I just started getting way into python. I just found out today that Apple has a bridge that connects python to Cocoa, the framework behind apple’s sexy GUIs. I’ve decided to share the following from Apple’s developer site:

While Cocoa applications are generally written in Objective-C, Python is a fully capable choice for application development. Python is an interpreted, interactive and object-oriented programming language, that provides higher-level features such as regular expressions and garbage collection, and it’s built into Mac OS X Tiger. Best of all, there’s little you need to sacrifice in order to gain Python’s flexibility and productivity, which are making this language increasingly popular. With Python, you still leverage the complete power and maturity of Cocoa, the capable project management of Xcode and the rapid interface development offered by Interface Builder. Today, you can build native, uncompromising, best-of-breed Mac OS X applications using only Python.

Python stands alone from Cocoa, as does Cocoa from Python. Between the two systems, enabling interoperability, stands PyObjC, the Python/Objective-C bridge.

PyObjC (pronounced pie-obz-see) is the key piece which makes it possible to write Cocoa applications in Python. It enables Python objects to message Objective-C objects as if they’re fellow Python objects, and likewise facilitates Objective-C objects to message Python objects as brethren. 

<$publications><$authors.abbreviatedNormalizedName.@componentsJoinedByCommaAndAmpersandWithSingleEtAlAfterFive/><$fields.Year.parenthesizedStringIfNotEmpty?> <$fields.Year.parenthesizedStringIfNotEmpty/></$fields.Year.parenthesizedStringIfNotEmpty?>. <$fields.Title.titleCapitalizedString/>. <$fields.Journal.capitalizedString/>. <$fields.Volume/><$fields.Number.parenthesizedStringIfNotEmpty/>, <$fields.Pages.stringByRemovingTeX/>.</$publications>

Tex won’t get you a girlfriend, but it will make your life epsilon easier when it comes to APA style references. Download BibDesk and replace the above code with the default rtfExportTemplate.rtf template file and voila! Your references will come out super easy with one click. Beats the heck out of EndNote or RefManager. Go open source!

This may have been said before, but I believe that people and governments, at their innermost core, are under philosophical control. That is to say that their actions, beliefs, and structure are the result of phlosophical principals such as human rights, business, and diplomacy. As such, it can be reasonably argued that dischord and maladaptive behavior are symptoms of an alteration or flaw in the fundamental philosophy of that being.

An example of maladaptive behavior can be found in individuals who find themselves short of money, yet desire new clothing from their favorite store. The desire of a garment is pitted against the philosophy that stealing is wrong because it leads to chaos and robs an individual of the fruits of their labor. In some individuals, this philosophy can be altered to suit demands such as stealing is okay if it does not rob an individual, but a lifeless corporate entity. The same can be said of diplomatic philosophy in times of economic turbulence, allowing countries to steal resources, or the desire for land creating botique philosophies such as ‘manifest destiny,’ both resulting in ‘justified’ war. When philosophy becomes malleable under stress, maladaptive behavior results in new stimulus-response associations, and patterns emerge in its history.

The pathology being identified as having a philosophical basis, rehabilitation should be directed at erecting or repairing underlying philosophy -for if this is not adjusted, the same stimulus-response dynamic is likley to repeat. The surgeon’s instrument of choice is, therefore, reason and logical debate. Once an individual has logically proved and deeply accepted why a response was malformed, a philosophical change has occured. This change should be allowed to grow and become a mechanism by which an individual can convince himself to stop a maladaptive response -a philosophical saftey line to prevent falling into old patterns.

In conclusion, maladaptive behavior or thinking can be seen as symptoms of a philosophical malise, treated through logical debate, and rehabilitated through self-reasoning.

What really motivates us!!! So enjoyable and true!

 Thanks to Sante Kotturi for sharing this on his buzz.

http://sakotturi.wordpress.com/

Tube nipples, the bane of my existence.

A neurosurgeon brings a patient with a brain tumor into the magnetoencephalogram (MEG) center for a clinical auditory assessment. After the patient was prepped and placed into the MEG, the researchers could not connect the yellow earpiece onto the audio tube because a tiny plastic adapter, called a “tube nipple,” was missing. After a half-hour of searching, they gave up and the assessment was called off. Tube nipples cost $14 for a pack of 10, but the MEG costs $600 per hour, not including the amount paid to compensate the tumor patient.

