If the hype is anything to go by, we’re entering a bright future where “Artificial Intelligence” will revolutionize medicine. This post provides some counterpoints to that hype, focusing on three issues:
patient privacy data sovereignty professional knowledge and its exploitation - image (1024x1024) generated using Stable Diffusion v2.0 (Linux, NVIDIA RTX 3090, 24GB) > prompt: creepy humanoid AI doctor looking at medical files, Steps: 20, Sampler: Euler a, CFG scale: 7, Seed: 403319539, Size: 512x512, Model hash: de2f2560, Batch size: 4, Batch pos: 0
Document editing and report creation are key parts of any researcher’s workflow. Thankfully, the tools to edit research documents have improved greatly over the last few years. While I used to write most documents with $ \LaTeX $, I’d now recommend that researchers use R Markdown.
R Markdown allows the use of R code in a Markdown format and is also easily extended with features. For example, there is the excellent bookdown package which allows easy use of book formatting, bibliographies and cross-references with Markdown.
Introduction Setting up wireless networking can be a frustrating experience. Available commercial routers vary greatly in their implementation and standards compliance, meaning that compatability between brands of router can be problematic. There are also significant security issues in implementing WiFi netowrks, compounded by the fact that manufacturers rarely update their firmwares and dump support for routers after only a few updates.
Thankfully, there are now some great options for installing open-source software on routers so that it’s easier to have cross-compatibility and maintain up-to-date security patches.
this post was previously published on my old website, there’ll be a few of those older but useful posts that I’ll be migrating over in the next little while…
Molecules are beautiful things, intricate and infinitely variable. As part of research publications it can be useful to catch them from their best angles. This short post gives some tips on how to present molecules in publications.
Our model for today is Prostaglandin-F2α There are a number of chemical databases and ways of expressing molecular identities.
Scientific computing increasingly involves handling large amounts of raw data. This is particularly the case for neuroimaging. With increases in processing speed outpacing storage speed, disk i/o has become a limiting factor in many computing applications, especially in multi-user systems.
For my consulting work with GreenAnt Networks, I was asked to build a high-speed disk array for scientific computing and virtual machine storage. This is a brief build log and performance test of the outcome.
Introduction Scientific publishing is undergoing a transition from corporate-controlled, for-profit publishing to more open models. While “Open-Access”, is part of this, there are a number of considerations important in true “Open Publishing”:
Free access to the public (who pays for most of the research via their taxes!) Free submission of articles by authors Open Data - where original data and analysis workflows are made public Creative Commons - otherwise known as Copyleft, where rights are retained by the author but the content is usable with attribution Transparency of the review process (with or without anonymity) Open software, standards and tools used in the publication process This post will focus on item (6) from above.
Sharing MRI data with colleagues can be a time-consuming exercise. The files are often large and data-sets need to be viewable in a flexible fashion. Rich features assist the communication of results, including:
3D viewing flexible slicing overlay of specific volumes While there are many software packages for running on the desktop, there’s nothing better than being able to point to a simple HTML link and open the scan data in a prepared format.