Taking a Different Road – Being a Statistics Major

The following is written by Sarah Lotspeich, University of Florida who attended the SAMSI Undergraduate Workshop focusing on Computational Neuroscience.

I declared my Statistics major in the eleventh grade, approximately halfway through my AP Statistics course. As everyone around me pondered medical school and the many types of engineering, I knew that my choice seemed unconventional. Now three years into my undergraduate degree, I have met only a handful of fellow Statistics majors to date. During the third week of October, however, this changed forever as I attended the SAMSI Undergraduate Workshop.

Duke Chapel

Duke Chapel.

It was a gorgeous fall day (a pleasant surprise for me, as my typical “fall” in Gainesville, Florida includes a few fallen leaves and a high temperature in the 80s) in Research Triangle Park, North Carolina. Budding statistics and mathematics students from across the country gathered to explore computational neuroscience, and to enjoy fantastic food. Always eager for an adventure, I flew in as early as possible the day before the workshop to get maximum exploring time in Durham. Perhaps a bit TOO eager, I walked over eight miles through Downtown Durham and to both edges of Duke University’s gorgeous gothic campus.

Dame's Chicken and Waffles

Excellent chicken and waffles place!

Fret not, however, as I was well fueled by Dame’s Chicken and Waffles and fondue from the Little Dipper. Needless to say the local area surpassed my every expectation and left me excited to wear scarves and learn more about statistics the following day. The mingling began at approximately 7:30am the next morning, as over thirty of my fellow “numbers people” bonded over bagels and oatmeal. I was so excited to hear from people who care as much about significance tests and p-values as I do!

The presentations commenced with an absolute bang as Dr. Ciprian Crainiceanu of Johns Hopkins University immersed us in “Neurohacking”. He outlined the basic principles of converting MRI images from picture to a system of numbers, and by the end of the hour left us with a data set and the necessary code to explore it independently. One of my favorite components of the workshop, actually, was the interactive nature of each presentation with the integration of R or Matlab code.

Guest lecturers introduced many fascinating facets of computational neuroscience, and I especially enjoyed how my knowledge on the subject compounded with each additional lecture. As the workshop progressed I found that I was relating information from one speaker’s presentation back to material I learned even hours previously, and even today I walked away with a nice basis on the topic. It very much feels as if I went from zero to one hundred with this material, and I appreciate the challenges posed to us by the complicated subject matter.

Beyond the presentations, the field trip to the laboratory for psychiatric neuroengineering at Duke University provided a “behind-the-scenes” glimpse at the processes of data collection that create the massive sets we dealt with during lecture. I was also just happy for any excuse to ogle the beautiful campus once more. Each new speaker and opportunity brought about new questions to ask and facts to learn, so I was happy for the constantly changing environment of the workshop from lecture to lecture, or even breaks for the field trip or panel.

students by SAMSI sign

From left to right: Jordan Zeldin, Eion Blanchard, Sarah Lotspeich, Michelle Zamperlini.

The many bus rides provided unexpectedly pleasant opportunities to meet new people, as well, as I was shuffled into new groups with each trip. I thoroughly enjoyed swapping stories about my university – about the weather, everyday dress code, the statistics department – with people from other schools! And I was even lucky enough to give suggestions about things to do and places to eat in Florida, as one of my new friends is planning a trip to the Sunshine State soon. Perhaps the most unexpected bonus to this experience was the people.

This was honestly one of the most incredible groups of students, and upon learning more about each person and their involvement I am absolutely honored to have been selected among them for the 2015 SAMSI Undergraduate Workshop. Though the workshop lasted only two day, the people I met and research I was immersed in will carry through my entire career. I cannot emphasize enough the importance of this experience and how strongly I recommend it.

There is a 100% probability that I would love to return to SAMSI sometime in the future.

Learning about Clinical Brain Imaging using R at SAMSI’s Computational Neuroscience Workshop

The following was written by Katy Wang, attendee from the University of California, Riverside, of the Undergraduate Workshop focusing on Computational Neuroscience.

Dr. Ciprian Crainceanu

Dr. Ciprian Crainiceanu.

Of the seven presentations at SAMSI’s Computational Neuroscience Workshop from October 19-20th, the one that was most memorable to me was given by Dr. Ciprian Crainiceanu, a professor in the Department of Biostatistics at Johns Hopkins University. Dr. Crainiceanu’s presentation on neurohacking in R really stood out to me because I learned how to preprocess images, read, write, plot, and manipulate neuroimaging data in R.

You may be wondering “what exactly is neurohacking and how is the application of statistics used in clinical brain imaging?” As defined by Dr. Crainiceanu, “neurohacking is the continuous process of using, improving, and designing the simplest open source scripted software that depends on the minimum number of software platforms and is dedicated to improving the correctness, reproducibility, and speed of neuroimage data analysis.” The goal of neurohacking is the “democratization of neuroimaging data analysis,” in other words, to make neuroimaging data analysis possible for all people to understand. Throughout the presentation, we were shown an image of an axial slice of the T1-w image of the brain that contained a multiple sclerosis lesion surrounded by a hyper intense ring, which indicated blood with a higher concentration of gadolinium chelate. After taking the region of interest, a matrix is used with numbers corresponding to that particular section of the brain in order to understand what the dynamics of blood flow are into the lesion. To see if anything has changed (e.g, Did the brain tumor get bigger? Was the cancer eliminated by surgery?), the follow-up T1-w is subtracted from a baseline T1-w volume. A template-based analysis is also used in which an MNI T1 template is used to see which parts of the brain it maps to. The results are later quantified and mapped to neuroimage.