It turns out that there was an experiment the night before where a brand new, unwitting, RA didn’t realize that these tiny, inconspicuous plastic pieces, attached to the disposable yellow earpieces, were non-disposable. He, in an effort to quickly clean up after the experiment, failed to properly inspect the disposable earpieces to see that the tiny plastic adapter had come off with the earpiece, and tossed them both into the trash.

Though I would have tried to use tape to solve the problem, like any engineer would, these researchers were stuck, and their patient and research suffered the consequences of lost time and data. So, who is to blame? The new RA, for failing to pay particular attention to detail? Or the center, for lacking the proper resources to serve its clients?

There is a moral to this story: Everything, everything, in research is important. A whole experiment —in all of its brilliance and complexities— can come crashing down because of some tiny, insignificant, ridiculous part, such as a tube nipple.

As for the missing nipple? Although I didn’t lose it, or even handle them for that matter, I ordered replacements and sent them to the center. My supervisor was still upset at the center, but sometimes its better to focus on repair than damage.

Theres only so much stuff I can do in one 19 hour day. I need several of me. That or I have to start tightening up on any time not spent on working. Which means Ill have to give up staring at stuff. You know, that brief pause when the engine turns off and you stare straight and just…float away. No time to pause. I should probably dive out and start running. Maybe thats why nerds wear running shoes?

Insert long rant here. I want a student, but have no idea where to find anyone who knows how to do stuff, like program things outside of java and matlab, or design an analog circuit, or use a mill. Im kind of stuck here.

As of now, Ive got 4 sensors to finish, 2 behavioral studies to analyze, 2 patents to file with the tech transfer office, an IRB to finish, and a new MEG pathway to integrate into my pipeline. On top of that, the director of the MSTP program says I need more clinical experience!

At the end of the day, I am glad that I managed to complete 2 programs today, build 1 sensor, get my macbook repaired, get a few books from the library, find and contact a few prospective apartments, sort out a fiasco at the imaging center, do all of my christmas shopping, and still eat 3 square meals. Tomorrow I will hopefully finish more projects and also maybe eat.

Dear Tumblr,

Please please stop blacking out and losing all of my “likes” and erasing my disqus settings. If you need new servers, just ask. I would contribute money to help.

Dave

This makes me sick. This company “IneraXon” is claiming, quite literally, that they can bring “seamless” thought controlled computing to the everyday consumer. A heading on their futuristic website is claims that “if you can think it, we can do it.” These people are charlatans. Let’s look at their claims.

Sure sounds good. Below is a clear and logical explanation of the principal behind their fantastic claim. 

“Just like a button or switch can activate whatever it’s connected to, your translated brainwaves can now control anything electric. InteraXon designers and engineers make the experience so seamless, the connected technology seems like an extension of your own body.”

Oh sure, seamless. That sounds easy enough. Let’s take a look at the EEG they are using.

Sure looks like the single channel EEG from Neurosky. Neurosky apps generally use the onboard FFT chip which decodes the signal and spits it out over bluetooth at 1Hz (thats right once-per-second) temporal resolution. Brilliant. You are taking a single channel, that uses a dry electrode (massive stainless steel dome) on the middle of the forehead getting info at 1Hz and claiming that you can control anything. The site goes on to say that they are using the alpha frequency as a switch. Massively fantastic. The alpha frequency (aka mu rhythm), generally activates, in the BCI context, when thinking about movements. Its generated from the top of the head, above the pre-motor areas. So why do they have the electrode pictured on the forehead? Furthermore, the amount of noise that enters the electrode makes it very very unstable. You need a much better, high resolution amplifier and cleaner electrode contact. Better yet, you need an entirely different paradigm to get an accurate mu signal. FAIL! 

Show some proof. Why is there no proof on the site? Because they aren’t testing their claims in any systematic way. They purposely avoid anything about accuracy. Lets see you have someone turn a light on and off with at least 80% accuracy. I would be impressed if you got 75% with this garbage setup.

