Un blog de Universitat Politècnica de Valéncia, Campus de Gandia.

Demystifying Columbia

Facebooktwitterredditpinterestlinkedinmail

Francisco Camarena tells us about his personal experience as a researcher in a pioneering lab in his field in the U.S.


Between September 2010 and August 2012 I did two research residencies lasting six and three months respectively in the Ultrasound Elasticity Imaging Laboratory (UEIL) at Columbia University in New York. During this period, my team at Campus Gandia and I worked on two projects related to the use of  in the field of medicine.

The first project tried to monitor, with the use of sound waves, the moment when a tissue was cauterized interestingly enough, also with sound waves. The second project tried to use ultrasound to open windows in the capillaries of the brain and this way introduces drugs for the treatment of certain neurological diseases.

The UEIL is a relatively young group. It was founded in 2004 and today it is one of the worldwide reference groups in the field of medical ultrasound. It publishes a dozen articles a year in high impact journals; it registers an average of three patents a year and maintains partnerships with large companies and research centers in the sector.

These are, in my opinion, some of the keys to its success:

1. The content is more important than the container.

The contents are the researchers and the laboratory equipment; the container is the building where they are located. When I arrived there was no place for me, literally, we couldn’t fit in the lab and we had to use the chairs and tables of our colleagues who were away on vacation, or at a conference, or sick. During my second residency they moved to a larger laboratory. Even after the move they were still tight for space.

It is true that square meters are a prized commodity in Manhattan and that Columbia has a serious space issue in a densely populated city, but still, the priority is to invest in researchers rather than infrastructure.

2. People come and go, but there are lots of people working.

The researchers we’re talking about typically haven’t studied at Columbia, many are not even Americans; they come, spend a couple of years working, and leave. The Asians and Europeans I knew got a research position in their respective countries when they finished their postdoctoral residency. Others were hired by companies to conduct R&D. In short, many people leave the group each year, but that doesn’t mean that the group stops investing in new researchers.

3. Meetings

If I could copy a single characteristic of UEIL I’d choose the format of their meetings. This is by far, the key that makes them such and efficient team. Since the analysis of this aspect is quite extensive, I’ll leave for a future blog entry. I’ll only advance that the meetings allows them to organize the work, anticipate problems, transfer knowledge among team members and stay on the cutting edge of research.

4.  Physical proximity to the problem

The UEIL laboratory is not on the Columbia University campus, but in the building of New York Presbyterian Hospital, more than thirty blocks away. It was normal to bump into patients and their families on your way to work every day. If you’re going to do medical research, it’s best to do it near patients, lest we lose sight of the goal. On the other hand, this proximity facilitated the use of the hospital’s own material, as well as conducting experiments with animals and people.

5. Pyramid organizational structure

The working group has a distinctly pyramidal structure. Many workers. One boss. The base of the pyramid is comprised of the students of Biomedical Engineering (9) plus PhD students (10), on the next level we find the postdocs (6), and lastly the leader of the team, sometimes assisted by a research associate. The UEIL is created to support the figure of the Principal Investigator, who they endow with students and infrastructure. Inverting the pyramid would mean that there were fewer researchers focused on experiments (decreased productivity) and more researchers concerned with defining the direction of the team (leadership issues).

 

Columbia

Columbia University (New York)

 6. A prepared, but young leader

I imagine that the people who decided to hire the director of UEIL took two things into consideration: first, to find a top researcher with a solid CV; and second, that he/she was young enough to carry out most of their career at that institution. Let’s say, less than thirty-five. This way they ensured reaping the best fruits of their career. Research takes time and physical energy, so, like in football, you should invest in someone that already shows promise, but has all their growth potential still intact.

7. Teamwork.

This is repeated time and again, but there’s no way around it. If you want to do great things, you need teamwork. It is not the only ingredient, but it is necessary: ​​many people, many achievements; few people, few achievements. All the UEIL projects include at least four researchers, and each one of them has a pyramid structure.

8. One-on-one discussions.

A climate of trust has to exist in order for people to talk about science, and that climate should be promoted by the team leader, and in the case of UEIL, he knew when to be quiet, to listen and argue with his students as equals. Let’s not forget that in many ways a senior researcher may know less than a student. For example, who knows more about the technical aspects of last ultrasound scanner acquired than the PhD student who spent eight hours a day trying to make it work? Refusing to listen the opinions of your research team is throwing away money and losing efficiency.

9. Accountability.

The director of UEIL was accountable to the institutions that invested money in the group. Among the expected results were the economic ones resulting from patents. Some in the form of reputation – research is one of the flags that American universities like to wave with the most pride. And others in the form of improvements in medical treatments that could be used directly in their hospitals. If there weren’t any results, I’m sure that they would close the laboratory, otherwise they wouldn’t work the way they do.

In short, we could say that to achieve a high level of scientific excellence you need certain organizational skills – inherent to any work team – and money, money is needed so that the structure is not affected, for example that the laboratory doesn’t become outdated, and above all, that the pyramid isn’t inverted. The rest, the scientific quality of the researchers, like the value in the soldiers, is assumed.

I didn’t see any geniuses, nowhere, but I found very enthusiastic and ambitious people committed to their work. Some of them are former students of ours, and I must say, they shine with their own light.


Francisco Camarena, professor and researcher at the Campus Gandia of the UPV .

You can follow him on: Blog / Twitter

 < Prev Post 
 Next Post > 
Facebook
Twitter
YouTube
LinkedIn
Instagram
RSS
Flickr
SlideShare
women and science comunica2 BANNER CIENCIA CERCANA_ING Scientific output R&D&I