Journal - June
Week 1: May 28-June 3
My parents and I drove from my home in North Carolina to Amherst, Massachusetts on the Friday and Saturday before my research was scheduled to start. It was a long trip, especially going through New York, but we eventually made it. I checked into my dorm on Saturday morning and spent the weekend watching movies and exploring Amherst. Since Monday was Memorial Day, I didn't start working until Tuesday.
My first week was full of reading papers to get background on the projects in the lab, meeting new people, getting keys and cards for buildings and offices, a short orientation with the computer science REU students who are here this summer, and learning to use Little-Jil (the language that the Laser lab has developed to describe processes. I also attended my first of the weekly Laser meetings on Friday. Each person reports on what they've been working on, what they're getting stuck on, what they're thinking about doing, etc. and everyone discusses.
Outside of research, I met REU students in other fields during a couple of get-togethers. First, there were sandwiches on Tuesday night, and there was an REU BBQ on Friday.
{back to top}Week 2: June 4-10
My first "real" weekend in Amherst was a lot of fun. On Saturday a big group of us went to Puffer's Pond, which is a local hangout spot. There was a nice little beach to lay on, and a short trail around the pond. It'll be even better later in the summer when the water's had a chance to warm up just a little. I also went for a short jog and walked to the farmer's market on Saturday morning, and we watched Zombieland on Sunday night.
During my second week in the lab, I learned about two potential projects, and decided on a project that involves continuing to develop a visualization of medical processes. It will entail creating mockups of the proces and working with a nurse to make sure it will guide nurses and doctors through complicated processes without distracting or annoying them. There will also be weekly lunches/lectures on Thursdays through the REU program. This week, there was delicious local pizza for lunch and a talk on reinforcement learning. Then, after lunch, Dr. Jensen gave the first of a 3 part lecture series on how to approach computer science as "real" science. They were both great talks, and I'm excited for more of them!
Social life in the evenings was relatively relaxed. We took advantage of the local theater's $5.25 movie day and saw The Hangover II on Tuesday, and Wednesday night was one of the REU students' birthdays, so we went out for dinner.
{back to top}Week 3: June 11-17
On Saturday I went to Yankee Candle. It was impressively large and had rooms decorated like a Christmas workshop, a pirate's lair, etc. and full of candles with every imaginable scent. It was a little bit of a sensory overload, and my nose was happy to be back in the fresh air when we left, but it's definitely worth checking out if you're in the area. We were slow to get moving on Sunday. I went for a short jog in the morning and spent the early afternoon reading. But we had big plans for Sunday evening - ice skating in the Mullin's Center! We skated for a couple of hours ... and I'm pretty sure we all fell a few times (think face plants ... or belly flops in frozen water). Despite my scraped up knees and elbows, it was a ton of fun! Sadly, it doesn't stay open all summer (we didn't realize that until someone else mentioned it to us at the rink) so that was the last day it was open for public skating.
During the week, I really got started with creating mock-ups of the medical processes. I'm starting off working on visualizing the prerequisites that certain steps have. Sometimes a prerequisite is another step, sometimes it is a simple condition, sometimes the nurse/doctor can satisfy the prerequisite themself, but other times another doctor/nurse has to do something before anything else can be done, so there's a lot to think about.
We had two more great lectures on Thursday. First, we heard about information retrieval research. It really does sometimes seem like Google (or other Internet-based search engines) have got this completely covered, but the speaker pointed out that there is actually a lot of room for improvement; for example, those search engines don't consider semantics, which is sometimes really significant. The second lecture was part two of professor Jensen's research methods talk. He spoke about what really makes science hard (like our own biases, relying on theories that may or may not be totaly true, etc.) and how to combat those issues.
{back to top}Week 4: June 18-24
Another really busy weekend! We got up on Saturday morning, armed ourselves with a hearty french toast breakfast, and headed out to do some strawberry picking. The picking was GREAT! Between six of us, we picked about 20 pounds of strawberries within just around fifteen minutes! We dropped off our booty and headed to Puffer's Pond for a couple of hours. It was a lot warmer out than last time, but the water was still chilly. We hope to get tubes or floats for next time. Around dinner time, we headed back, cleaned up, and got back together to go to Taste of Amherst. Taste of Amherst is a sort of festival where all of the local restaurants set up stands in a big field and sell smaller portions of their meals for (sort of) low prices so that you can try lots of different food. I had lo mein from a Chinese restaurant that I've heard good things about and fried dough with powdered sugar. Yum! Sunday was much more relaxed, but not devoid of fun by any means. We put those strawberries to good use by making strawberry shortcakes, and watched Almost Famous.
I kept working on the visualization this week. I've made updates that allow multiple patients and multiple processes per patient to be displayed. I'm working on mocking up a few different visualizations of preconditions. That turns out to be a harder problem than I expected. As I get further into the visualization, it also becomes harder and harder to keep track of the decisions I've made about how to represent the processes; I think I will need to create a key of some sort very soon.