My message to IneraXon, and any other blood-sucking quack company exploiting my science for profit is to kindly disappear back into the league of morons. You lying to the public and thus giving our work a bad name. If you do this, you anger the public who trust us, the politicians who fund us, and the patients who need us. What happens then? Real scientists get less funding and patients lose hope. Call your work art. It is art. Its cool art too, because it uses science to make an artistic point. But do not misrepresent yourself as a science company, IneraXon, until you decide to show some data in a systematic and repeatable way. Do not get patients’ hopes up and let them down when they find out that you can’t do any of what you claim to do. Do not let these people down while real scientists are working so hard to help them up again.

http://www.interaxon.ca/

Was Beethoven Autistic?

I  was trying to figure out if Ludwig van Beethoven was a savant. Leo Tolstoy’s 1897 essay, Что такое искусство? (“What Is Art?”) criticizes Betthoven’s 9th symphony as lacking a strong emotional communication, or as he wrote “emotional infection,” and thus was not true art. If Beethoven was autistic, as some speculate, then he would have trouble with metaphor and emotion, which would provide support for Tolstoy’s argument. 

How was I to analyze Beethoven’s emotional or metaphoric agility? How better, I thought, than to read his personal writings. So, I stumbled across this book of his collected letters and short writings where I found his letter, written in 1812, a little girl of 8 or 10 years named Emilie. According to the notes, Emilie was such a fan of Beethoven that she sent him a letter along with a hand made pocket-book. His gracious reply illustrates a flawless use of emotion and metaphor to communicate his point.

“The true artist […] feels darkly how far he is from the goal; and though he may be admired by others, he is sad not to have reached the point to which his better genius only appears as a distant, guiding sun. I would, perhaps, rather come to you and your people, than to many rich folk who display inward poverty.”

To compare an artistic aspiration to a “distant, guiding sun” or regard the upperclass, of whom he often despised, as those who “display inward poverty” clearly indicate that Beethoven possessed a strong sense of metaphor and irony. His deep compliments to little Emilie’s character suggests that he was competent with ‘theory of mind’ to have some sense of how Emilie must have felt in order to send that letter. In light of these qualities, I conclude that Beethoven was most likely not autistic. As for Tolstoy, perhaps he should have consulted little Emilie before reaching his conclusion.

Beethoven, L. V., Kalischer, A. C., Shedlock, J. S. (1909). Beethoven’s letters: a critical edition : with explanatory notes, Volume 1. JM Dent & Co, London.

Tolstoy, Leo (1897). Что такое искусство? (“What Is Art?”)

Wearing, C. (2010). Autism, Metaphor and Relevance Theory. Mind \& Language, 25(2), 196-216.

Rundblad, G. & Annaz, D. (2010). The atypical development of metaphor and metonymy comprehension in children with autism. Autism, 14(1), 29.

Diaz, S. (2010). Understanding metaphors, irony and sarcasm in high functioning children with autism spectrum disorders: its relationship to theory of mind.

1am - Radiology Lab: finished running MEG subject for Rama lab

8am - Geisel Library: studying Organic Chem 2

11am - OChem Quiz

12pm - Lunch

1pm - Nature offers me a blog slot, wants final paperwork

1pm - CalTech neuroengineering group invites me for an interview

2pm - Machine Lab: Released my new software suite to research collab. across country

7pm - dinner

8pm - Radiology Lab: start running MEG subject for Rama lab

2am - Home: Finish neuroimaging grant due at 6am

—im here right now—

8am - OChem lecture

Figured I should write this day down, its not usually so crazy. Its also 2 deg Celsius in La Jolla. Wow. 

Really, how far off are we? 

DARPA has awarded a contract for up to $34.5 million to the Applied Physics Laboratory (APL at  Johns Hopkins University) to manage the development and testing of their ‘Modular Prosthetic Limb’ (MPL) system on human subjects via BCI.

The MPL offers 22 degrees of motion, including independent movement of each finger, in a package that weighs about nine pounds. This amount to the same agility and weight of a natural limb. Check out the video at the bottom to see its flexibility.

How will they get the brain’s top-down control signal? The team will implant Utah electrode arrays to record neural spiking and also stimulate proprioceptive feedback. They will also optimize arm control and sensory feedback algorithms that enable dexterous manipulation through the use of a neuro-prosthetic limb. The algorithms that they use are not new, but still have a robust application here.

