There are many open-source solutions for BCI projects like BCI2000 (http://www.schalklab.org/research/bci2000), OpenVibe (http://openvibe.inria.fr/) and EEGLab (https://sccn.ucsd.edu/eeglab/index.php). That's based on the kinds of tools that our customers use. Most of these, aren't as pretty as tool like Neurovis, but in reality most of the information that we make use of to control prosthetics or signal intention involve looking at the temporal-frequency relationships between and within broad regions of the brain. There isn't a lot that you gain from just looking at the brain light up like this for BCI — the main use for a visualisation like this, is as the docs say, for diagnosis and determination of epilepsy since in epilepsy the activity you'd see is much higher than usual.
EEG has better temporal resolution than FMRI; you are measuring the electrical activity rather than vascular changes, the former changes more rapidly than the other. EEG, however, is just the surface activity of the brain, so you don't get information about 'deeper' (physically) brain processes; this is where FMRI is invaluable. EEG is also limited to the size of the electrodes and how many electrodes you can physically place in one location. 256 electrodes on an EEG cap is about the limit you can get to.
Electrocorticography (ECog) involves implanting electrodes on the dura, this can substantially improve the density of electrodes in a given area, however this doesn't measure deep brain activity and we have no way of leaving the electrode grid in place for long periods of time without risking infection. For BCI, we've been able to classify more classes of data using ECog than EEG — research by Kyousuke Kamada and Gerwin Schalk are informative. It's a very promising area if we can work out how to implant the electrodes, seal the skull and telemeter data out.
Magnetoencephalography (MEG) can help with measuring deep brain activity, but there are other tradeoffs to consider. Essentially, the point where we are now is combining multiple techniques to get the best temporal, spatial and frequential compromises.
Thus in answer to your question; no FMRI is not great for realtime responses, measuring the electrical activity has better temporal properties so EEG and ECog work better here. Sensor density is part of the problem, but once you solve density you then need to consider how you'll deal with deep brain measurements.
Background; I own a company that distributes BCI equipment for g.tec medical engineering in the UK. We've been operating in this space for 8 years. I have a PhD in Pharmacology and a speciality in electrophysiology.
True, plus, I forget the legislation but you are effectively breaking into the computer first which is a crime. Committing a crime for a noble outcome is still a crime.
Incentives is a real issue here and those that provide the patch would, reasonably, expect a reward i.e. MS for updates, AV provider for testing, finding and securing the vulnerability and a whitehat for disclosure.
However, there is no reason why a "charitable" hacking group wouldn't do this as part of some sort of digital vigilantism. Sometimes people do things without extrinsic reward and the thrill here is that it is as hard as cracking, but you get to know that your efforts could be immediately applied.
What gets me is why we don't see more viruses that _deliver_ the patch to fix the vulnerability.
It's perhaps a little more difficult as you'd need a vulnerability to keep spreading the innoculation. Arguably, though you release the virus, let it spread and then trigger the innoculation using a mechanism like calling out to a webserver, just as the kill switch worked here.
While it doesn't have an inherent advantage, it has the mindshare and momentum of a community that has these tools now.
R could be just as capable as Python, but I think Python has largely won the race to be the most popular language for data analysis which in turn encourage more developers to commit to it, cementing Python's advantage.
R still has solid lead in statistics and a good mindshare amongst academics.
I'm a Pharmacologist and I've worked in a number of Biotechs. I have been part of pricing discussions to value our drugs and your assessment is correct — price discrimination, as in most industries, is based on what a particular market can bear. Factors that influence this are the way that healthcare systems are run (i.e. public healthcare, insurance based), the cost of compliance and competition. I wouldn't say that the industry subsidises poor countries via sales in the developed world, it's more a case of it's better to be paid something rather than be paid nothing. I personally, however, do like the happy side-effect this has for patients in poor countries.
Increased regulation costs more to comply with and that cost of course is born by the consumer. The cost of regulatory compliance also varies on the market you sell to, the US being one of the most expensive. Just as with software or hardware, the price reflects the cost of production and maintaining the product.
Notably with biologics (Insulin, any hormone ...) as opposed to small chemical entity (paracetamol, aspirin etc) is that broadly they are harder to keep in their stable active form. The are heat sensitive, chemically sensitive and have a tendency to stick to themselves. This tends to increase the cost of storage, logistics and compliance. It also means that if you can make a worthwhile biologic it will generally experience less competition.
