NeuroTech...

This New Scientist Tech News Item piqued my interest today...

Man and machine vision in perfect harmony
08 July 2006
Lakshmi Sandhana
Magazine issue 2559
SCANNING through large amounts of images or video footage can be time-consuming and laborious. Now a machine that taps into the processing power of your brain can help you do it much faster.

The device, called the cortically coupled computer vision system (C3Vision), was developed by Paul Sajda and colleagues at Columbia University in New York. It uses an electroencephalogram (EEG) to measure electrical activity in the brain via skull-cap electrodes, while the user scans through images. If the user sees an image that grabs their attention, such as a picture of a crime suspect, the EEG will pick up a spike of electrical activity in their brain 300 milliseconds later, something known as a P300 response. This occurs well before the user can click a button to say they saw the picture. A connected computer can then collect all the images associated with a P300 response.

I was interested in the subject matter, but didn't really want a subscription to their magazine, so I thought I would see what I could see about this little piece as it's another step towards human and machine integration.

1st Link
This was a completely random find, but it at least verifies the veracity of the New Scientist piece.

Associate Professor Paul Sajda of Biomedical Engineering has been awarded a one-year, $750,000 Phase 1 grant from DARPA/ NGA entitled “Cortically-coupled computer vision.” The goal of the project is to interface single-trial real-time neuroimaging with computer vision algorithms to improve object recognition and image search by humans. Columbia co-Investigators include Professor Truman Brown (BME and Radiology) and Postdoctoral Research Scientist Robin Goldman of Radiology. Outside collaborators include City College of New York, Oregon Health Sciences University and Siemens Corporate Research.

Oh look it's our friends at DARPA (wonder why NS left them out of their blurb?) funding more research that will benefit the private sector - after the Military gets their version of the interface linking man with drone capable of making instant decisions and capable of analyzing all visible terrain data at the speed of thought...

Second Link
I found the mini-article below and was pretty impressed with the entire Seeing With Sound website. There's a lot of in-depth information and access to even more information on the site.

Cortically coupled computer vision for rapid image search.

IEEE Trans Neural Syst Rehabil Eng. 2006 Jun; 14(2): 174-9
Gerson AD, Parra LC, Sajda P

We describe a real-time electroencephalography (EEG)-based brain-computer interface system for triaging imagery presented using rapid serial visual presentation. A target image in a sequence of nontarget distractor images elicits in the EEG a stereotypical spatiotemporal response, which can be detected. A pattern classifier uses this response to reprioritize the image sequence, placing detected targets in the front of an image stack. We use single-trial analysis based on linear discrimination to recover spatial components that reflect differences in EEG activity evoked by target versus nontarget images. We find an optimal set of spatial weights for 59 EEG sensors within a sliding 50-ms time window. Using this simple classifier allows us to process EEG in real time. The detection accuracy across five subjects is on average 92%, i.e., in a sequence of 2500 images, resorting images based on detector output results in 92% of target images being moved from a random position in the sequence to one of the first 250 images (first 10% of the sequence). The approach leverages the highly robust and invariant object recognition capabilities of the human visual system, using single-trial EEG analysis to efficiently detect neural signatures correlated with the recognition event.

That's basically saying the same thing the NS article said, only using the technical jargon of the original publishing. Sometimes, I find reading these journal publishings really cause the experiment to sink in - this is a computer monitoring a human's thought process and recognizing a spike in his or her EEG as recognition in the human brain. The computer then grabs the image that caused the peak for later application...

also from the site, but unrelated to this article, was this little tid-bit:

IEEE Trans Neural Syst Rehabil Eng. 2006 Jun; 14(2): 246-50
Wilson JA, Felton EA, Garell PC, Schalk G, Williams JC

Most current brain-computer interface (BCI) systems for humans use electroencephalographic activity recorded from the scalp, and may be limited in many ways. Electrocorticography (ECoG) is believed to be a minimally-invasive alternative to electroencephalogram (EEG) for BCI systems, yielding superior signal characteristics that could allow rapid user training and faster communication rates. In addition, our preliminary results suggest that brain regions other than the sensorimotor cortex, such as auditory cortex, may be trained to control a BCI system using similar methods as those used to train motor regions of the brain. This could prove to be vital for users who have neurological disease, head trauma, or other conditions precluding the use of sensorimotor cortex for BCI control.

To give you a better idea what Electrocorticography is and what they mean by "minimally-invasive", here's the wikipedia definition of Electrocorticography:

Electrocorticography
From Wikipedia, the free encyclopedia

Electrocorticography (ECoG) is the practice of using an electrode placed directly on the brain to record electrical activity directly from the cerebral cortex. By placing the electrode directly onto the cortical grey matter one can record signals from neurons much more effectively than through electroencephalography (EEG). One of the main limitations of EEG is the poor spatial resolution, because the skull acts as an attenuator of neural signal, thus filtering out high frequency signals and lowering the signal-to-noise ratio. The biggest drawback to ECoG is the requirement of surgery in order to place the electrodes under the dura mater directly onto the brain's surface.

YIKES!!!! Place an electrode directly ON the brain...That's minimally invasive?!? So what do you suppose invasive would be? Taking a silicone chip and planting it directly in the cerebral cortex? Nevermind...that's probably exactly what invasive would be. And probably the next step in this developing tech...