BRAIN RESEARCH WITH THE ATARI
Mapping brain receptors at medical schoolby TOM RAINBOW, PH.D.
You bought an Atari ...with the taxpayer's
money?! exclaimed the electronics technician at my university. You
should be arrested!"
"Er, I gather then you won't assist me in interfacing the Atari to my laboratory equipment."
He started to laugh. "Here at the university electronics shop, we don't work on toys."
A toy! How could anyone call an Atari 800 a toy? Granted that it played great games. But a common lab micro, such as the Apple II, had only one microprocessor, the 6502. An Atari had four microprocessors, the 6502, ANTIC, GTIA and POKEY. The ANTIC co-processor gave the Atari incredible graphics abilities-display lists, color palettes, single-bit scrolling-things the Apple simply couldn't do. Plus the Atari was cheaper than an Apple. An Atari with a disk-drive and printer would cost me roughly $1000, while an Apple with the same peripherals would easily cost twice as much. In an era of declining Federal support, you had to be a penny-pincher.
I wanted to use the Atari as a computerized light meter. My research involves making chemical pictures of the brain by a technique called autoradiography. Thin slices of a rat brain or human brain are exposed to a radioactive drug or hormone. The drug or hormone then attaches itself to protein molecules on the outside of brain cells called receptors. By placing the brain-slice against a special photographic film that responds to radioactivity, it's possible to take a picture of where a receptor is in the brain.
The Atari would tell me how dark or light portions of the autoradiogram were. The higher the optical density of a brain region, the more receptor there was, so I would essentially learn from the Atari the biochemical concentration of receptor within that brain area. The simplest way to do this was to interface a photocell with the Atari. I would project a light through the autoradiogram and move its image over a stationary photocell. The less light transmitted to the photocell, the more receptor was in that brain region.
Originally, I had in mind connecting a photocell through the PIA joystick ports. This got nixed by the supportive, enthusiastic reaction of my campus electronics shop. So I opted for a commercial analog-to-digital converter that could connect with the Atari through the RS-232 serial port on the 850 interface. The one I bought was the EI-100 unit from Cambridge Development Laboratories, Watertown, Mass. This unit was nice in that you could purchase a plug-in photocell for it, and most significantly, Cambridge Development Labs had actually prepared a separate manual for the Atari 800, complete with sample programs written in Atari BASIC.
With the sample listings, it was a cinch to write a program that would open the RS-232 port and take light-readings from the photocell. There are several other sensors available from Cambridge Development Laboratories, so you could also use the Atari to take pressure or temperature readings, for instance.
We got cute with our program, choosing a light pen for user input, and trying to use color and mixed display-list modes as much as possible. Essentially, the program is designed to let the user set up "laundry lists" of brain structures. The Latinized medical name of the brain structure is displayed on the top line of the TV screen in Graphics 0. Below, in Graphics 1, is the current photocell reading, represented as a number from 0-255. When the user wants to take a reading, he touches the "Keep Value" spot with the light pen. The program will then average subsequent readings until the user touches the "Exit" spot.
The name of the next brain structure in the laundry list is then displayed for analysis, and so on, until the optical densities of all the structures on the original list have been measured. The program then does some algebra to convert the density values into the actual concentration of receptor in a brain region. The laundry list of structures with the associated receptor concentrations is then displayed in Graphics 0 and dumped to an Atari 825 printer.
The program is long, occupying essentially all the available RAM on a 48K Atari 800. It is messy to write such long programs in Atari BASIC, with its lack of Trace features and whatnot. The available enhancement software that improves the editing features of Atari BASIC takes up too much memory to use with our program. If we had to do it all over again, we would probably write the program in some version of C Language that supports floating-point on the Atari. We could have then used a real text-editor and compiled the program. However, Atari BASIC was really the best choice when we wrote the thing, and it wouldn't be such a bad choice even now.
I would like to think that I may make some significant scientific discoveries with my Atari. We've published about 20 research papers so far where we've used the Atari to analyze brain autoradiograms. None of these studies has yet won me a Nobel Prize or cured a Dread Disease, but maybe they are making some incremental contribution to our knowledge about how the brain works, and certainly, they would have been done much less well without the Atari.
The other important use for our Atari is as a low-cost word processor. All of those 20 research articles plus about $200,000 worth of research grant proposals were written on an Atari. We are big fans of the Letter Perfect word-processor by LJK, which works nicely with our 18 characters-per-second el-cheapo Comrex CR-1 daisy-wheel printer. With Letter Perfect, we can make the Comrex underline, superscript, subscript, and do a one-word boldface-all the printer functions you need to publish a scientific paper-and done with a $70 word-processor and an under $500 daisy-wheel. We currently have two Atari 800's in the lab, one is our "home-made" densitometer which doubles as a word-processor and the other basically just functions as a word-processor.
I suppose that like many Antic readers, I get frustrated when the rest of the world doesn't recognize the superiority of my computer. However, I use my Atari for scientific research, a very serious purpose, and I like it. It's truly a very serious computer. And you know what else? it plays great games!
Tom Rainbow, Ph.D. is an Associate Professor of Pharmacology at the
University of Pennsylvania School of Medicine.