I'm a Pharmacologist and I've worked in a number of Biotechs. I have been part of pricing discussions to value our drugs and your assessment is correct — price discrimination, as in most industries, is based on what a particular market can bear. Factors that influence this are the way that healthcare systems are run (i.e. public healthcare, insurance based), the cost of compliance and competition. I wouldn't say that the industry subsidises poor countries via sales in the developed world, it's more a case of it's better to be paid something rather than be paid nothing. I personally, however, do like the happy side-effect this has for patients in poor countries.
Increased regulation costs more to comply with and that cost of course is born by the consumer. The cost of regulatory compliance also varies on the market you sell to, the US being one of the most expensive. Just as with software or hardware, the price reflects the cost of production and maintaining the product.
Notably with biologics (Insulin, any hormone ...) as opposed to small chemical entity (paracetamol, aspirin etc) is that broadly they are harder to keep in their stable active form. The are heat sensitive, chemically sensitive and have a tendency to stick to themselves. This tends to increase the cost of storage, logistics and compliance. It also means that if you can make a worthwhile biologic it will generally experience less competition.
Ian is linking from twitter for the images so your company is probably blocking twitter. Evidently, however, not the equally large time-sink that is Hacker News!
I thought the vibe of the conference was the best we could have hoped for.
Python is crazy popular now — the number of people we had from big blue chips and the fact that Sainsburys (a nationwide grocer for you US chaps) sponsored us was telling of just how deep Python has penetrated the enterprise market.
Animal experimentation is repellant to many people, I can see why. The point is that society has decided that they'd prefer well-researched, empirically-observed working medicines.
Regulation must strike a compromise between allowing research to continue with enough oversight to limit abuse. It is the sort of compromise developers use with security; enough to limit abuse, but not too much that no-one can use the system!
Your point is quite right, I don't think it's self serving at all. Complex physiological diseases like cancer can be observed as tumours grafted to a scaffold in vitro, but you cannot observe the effect of drugs, it's relationship to the whole organism, without using a whole organism.
Monkeys have immune responses that are as similar to humans as we can hope for. Further, Mice while they have some conservation of immune related genes there are many divergent expression of immune related genes including alterations in master-gene (cis) and innate immune gene (CD4) expression [1] which may explain why they don't translate very well into humans. We have an awful time trying to get decent responses to allergic stimuli with mice and guinea pigs! Rodents have very robust immune systems in my experience.
Unfortunately, science isn't the sort of endeavour where anything can resolutely be considered necessary to start with. Arguably there should be some doubt about whether your research will lead to something important else you can't be doing anything novel. There is nonsense in any endeavour and that is why there are layers of ethics committees, reviews and audits within animal research that limits wanton nonsense. The price of any endeavour is that there will be waste — humans cannot help it — it is unfortunate that the waste is at the expense of animal life.
The first article you link to is about the abuse of frequentist statistics, it is not a piece of nonsense primary research (1182327). The article is correct and an interesting read, but not in support of poor use of animal experiments. The second article (8124111) is about clinical research, not pre-clinical animal research although the arguments raised in the article are applicable to pre-clinical research. It is very difficult to find and expose nonsense research!
I can confirm that this largely rings true for the UK. The Home Office is the body that deals with the certification of a specified procedure and provides the oversight that animal experimentation is done in the most humane manner that is practical for a given experiment.
I recently completed my doctorate in Pharmacology and I have a Home Office license to perform specific surgical procedures on animals. I have training in surgery, anaesthesia and euthanasia which is routinely assessed to maintain my Home Office license.
The details of the experimental protocol that the poster lists - deprive resource, validate behaviour and reward - are probably correct, sensationalised, but correct. I say that, because this reads like the sort of pain protocols I have been involved with in the past. The simple fact is that to explore whether a pain medication works, you must inflict pain, see whether your drug modifies pain behaviour and repeat to sufficient statistical power.
The details of the experimenter dispatching (killing, euthanising ...) the animals with chloroform, en masse, in a plastic bag is a violation of his license and his colleagues license, it is not a listed method on Schedule 1 Method of Euthanasia. It is one of the facets of animal experimentation regulation that I am proud of, the humane killing of animals ensures minimal distress. The euthanasia regulation for animal experimentation are stricter than those in the food industry and far more stricter than the exemptions that religious faiths have when preparing meat.
Non-vertebrates animals are covered by the ASPA 1986 guidelines in the UK for euthanasia and commonly you must present an anaesthetic regime for all uses of laboratory animals. Octopuses are explicitly named as they are one of the oldest laboratory animals in use — they were pivotal in understanding the role of ions and ion channel pumps in nervous transmission.
