How Dogs Love Us: A Neuroscientist and His Adopted Dog Decode the Canine Brain (18 page)

While Melissa worked with McKenzie to get comfortable around the magnet, I settled in with Robert and Sinyeob to plan the afternoon’s scans. Based on what we had learned from Callie’s dress rehearsal, we would implement several changes in the scan protocol.

From the control room I could see Melissa coaxing McKenzie into the scanner. Tentative at first, McKenzie carefully walked up the steps and plopped down on the patient table. She seemed unsure of what to do. Melissa went around to the other end of the magnet and, using some hot dogs, lured McKenzie in. As soon as Callie caught sight of that, she ran up the steps and climbed over McKenzie to stake out her position inside the magnet.

Sinyeob started laughing. “Look, two for one!”

McKenzie was not amused.

Turning to Helen, I said, “Could you get Callie and keep her out of the magnet until we’re ready?” Helen beamed, happy to have a job to do.

As had been done at the dress rehearsal, Sinyeob programmed three types of scans. The first would be the localizer image to see where the dog’s brain was. This would allow us to set the field of view for the subsequent scans. Next would come the functional scans.

“How long do you want to scan?” asked Robert.

For human studies, we normally scan in blocks of about ten minutes. That is about as long as a person can stay engaged in a task while in the scanner. We usually give the person a break of a few minutes and then repeat. I didn’t expect the dogs to stay in the scanner for that long, though.

“Let’s go with five minutes,” I said. With a complete scan through the brain taking two seconds, that would give us 150 images. Our goal was to get a consecutive sequence of ten images without movement somewhere in that block.

If we had time, and the dogs were up for it, the final scan would be a structural. It would be quick, but they would need to hold still for thirty seconds.

I turned to Andrew and said, “Are you ready?”

“Let’s do this!”

I looked at Callie and said, “Hey, Callie! Wanna do some training?” She jumped up to lick my face.

Melissa put McKenzie in a pup tent that she had brought. This let her relax while we worked with Callie.

Rebeccah followed Callie and me into the magnet room and took up a position at the end of the patient table where she could closely monitor Callie for any signs of distress. I slipped the earmuffs over Callie’s head and motioned for her to go into the magnet. As she shimmied in, I ran around to the back end so that I was facing her. Callie scooted into the head coil and placed her head on the chin rest.

“Good girl!” I said, and gave her a hot dog. She wagged her tail and looked at me expectantly. Meanwhile, Andrew had taken up a position just to my left. I glanced at him.

“Do it.”

Andrew raised his hand and pointed to Robert, who was observing through a glass window in the control room. The scanner came to life.
Click. Click. Click.
Then the buzzing of a million bees. The software, not recognizing a canine occupant, was beginning the
shimming procedure to compensate for the distortion of the magnetic field that Callie’s canine form was causing.

Callie’s eyes narrowed to slits. I held up a piece of hot dog but it was already too late. She had started to back out. Further confused by the lack of anything inside it, the scanner continued buzzing for a few seconds until it gave up and aborted the sequence.

Rebeccah consoled Callie by stroking her chest. The earmuffs had slipped back and dangled uselessly around Callie’s neck.

I gave her some more hot dog pieces and repositioned the earmuffs to try again.

Callie went inside and once again assumed the sphinx position in the head coil. And once again, as soon as the scanner started shimming, she backed out.

After two more tries, I was beginning to get frustrated. I sat on the doggie steps and petted Callie on the head. She just looked at me as if to say,
I’m trying.

Even with the extra training, the noise was worse than we had anticipated. Plus, the earmuffs kept sliding off. Maybe if we could get the earmuffs to make a better seal and stay on, the noise would be tolerable.

Rebeccah had the same thought. She rummaged through her vet tech gear and pulled out some gauze pads and a roll of nonadhesive tape—a clingy material called vet wrap. While I fed Callie a constant stream of hot dog, Rebeccah carefully placed a gauze pad between each of Callie’s ears and the earmuff. To keep it in place, Rebeccah wrapped her whole head with the vet wrap. Amazingly, Callie didn’t seem to mind this procedure. She ended up looking like she’d sustained a serious head injury, but the earmuffs were secure.

“Let’s try it again,” I said. Callie didn’t hear me, which was a good sign. I just motioned to the magnet, and she trotted in.

The magnet went through its click-whirrings, and I braced myself for the swarm of bees. I gave Callie some hot dog, and this time she
stayed put. It seemed to go on forever. After thirty seconds, the buzzing was replaced by three brief klaxon sounds of the localizer. That meant the scanner had successfully shimmed the magnetic field and acquired an image. I gave Callie some more hot dog and ran to the control room to see what we had gotten.

Rebeccah giving Callie the wrap treatment.
(Bryan Meltz)

Callie wrapped and ready to rock!
(Bryan Meltz)

First localizer image of Callie.
(Gregory Berns)

Robert and Sinyeob were smiling and giving the thumbs-up sign through the window.

