传统影像组学研究对电脑硬件的要求并不高,以3D Slicer软件为例,官方推荐配置要求:
Memory: more than 4GB (8 or more is recommended). As a general rule, have 10x more memory than the amount of data that you load.
Display: a minimum resolution of 1024 by 768 (1280 by 1024 or better is recommended).
Graphics: Dedicated graphics hardware (discrete GPU) memory is recommended for fast volume rendering. GPU: Graphics must support minimum OpenGL 3.2. Integrated graphics card is sufficient for basic visualization. Discrete graphics card (such as NVidia GPU) is recommended for interactive 3D volume rendering and fast rendering of complex scenes. GPU texture memory (VRAM) should be larger than your largest dataset (e.g., working with 2GB data, get VRAM > 4GB) and check that your images fit in maximum texture dimensions of your GPU hardware. Except rendering, most calculations are performed on CPU, therefore having a faster GPU will generally not impact the overall speed of the application.
Some computations in 3D Slicer are multi-threaded and will benefit from multi core, multi CPU configurations.
Interface device: a three button mouse with scroll wheel is recommended. Pen, multi-touchscreen, touchpad, and graphic tablet are supported. All OpenVR-compatible virtual reality headsets are supported for virtual reality display.
Internet connection to access extensions, Python packages, online documentation, sample data sets, and tutorials.
主要包括:
配置要求:
计算机需要足够的内存和图形功能将原始数据和处理结果保存在其内存中。
内存:超过4GB(推荐8G以上)。
显示:最小分辨率为1024×768(推荐为1280×1024或更好)。
专用图形硬件和内存(建议使用1GB)。
Slicer中的许多计算都是多线程的,并且将受益于多核,多CPU配置。
接口设备:推荐使用带滚轮的三键鼠标。
根据笔者实际经验,如果内存不足可能导致软件罢工,其余显示、GPU等均非必要。
因此笔者推荐16G内存电脑,且越大越好。
如涉及深度学习等方法的研究需根据实际需求配置电脑,无法一概而论。