Correct me if I'm wrong but perhaps you had the same issue as me. The documentation is plentiful, lots of good examples and the book, similarly, increases with a nice linear complexity from basic "how do I select a (cell | row | column) ..." to full blown how do I do timeseries analysis on a dataseries pulled in from a remote source.
The issue I had was not the documentation but the language of pandas mirrors the language used in R (I think this is something Wes McKinney intentional did) and it's the burden of all that new verbage that makes the documentation harder to sift through. Some choice exampels; "melt", "stack/unstack" and "reindex" — necessary, I grant you, so that functions can be aptly named and in turn encapsulate vectorised procedures that are composable.
I found that the documentation was harder to search because I lacked the domain language and the documentation, for better for worse, doesn't dawdle with educating the reader about the verbage — worked examples often provide a easier route. It reads like a mathematical proof rather than prose and I used to think that the documentation was too terse but now I appreciate that probably just succinct.
Hey Laurie. To be honest, if the undergrads can knock out presentations as good as your presentation on DOTA2 analysis then, we, at PyData London should be actively trying to get more undergrads to come to our meetup.
An unbalanced group is, as you say, a problem. The key is to get a wide gamut; people with ideas and some idea of implementation, flat-out implementation people (devs and designers), business people and investors. A lot of the undergrads I know are green but they often have a lot of energy, ideas and can be fine coders.
I am committee member of the PyData London Meetup and I'd argue that the tech scene in London is simply more vibrant outside of the universities at the various meetups, hacker spaces and conferences so often you are better served going to those than a University society.
There is a certain amount of red tape organising events at Universities (my basis for this is KCL and Imperial) unless they are either explicitly educational and you often have to liase with security so we've had a lot less friction by engaging tech companies to host meetings. You get the industry links by making use of their facilities, you get students, devs and business types and in all honesty companies in Silicon Roundabout are gagging to look cool and host a meetup.
Saying that, I'm about to start post-doc work at KCL (been here since Undergrad) so I'll be kicking around for the next 3 yrs so I'll be signing up.
Seconded - that series illustrates a number of dystopian fantasies based on rapid technological change.
Forgive the tangent but I felt the "The Entire History of You" episode is particularly poignant at illustrating one of the most damage side-effects of always being connected and that is the losing the ability to forget. The ability to recall (via images, videos) at a whim can lead to being obsessed with matters that occurred in the past while living in the present. It's a problem unique to our generation since previous generations weren't able to capture, as ably as we can capture now, all the unnecessary specifics of daily life.
I upgraded from 12.04 LTS and I had a few niggles - the configs for Compiz were messed up and the AMD proprietary driver wasn't working, oh and grub got nuked.
Ultimately, blowing away the compiz configs (rm .compiz*), a dpkg --configure -a fixed the AMD driver and boot-repair from a Ubuntu LiveCD was a wonderfully simple way to redo grub.
Not painless but not particularly difficult to fix.
EEG has better temporal resolution than FMRI; you are measuring the electrical activity rather than vascular changes, the former changes more rapidly than the other. EEG, however, is just the surface activity of the brain, so you don't get information about 'deeper' (physically) brain processes; this is where FMRI is invaluable. EEG is also limited to the size of the electrodes and how many electrodes you can physically place in one location. 256 electrodes on an EEG cap is about the limit you can get to.
Electrocorticography (ECog) involves implanting electrodes on the dura, this can substantially improve the density of electrodes in a given area, however this doesn't measure deep brain activity and we have no way of leaving the electrode grid in place for long periods of time without risking infection. For BCI, we've been able to classify more classes of data using ECog than EEG — research by Kyousuke Kamada and Gerwin Schalk are informative. It's a very promising area if we can work out how to implant the electrodes, seal the skull and telemeter data out.
Magnetoencephalography (MEG) can help with measuring deep brain activity, but there are other tradeoffs to consider. Essentially, the point where we are now is combining multiple techniques to get the best temporal, spatial and frequential compromises.
Thus in answer to your question; no FMRI is not great for realtime responses, measuring the electrical activity has better temporal properties so EEG and ECog work better here. Sensor density is part of the problem, but once you solve density you then need to consider how you'll deal with deep brain measurements.
Background; I own a company that distributes BCI equipment for g.tec medical engineering in the UK. We've been operating in this space for 8 years. I have a PhD in Pharmacology and a speciality in electrophysiology.