-
Towards Automated Semantic Segmentation in Mammography Images
Authors:
Cesar A. Sierra-Franco,
Jan Hurtado,
Victor de A. Thomaz,
Leonardo C. da Cruz,
Santiago V. Silva,
Alberto B. Raposo
Abstract:
Mammography images are widely used to detect non-palpable breast lesions or nodules, preventing cancer and providing the opportunity to plan interventions when necessary. The identification of some structures of interest is essential to make a diagnosis and evaluate image adequacy. Thus, computer-aided detection systems can be helpful in assisting medical interpretation by automatically segmenting…
▽ More
Mammography images are widely used to detect non-palpable breast lesions or nodules, preventing cancer and providing the opportunity to plan interventions when necessary. The identification of some structures of interest is essential to make a diagnosis and evaluate image adequacy. Thus, computer-aided detection systems can be helpful in assisting medical interpretation by automatically segmenting these landmark structures. In this paper, we propose a deep learning-based framework for the segmentation of the nipple, the pectoral muscle, the fibroglandular tissue, and the fatty tissue on standard-view mammography images. We introduce a large private segmentation dataset and extensive experiments considering different deep-learning model architectures. Our experiments demonstrate accurate segmentation performance on variate and challenging cases, showing that this framework can be integrated into clinical practice.
△ Less
Submitted 18 July, 2023;
originally announced July 2023.
-
GRSUS: Gerenciamento De Recursos De Saúde, Um Estudo Sob A Ótica Da Portaria GM/MS 1631/2015 No Estado do Pará
Authors:
Paulo Sérgio Viegas Bernardino da Silva,
Lucas Vinícius Araújo Caldas,
Antônio Fernando Lavareda Jacob Junior,
Fábio Manoel França Lobato
Abstract:
Investments in public health had an increase of about R$ 20 bi in recent years. Even with the dynamism of the Unique Health System (SUS), only after 13 years the criteria and parameters for the planning and programming of health services have been updated. The calculations for health resources division are complex due to the nature of the SUS administrative organization, which has three administra…
▽ More
Investments in public health had an increase of about R$ 20 bi in recent years. Even with the dynamism of the Unique Health System (SUS), only after 13 years the criteria and parameters for the planning and programming of health services have been updated. The calculations for health resources division are complex due to the nature of the SUS administrative organization, which has three administrative levels. Despite providing the criteria and parameters for the calculations, it was not provided any information system that would automate this process and provide reliable information for decision making. In order to fill such gap, this paper presents a system for health resource management from the perspective of GM/MS 1631/2015 ordinance. The tool has been validated using as case studies two municipalities in the interior of the state of Pará. The results were promising, with latent market potential, being possible to simulate various scenarios for a medium and long-term predictions.
△ Less
Submitted 26 November, 2018;
originally announced November 2018.
-
Area and Throughput Trade-Offs in the Design of Pipelined Discrete Wavelet Transform Architectures
Authors:
Sandro V. Silva,
Sergio Bampi
Abstract:
The JPEG2000 standard defines the discrete wavelet transform (DWT) as a linear space-to-frequency transform of the image domain in an irreversible compression. This irreversible discrete wavelet transform is implemented by FIR filter using 9/7 Daubechies coefficients or a lifting scheme of factorizated coefficients from 9/7 Daubechies coefficients. This work investigates the tradeoffs between ar…
▽ More
The JPEG2000 standard defines the discrete wavelet transform (DWT) as a linear space-to-frequency transform of the image domain in an irreversible compression. This irreversible discrete wavelet transform is implemented by FIR filter using 9/7 Daubechies coefficients or a lifting scheme of factorizated coefficients from 9/7 Daubechies coefficients. This work investigates the tradeoffs between area, power and data throughput (or operating frequency) of several implementations of the Discrete Wavelet Transform using the lifting scheme in various pipeline designs. This paper shows the results of five different architectures synthesized and simulated in FPGAs. It concludes that the descriptions with pipelined operators provide the best area-power-operating frequency trade-off over non-pipelined operators descriptions. Those descriptions require around 40% more hardware to increase the maximum operating frequency up to 100% and reduce power consumption to less than 50%. Starting from behavioral HDL descriptions provide the best area-power-operating frequency trade-off, improving hardware cost and maximum operating frequency around 30% in comparison to structural descriptions for the same power requirement.
△ Less
Submitted 25 October, 2007;
originally announced October 2007.
