NeuroImaging and statistical analysis workstation

We were consulted by a client who was considering introducing a machine for medical image analysis, particularly NeuroImaging.

The software we plan to use is as follows:

Image analysis	FreeSurfer, FSL, ANTs, Matlab, SPM, CONN (SPM/ Matlab environment)
Numerical analysis statistical processing	R, Matlab, Python, Julia

A GPU with 16GB or more of VRAM is required to support CUDA-compatible analysis tools, inference using deep learning, and experimental learning processing.

As for the chassis, the Define 7 (240mm wide x 475mm high x 547mm deep) is too large, so an ATX-compatible chassis, specifically the Define 7 Compact (210mm wide x 474mm high x 427mm deep), is recommended.

In addition, the motherboard must have a power phase of 9000+14+2 or more and four or more SATA ports, while taking into consideration compatibility with the Ryzen 1 series.

In addition, we would like to use the service that does not require the HDD to be returned.
*For details on HDD detailed service,'Kensetsu Fair Hokuriku,'

We proposed a versatile configuration that is centered around NeuroImaging tools and can also be used for the experimental introduction of statistical analysis, signal processing, and deep learning.

CPU	AMD Ryzen 9 9950X 4.30GHz (up to 5.70GHz at Boost) 16C/32T
memory	Total 128GB DDR5 5600 32GB x 4
Storage 1	4TB SSD M.2 NVMe Gen4
Storage 2	20TB HDD S-ATA
Video	NVIDIA GeForce RTX5080 16GB
network	On board (2.5G x 1) Wi-Fi x 1
Housing + power supply	Mid-tower chassis + 1000W 80PLUS PLATINUM
OS	OSなし
security	No HDD return required service for 1 year

GPU configuration and support for analysis applications

The NVIDIA GeForce RTX 16 GPU with 5080GB of VRAM was selected for use with CUDA-compatible image analysis tools and lightweight learning processing.
It meets the requirements of major research frameworks and is flexibly configured to accommodate future expansion of applications.

Selection of enclosure with safety in mind

Although a small enclosure was requested, we decided against adopting it due to concerns about interference and heat generation.

In recent years, GPUs have generally been designed with auxiliary power terminals on the side, and in compact cases, the gap between the terminal and the panel is narrow, making it difficult to route cables.
As a result, excessive tension is applied to the connector, leading to serious problems such as heat generation and burning, as have been reported both in Japan and overseas.

To avoid these risks and ensure an environment that can withstand safe and long-term continuous operation, we have designed a roomy interior.Fractal Design Define 7 Mid-Tower CaseWe proposed the following.

Regarding motherboard selection

Based on customer requests, we have developed a motherboard.ASUS TUF GAMING X870E-PLUS WIFI7was selected.

This model is16+2+1 power phases (16 phases for CPU) and four S-ATA 6G portsThis design minimizes voltage fluctuations even during continuous operation, making it easier to avoid problems during processing.

Furthermore, we have not received any reports of malfunctions with the ASRock X870 Steel Legend WiFi to date, and it has a proven track record of stable operation.
We provide the following for all PCs:High-load testing before shipping　

Even if you use ASRock products,Update the BIOSWe provide the necessary settings after making the necessary changes.
We can also adjust the specifications to suit your needs, so you can make your selection with confidence.

For those who are active in these fields Medical Image Analysis Neuroscience Clinical research Medical Engineering Computational Neuroscience

We also accept customizations based on the tools and environment you plan to use. Please feel free to contact us first.

Keyword ・What is NeuroImaging analysis? Neuroimaging analysis is an analytical method used by researchers in fields such as medicine, psychology, and information science to diagnose brain diseases and visualize mental activity. In particular, it is used to detect abnormalities and pre-symptomatic changes in brain function, and to extract disease characteristics using machine learning. ・What is FSL (FMRIB Software Library)? FSL is a toolset used primarily by researchers in the fields of chemistry and life science to elucidate the ultrafast reaction mechanisms of molecules and proteins. It is used to observe reaction transition states and structural dynamics in real time and to investigate the details of biological phenomena and material changes. reference:F.S.L. *Jumps to an external site ・What is R? It is an open-source statistical analysis software used by researchers in a wide range of fields, including medicine, bioinformatics, social sciences, and engineering. It is widely used for a variety of purposes, including data analysis, statistical model building, and visual display of results. reference:The R Project for Statistical Computing *Jumps to an external site

■ Click here for details and inquiries about this PC case NeuroImaging and statistical analysis workstation * Please enter the name of the case or your desired conditions.