An Energy-Efficient Programmable Manycore Accelerator for Personalized Biomedical Applications

Kulkarni, Adwaya; Page, Adam; Attaran, Nasrin; Jafari, Ali; Malik, Maria; Homayoun, Houman

doi:10.1109/tvlsi.2017.2754272

Cited by 19 publications

(9 citation statements)

References 57 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“… 171 FPGA implementation may, however, be problematic due to increased implementation complexity. Custom-designed accelerator chips 172 and Application-Specific Integrated Circuit (ASIC) 173 offer even higher performance but at even higher implementation complexity.…”

Section: Distributing the Computational Loadmentioning

confidence: 99%

Embedded deep learning in ophthalmology: making ophthalmic imaging smarter

et al. 2019

View full text Add to dashboard Cite

Deep learning has recently gained high interest in ophthalmology due to its ability to detect clinically significant features for diagnosis and prognosis. Despite these significant advances, little is known about the ability of various deep learning systems to be embedded within ophthalmic imaging devices, allowing automated image acquisition. In this work, we will review the existing and future directions for ‘active acquisition’–embedded deep learning, leading to as high-quality images with little intervention by the human operator. In clinical practice, the improved image quality should translate into more robust deep learning–based clinical diagnostics. Embedded deep learning will be enabled by the constantly improving hardware performance with low cost. We will briefly review possible computation methods in larger clinical systems. Briefly, they can be included in a three-layer framework composed of edge, fog, and cloud layers, the former being performed at a device level. Improved egde-layer performance via ‘active acquisition’ serves as an automatic data curation operator translating to better quality data in electronic health records, as well as on the cloud layer, for improved deep learning–based clinical data mining.

show abstract

Section: Distributing the Computational Loadmentioning

confidence: 99%

Embedded deep learning in ophthalmology: making ophthalmic imaging smarter

et al. 2019

View full text Add to dashboard Cite

show abstract

“…precisas y rápidas. En base lo anterior, en (Kulkarni et al, 2017) se propone un acelerador de múltiples puntos pequeño, eficiente en energía y específico de dominio, denominado nano clústeres de potencia eficiente (PENC), para mapear y ejecutar los núcleos de estas aplicaciones. Los resultados de la simulación muestran que el PENC puede reducir el consumo de energía hasta en un 80% y 25% para los kernels DSP y ML, respectivamente, cuando se paraleliza de manera óptima.…”

Section: Plataformas Heterogéneas De Dispositivos Paralelosunclassified

Implementación de Algoritmos de Procesamiento Digital de Señales en Hardware Paralelo: Artículo de revisión

Martínez

Aceves

Argüelles

et al. 2018

CULCyT

View full text Add to dashboard Cite

show abstract

“…Network on a chip which uses some processors for parallel calculation can be a suitable solution for the above. However, the metal connections ordinarily that are used in the network on the chip changed in limited bandwidth, long release delay, and high energy consumption [1][2][3]. Photonic network on chip is a hopeful solution to overcome these limitations.…”

Section: Introductionmentioning

confidence: 99%

Non-Blocking and Multi Wavelength Optical Router Design based on Mach-zehnder Interferometer in 3-D Optical Network on Chip

Asadinia

Meheabi

Yaghoubi

2021

mjee

View full text Add to dashboard Cite

Due to the increasing number of cores in a chip, electronic networks on chip cannot be an effective solution for using multi-core processors. The use of optical connection technology for networks on 3D chip is an ideal choice recommended in response to delay and reliability. In optical network architecture on 3D chip, 7-port routers are used to data transfer. The structure of the optical routers in the network on the 3D optical chip affects the Performance transmission of the entire network so that the provision of optical router with minimal loss and power consumption is considered by researchers in this domain. In this study, a seven-port Non-blocking optical router based on the Mach Zehnder interferometer optical switch in the network on a 3-D optical chip is presented. This router consists of 18 Mach Zehnder interferometer optical switches that can transfer multiple wavelengths concurrently. To evaluate the proposed 7-port router in the network on the 3D optical chip, the parameters of insertion loss, bandwidth density and power consumption are considered and the simulation results represent that this router decreases the loss as much as possible and improves the use of the wavelength channel comparing to available router and has the ability to transfer 4 wavelengths simultaneously with a wavelength range of 1525-1565 nm and a data transfer rate of 20Gbps for all 42 optical links. So it is useable for optical connection on the chip.

show abstract

An Energy-Efficient Programmable Manycore Accelerator for Personalized Biomedical Applications

Cited by 19 publications

References 57 publications

Embedded deep learning in ophthalmology: making ophthalmic imaging smarter

Embedded deep learning in ophthalmology: making ophthalmic imaging smarter

Implementación de Algoritmos de Procesamiento Digital de Señales en Hardware Paralelo: Artículo de revisión

Non-Blocking and Multi Wavelength Optical Router Design based on Mach-zehnder Interferometer in 3-D Optical Network on Chip

Contact Info

Product

Resources

About