Meetings this week were especially interesting. A student working with another group in Boston came to the medical safety meeting on Monday. Their group is interested in using the Laser group's Little-Jil language and other software in conjunction with technology they are developing (or helping with) that monitors the various machines used in medical processes, so he had lots of questions about how to use the language and tools effectively, and the discussion also generated interesting ideas for the Laser group. For example, the tools currently only work with static analysis of the problems (i.e. they have the process and can use the tools to find potential sources of error), but not dynamic analysis (i.e. they don't currently support the ability to keep track of a process in real time and figure out when an error has occured). I went out to lunch with my mentor, Lori Clarke on Tuesday. We talked a little bit about graduate school (I'm sure we'll talk more as the summer goes on), and also about projects in the lab. I was really curious how the lab got involved in the medical safety project. The story's actually pretty neat; one of the professors in the lab was department head for a while and, therefore, sat on stage during commencement. He was right next to someone from the nursing department and they started talking about all of the potential soures of error that are the result of really complicated processes.
The Thursday lectures were great again this week! We first heard about some work that's going on in the department with mobility and manipulation for robots. One interesting decision they've made with their robot is to have it automatically balance itself, much like a segway. This means that it can have a smaller base (smaller footprint) and not be as obtrusive in a crowd of people as bigger machines, and it's also more stable (and the taller the robots get, the more stable they'll be). We saw quite a few videos of the robot performing and learning various maneuvers and heard about some of the specific problems they've been trying to solve.
Lori gave this week's longer lecture. She outlined how to apply to graduate school and what kinds of choices there are. There was a lot of great information!
{back to top}Week 5: June 25-July 1
I spent this weekend in New York City! It was my first time visiting The City that Never Sleeps, and I had a great time! I took a bus that left from the Hadley mall (about a 5 minute drive from the dorms) at 4:30am and arrived in NYC around 8am. I stayed with a friend of mine, Victoria, has a journalism internship there this summer. She met me at the bus stop, and we walked back to her place in Union Square. We did a little bit of shopping in Union Square while waiting for another friend, Virginia, to arrive. A little after noon we left to pick Virginia up from her bus stop, conveniently located near the Empire State Building and Madison Square Garden. Check and check on the tourist to-see list.
Yet another friend from school, Allison, was also coming for the weekend, but Virginia and I decided to explore the city while Victoria caught up with her. We took the Metro to Times Square (wow!) and saw, among other things, the Disney Store (I bought a t-shirt with Minnie Mouse dressed up as the Statue of Liberty), the Toys R Us with a ferris wheel in the store, and tons of people in ridiculous outfits. We took a short ice cream break in McDonalds while deciding where to go next. I suggested we find the FAO Schwarz that I've heard so much about. We didn't realize until we were out of the Metro station near FAO Schwarz that we'd conveniently also located the famous Central Park and 5th Avenue! The toy store and the area in general totally lived up to my expectations. The atmosphere in that part of New York is remarkably different from other areas; it's kind of bizzare. We finally headed back and met up with Victoria and Allison in time for a late dinner at an Italian place.
We got a little bit of a late start on Sunday. Our first stop was the Gay Pride Parade, which was very entertaining. And I thought there were a lot of goofy outfits in Times Square. Virginia and I, once again, took off on our own after just an hour or so of the parade. This time our destination was the Statue of Liberty. We took a ferry to the island and walked around Lady Liberty; it's almost impossible to get tickets to actually climb to her crown. Even though I feel like I saw a LOT of NYC, my to-see list is still really long: Grand Central Station, Ellis Island, Coney Island, Wall Street, the Rockefeller Center, Stanton Island, and Ground Zero. Good thing some of the REU students are hoping to go later this summer!
I didn't get much actual work done this week, but I definitely learned a lot. Researchers from France visited the lab. They have similar research interests - they're looking into software development processes (which, it turns out, have some similar properties/challenges to the human-intensive processes that the Laser lab is interested in). At our weekly Medical Safety meeting this Monday, I presented the progress I've made on the visualization of medical processes. I got some really good feedback on what was good and what still needs work.
There are some interesting disagreements and conflicting goals. For example, when a Little-Jil process runs, there's a concept of the step being "started" and "completed." Sometimes this is also a useful way for doctors and nurses to think of the process; the actual administration of chemo or transfusion of blood take a while, so it makes sense to start and end them. Many other steps, like checking a patient's ID, however, are atomic events and don't need to be thought of by nurses/doctors as being started, just completed. In general, doctors and nurses want a system that requires as few clicks as necessary. However, this sometimes conflicts with other interests. The physician that came to our medical safety meeting, for example, suggested that it is a bad idea to be able to switch between the processes for different patients simply by clicking another tab (as we have proposed), because that could lead to performing the process steps on the wrong patient. In this case, we will probably need to introduce more clicks or some other sort of check to switch between patients. The physician who collaborates with the lab and was at the meeting invited me to come see the current system at Baystate Medical Center on Friday. I think that will help a lot with my work, and I'm really excited to go!
The rest of the week was devoted to presentations. The visiting researchers gave presentations on their work, and everyone in the Laser lab gave a presentation on their current work. It was really neat to see exactly what everyone is working on! This Thursday's REU lunch talk was given by a graduate student from the PLASMA lab on how modern processors work and how to improve their performance.
{back to top}