Students interacting with the lecturer

Students interacting with the lecturer.

Although there was not enough time to actually work with the data during the presentation, Dr. Crainiceanu offered a clear explanation on neuroimaging, an impressive tutorial with Powerpoint slides on how to set up the data, information on data structure and operations (working with various file types, visualization and data manipulation) preprocessing (inhomogeneity correction, intensity normalization, tools in R), registration, segmentation, dynamic visualization in R, and many resources in order to work with and become more familiar with working with the data. Furthermore, we were given suggested prerequisites and coursework, such as (1) Linux/Unix; (2) a basic knowledge of programming; (3) a basic knowledge of array data structures (e.g. 2d and 3d arrays), and most importantly (4) an interest in “hacking” with neuroimaging data! You may also find these Coursera Data Science Specialization courses offered on behalf of Johns Hopkins University as a helpful resource.

All in all, SAMSI’s undergraduate workshop was truly a great learning experience! I went into the workshop with very limited knowledge in computational neuroscience but came out of the workshop with several Word documents of notes, many data files/tutorials, and resources to enhance my knowledge of mathematical and statistical methods in neuroscience.

Impressions of the SAMSI Undergraduate Workshop on Computational Neuroscience – Part 1

Several undergraduate students recently attended the SAMSI Undergraduate Workshop focusing on Computational Neuroscience. Here is Suzanna Mostaghim’s impressions of the workshop. Suzanna attends Virginia Tech. She has a wonderful blog you should check out called Inklings and Tea.

The Statistical and Applied Mathematical Sciences Institute (SAMSI) holds a two day undergraduate workshop in the fall and spring as well as a week-long one in the summer.

This year, the fall workshop from October 19-20th focused on computational neuroscience. Over one and a half days there were seven scientific talks, which was incredibly intense. I personally have over six pages of notes.

I got in on Sunday night, a lot of other participants got in in the morning and afternoon due to flights and better planning than myself. As I was the only girl from Virginia Tech I was paired with a random roommate (who is in fact an awesome triple major, check out her blog here).

SAMSI was great and paid for our hotel rooms so we wouldn’t have to worry about waiting to get reimbursement for the room for a few weeks. We were staying only about 10 minutes from NISS and about 25 minutes from Duke.

Our first day started early, we shuttled over to SAMSI at 8:15, registered, and began talks at 9:00.

 Ciprian Crainiceanu from John Hopkins University

Ciprian Crainiceanu from John Hopkins University.

The first talk was by Ciprian Crainiceanu from John Hopkins University; he talked about clinical brain imaging from a statistical view, pointing out that data scientists and electrical engineers would use different techniques and names. Look out for his MOOC on Coursera early next year.

For me, I recognized this area as image deblurring – the data science point of view. This was something I had touched on in a Computer Modeling and Data Analytics course at Virginia Tech taught by Julianne Chung whose research is in image deblurring.

We sped through the first slideshow he showed on Neurohacking and moved onto Data Structures and Operations which was based on classification typing. It was intense to go through so many slides in less than two hours and absorb as much information as I possibly could.

After Dr. Crainiceanu was finished, we had a five minute break and then Ana-Maria Staicu of NC State stepped up to the plate and gave us an impressive technical primer on Tensor Imaging Study of MS.

I don’t think I ever wrote so many notes on a singular lecture outside my classes until that point. She brought down the information to the level that an undergraduate would understand and expanded upon it to show us how to approach the problems she faced while analyzing diffusion anisotropy.

students walking into building

Entering the Center for Neuroimaging at Duke.

When she finished we broke for lunch and traveled to the Center for Neuroimaging at Duke and did an overview of the facilities and once again listened to a talk. I won’t go into detail, but I’ll admit you should definitely visit them for grad school if you’re interested in biostats and psychological disorders.

In full disclosure, I think that it could have been organized a little better for the field trip. However, it was still a wonderful experience and good break in between intense scientific talks

Next we had an R tutorial from Benjamin Risk, who works at SAMSI. He sped through and gave a thorough tutorial. However, due to the faulty internet connection a lot of students weren’t able to follow completely.

After R, we got to one of my favorite talks by Sarah Vallelian. Which was about computed tomography using MATLAB and Python; something I’ve covered at Virginia Tech. But, I won’t expound upon the details. She showed us the equations for attenuation and how to use Radon Transforms when doing compute tomography.

When the talks for the day were over we attended a panel discussion hosted by postdocs and graduate students about graduate school. One in particular was very sarcastic and, while amusing, it got a bit tiring after a while. However, overall they were very helpful and gave a good amount of advice for graduate school considerations.

We then shuttled back to the hotel and groups of us conglomerated together and bonded, exchanging contact information in order to keep in contact after the workshop ended the next day.

The second day started just as early as the first and we started with a talk on models for muscle activation by Lauren Miller from UNC – Chapel Hill. She covered the simplistic models and moved onto more complex ones such as the Three-Element Hill Model, explaining on a level that we understood.

Ezra Miller at SAMSI workshop

Ezra Miller, Duke University.

Our very last talk was on Topology for Statistical Analysis of Brain Artery Images by Ezra Miller. I knew very little about topology, and when he was done I understood a lot more than I had ever learned prior to that talk.

He was incredibly engaging and broke down topology to the point where I believe almost anyone could understand. I honestly am considering applying to Duke just to have him as a professor one day.

With the end of the Ezra’s talk, the workshop came to an end. I had to go back to Blacksburg and part ways with my new found friends. But, it was definitely an experience I will never forget.