Sure enough, there on the scanner console was a blurry but clearly recognizable image of Callie in profile. Her brain and spinal cord were unmistakable. Everyone gazed in amazement at what was surely the first MRI image of a dog in a natural position. Every other image like this had been obtained in anesthetized dogs who had been intubated and who had their necks hyperextended unnaturally. It was eerie seeing Callie’s brain transition into her spinal cord and how it ran down her neck just behind her trachea.

Next up were the functional scans. Robert opened a box on the screen to set the field of view. We oriented the FOV so that it was as if Callie’s head were a loaf of bread; the MRI would digitally slice her brain face-on, in what is called the
coronal
plane. Each slice would be 3.5 millimeters thick and, with twenty-five slices, that meant the depth of the box would be 87.5 millimeters—just under 3.5 inches. It didn’t leave much margin for error. If Callie placed her head in a different location, her brain would be out of the FOV, even if she held it perfectly still. I hoped the chin rest would do its job.

I looked at the clock. It was three o’clock. We had already burned up two hours just to get to this point, and we hadn’t even done anything with McKenzie yet. I made a mental note to switch to McKenzie at three thirty.

Callie’s energy level had noticeably decreased. She trotted up the stairs into the magnet, but now she wasn’t wagging her tail as much. This was actually a good thing. When she wagged her tail in the magnet, the fishtailing motion caused her head to move in rhythm with her tail.

Andrew took up his position at the rear of the magnet. He held a small box with four buttons. The button box was connected to a computer in the control room. Each time Andrew pressed a button, the computer would log the exact time after the beginning of the scan. One button represented the beginning of each trial, when Callie placed her head in the chin rest. Andrew would push a second button when I put up the hand signal for hot dogs, and he would push a third button when I gave the signal for peas. A fourth button would be pushed when I actually delivered the reward. By logging the time of each of these events, we would be able to match them up to the corresponding scan number.

Callie had settled in and looked at me expectantly. I gave her a hot dog and yelled, “Good girl!” I nodded to Andrew to indicate she was ready.

The scanner emitted a few clicks and then launched right into the jackhammer sounds of the functional scans. This was the sequence Callie had been practicing with for the last month, but it was much louder than the localizer. She lifted her head up from the chin rest and started to retreat, but then she paused. Half in and half out, she stared at me. I held up a piece of hot dog. Callie thought about this for a second, and then scooted forward to lick the treat out of my hand. Satisfied that nothing bad was going to happen, she settled into the chin rest.

Out of the corner of my eye, I could see Andrew hit the first button. He was holding up one finger—the sign for hot dogs.

In my head, I counted two Mississippis and put up the hot dog sign. Callie’s eyes dilated. I tried to count another six Mississippis before giving her the treat. It was just like practicing at home.

We kept at it. Not every repetition was a success, though. One time, I bumped the patient table as I reached in to give Callie her food. The slight jarring spooked her, and she backed out. Amazingly, she retreated only partway and came back when I held up the food.

After what seemed like an eternity, the scanner stopped.

Callie was waiting for me at the end of the patient table. I gave her a big hug and a handful of hot dog before escaping to the control room, where Robert was scrolling through a series of images on his screen. Most had snow. Occasionally something doglike would appear at the bottom of the screen, only to disappear a few images later.

Nothing.

“She wasn’t in the field of view,” Robert said.

My heart sank. Callie had performed so well, but her position during the localizer had been different. Without knowing it, we had programmed the scanner for the wrong location. Callie’s brain was nowhere to be seen.

We had come too far to give up, and Callie showed me that she could do this.

“Let’s switch to a dorsal orientation,” I said.

Staring at Callie’s localizer image, I realized what should have been obvious all along. The dog brain is longer front to back than it is from top to bottom. To better match the flattened shape of the brain, it made sense to take slices from top to bottom, which is called the
dorsal
orientation. Unless we took very thick slices or a lot of them, the FOV is a rectangular brick that is larger in the plane of the slices. By rotating the FOV to better match the flattened shape of Callie’s head, we would be much more likely to capture her brain regardless of where she put her head down.

Using the cursor, Robert rotated the FOV ninety degrees. It was now aligned parallel to Callie’s brain.

Between the scan and the fiddling at the console, we had burned up another thirty minutes. I broke my internal vow to switch to McKenzie. “Let’s try this one more time. Then we’ll give McKenzie a shot.” Callie was lying on her side next to me. She was tired. I reached down to stroke her. Her tail thwacked the floor, indicating that she still had some juice left.

By now the team had settled into a routine. Rebeccah secured the earmuffs and Andrew took up his position at the rear of the scanner. Callie, now either bored or depleted of energy, sauntered into the magnet. She didn’t so much as flinch when the sequence started.