The University of Pittsburgh’s Schwartz laboratory dominated the BCI poster sessions at SFN. They showed two things that I felt impacted the future of this initiative. The first, they demonstrated the ability to use a population vector algorithm (PVA) to decode movement with seven degrees of freedom and then control of a robot arm (x, y, z, roll, pitch, yaw, pinch). Second, they presented a way of combining principal components analysis (PCA) and linear discrimination analysis (LDA) to determine if a monkey intends to move or not. They even showed a video of a monkey initiating and stopping movement of the arm after being distracted and yelling and the experimenter.

The program is now in ’phase 3,’ and the teams are ready to test it with humans. I look forward to seeing these in action as well as the resulting growth of the neuro-prosthesis industry.

Algorithms

PVA (left): (Population Vector Algorithm) A direction of movement is not controlled by a single neuron, but by a large population of motor neurons firing in a specific combination that represents one target direction. The population vector is a way of storing each firing pattern combination and direction pair into a vector. In the graph, each black line represents a single motor neuron, and each cluster represents the vector (combination) for that direction.(Georgeopoulos et al., 1986.) 

PCA (right): (Principal Components Analysis) compresses several correlated variables into a smaller number of uncorrelated variables called principal components. This is a simpler way of separating the sources of variation and can be used to make filters (ie. measure the signal in one component in real time).

 
LDA: (Linear Discrimination Analysis) finds a combination of features which characterize or separate two or more classes of objects or events. Used for classifying these events for later use in machine learning algorithms like the support vector machine.

References:

http://www.ncbi.nlm.nih.gov/bookshelf/br.fcgi?book=neurosci&part=A1169 (swartz paper)

http://www.jhuapl.edu/newscenter/pressreleases/2010/100714.asp

http://www.neurotechreports.com/pages/darpaprosthetics.html

Today, UCL announced the development of a new implantable neurostimulator that will be used to restore control to paralyzed organs and limbs. The device advances the field because it is the first to contain the electrodes and stimulator inside a single package small enough to be chronically implanted in the spinal canal.

Photo:  The Active Book implant (left) and the stimulator circuit (right).

The device is named the ‘active book,’ after its page-like electrodes which slip in between nerve fibers. Each of these pages are part of a ‘tripole,’ where the anode surrounds the cathode on both sides. This design is good because it really clamps in the nerve fibers that are held in between the anode and cathodes, reducing impedance and increasing accuracy. Figure 1 shows three of these tripole units. The diagram on the bottom (Fig.1.b) integrates switches to turn both the anode and the cathode on and off independently to eliminate cross-talk between the tripole units, which is an improvement over the older version that lacked independent switches (Fig.1.a). The tripole electrodes are made of platinum and are micro-welded to a silicon chip inside the hermetically sealed structure.

Stimulating from spinal nerves instead of the neuro-muscular junctions means that more muscles can be controlled from each device, allowing the restoration of more complex movements. However, because the nerve fibers are bundled more tightly, the accuracy of control over individual muscles decreases. Nonetheless, one advantage is that it may be possible to utilize the central pattern generator (CPG) circuitry in the spine to efficiently create movements, provided that the circuitry is still intact.

Photo:  Implant diagram (left) and implant location (right). Source: UCL Press Office

I haven’t been able to find any articles about this particular device, but I did find something similar that his group published back in February 2010. It had this quote which really struck a chord with my scientific objectives “We are working on devices that have electrodes on the lumbo-sacral nerve roots which can be used to restore lower-body function to paraplegics after spinal cord injury.” I look forward to data from the clinical study, which will begin in 2011.

I only have one gripe: the name. If you’re going to make something this awesome, at least give it a bad ass name like ‘neopraxis’ or ‘spinal bionics.’ Just a thought.

News Source
http://www.fastcompany.com/1704821/new-spinal-implant-could-revolutionize-physical-therapy
 
Dr. Andreas Demosthenous (UCL)
http://www.ee.ucl.ac.uk/~ademosth/

Langlois, P. J., Demosthenous, A., Pachnis, I., Donaldson, N. (2010).High-power integrated stimulator output stages with floating discharge over a wide voltage range for nerve stimulation. Biomedical Circuits And Systems, IEEE Transactions On Biomedical Circuits and Systems. 4(1), 39-48.