I have written about automated decision making or machine learning for Computer Weekly, in particularly the numerous problems with using it. The biggest set of issues is summed up nicely by Joanna Bryson of Bath and Princeton universities: “The reason machine learning is working so well is it is leveraging human culture. It’s getting the bad with the good.”
Volunteers in Oxford agreed to infect themselves with typhoid, as part of a recently-completed trial of a vaccination that could save the lives of hundreds of thousands of people.
The Oxford Vaccine Group, part of University of Oxford, gave 99 volunteers a drink laced with live Salmonella Typhi bacteria a month after vaccinating them. A third had been vaccinated with Typbar-TCV, a new conjugate typhoid vaccine, a third with established vaccine Typhim Vi and the rest with meningitis vaccine Menveo, which does not protect against typhoid. Neither the volunteers nor the doctors carrying out the injections knew who was getting which vaccine. Continue reading “Oxford vaccine test volunteers infected with typhoid”
Central Manchester University Hospitals NHS Foundation Trust has diverted 40% of its clinical waste into a cheaper-to-treat category, by introducing special ‘tiger’ bags.
The distinctive yellow and black striped bags are for clinical waste such as dressings and incontinence pads used by non-infectious patients. Previously all clinical waste had been heat-treated in an autoclave, but this process – which adds £100 per tonne to the cost – is now used only for waste from infectious patients. Amounts going through the more expensive process have fallen from around 130 tonnes in April 2014 to around 60 tonnes in December 2016. Continue reading “Central Manchester tames clinical waste with tiger bags”
Britain is a world-leader in genomics. When I heard George Freeman MP, the chair of the prime minister’s policy board, tell a conference that despite having voted Remain he thought Brexit could lead to a better regulatory framework for genomics, it sounded like a story. Computer Weekly published the resulting article last week.
UK organisations concerned with genomics didn’t seem to want to discuss it, however. Apart from a short statement from the Department of Health and Mr Freeman’s comments, I interviewed a specialist lawyer and Kari Stefansson, the founder of Icelandic genomics firm deCode Genetics, who suggested that public healthcare should only be available if patients participate in genomic research.
Organisations should be considering and talking about the opportunities Brexit brings, as well as working to mitigate its dangers. In public sector IT, some suppliers already serve the UK and other Anglophone countries such as Canada, Australia and New Zealand, given similar legal and political structures; new trade deals could help expand this. And if Brexit allows regulatory changes that boost genomic research, they are surely worth considering. Continue reading “For organisations, Brexit means deal with it”
Earlier this month I took part in the Milton Keynes self-driving car trial, for a forthcoming article for Computer Weekly. Sitting in a car that steers its own wheel was enormous fun. However, it was striking that those I spoke to for the article thought that fully driverless cars on public roads are a long way off. Such technology is generally better-suited to making operators more productive rather than replacing them.
There is something of an obsession within tech for eliminating people completely, whether drivers or customer service staff. But with the latter, where the process has gone a lot further, the result is often frustrated rather than delighted customers, and the same is likely to be true of any process that involves human users – which is most of them. As Izabella Kaminska argues in the FT’s Alphaville blog, among many other drawbacks, properly driverless taxis would quickly end up twice as filthy as night buses at dawn with no-one to supervise them. Continue reading “IT should focus on productivity not eliminating people”