-GSC beamformer with Wiener-based filters has strong potential for ultrasound elastography applications, improving image quality and reducing shear-wave velocity estimation errors.