Abdullah-Al-Zubaer Imran

From Fully-Supervised Single-Task to Scarcely-Supervised Multi-Task Deep Learning for Medical Image Analysis

University of California, Los Angeles (UCLA)

Hi! Tell us a bit about yourself – what stage of your PhD are you in and what’s your project about?

Hi! I am Abdullah, born and raised in Bangladesh. After completing my Bachelor’s in Computer Science & Engineering from the Rajshahi University of Engineering & Technology (RUET), I moved to the States to pursue an MS in Computer Science. While working on my Master’s thesis at Delaware State University (DSU), I got fascinated in the machine learning for medical imaging research and decided to go for a PhD and further explore the field. I have finished my PhD in Computer Science at UCLA in Spring this year and I have recently started my first postdoc position at Stanford University. I loved my PhD work! I developed a number of deep learning-based algorithms, with special focus in learning from limited labeled data, ranging from single to multi-tasking. In general, my research work focuses on exploiting artificial intelligence, specifically deep learning techniques for various tasks in the medical imaging spectrum such as, image generation/reconstruction, image quality assessment, diagnostic classification, anatomical structure/lesion segmentation, disease quantification, etc.

What does a typical day or week look like for you?

This depends on the stage of any project (literature review, experimental design, implementation, or paper writing). For example, when I am working for any submission and close to the deadline, I work almost 24/7 checking the final results, doing any last moment analyses, or writing/editing the paper. Usually, I try to wake up very early in the morning and get ready for work (from home though for the time being). I start with checking my emails and to do list. Then, I like to read (or at least skim through) couple of papers while enjoying my morning coffee. Every week, I usually attend several meetings (lab group, joint group, 1:1, project), and seminars and journal club.

What’s one thing that you’ve enjoyed the most during your PhD?

Except the struggle in the early part, I enjoyed every bit of it. Especially, I would say internships and Summer schools. I research interned at three different companies during my PhD studies. Through the internships, I could explore what’s happening in a company, learn from the peers in a corporate settings, and get exposed to unprecedented problems/technologies. I also attended couple of machine learning/medical imaging Summer schools in different European countries. While learning from the experts in the field, I could also make a lot of new connections and learn from other PhD students coming from different parts of the world.

What’s been the most challenging part of it?

To overcome my personal, mental hurdles. Sometimes I was very hard on myself as I felt whatever I was doing probably not good enough. Another challenge was to come up with a concrete proposal for my dissertation, something that I could defend in front of my committee and accomplish the milestones within the time-frame. While overcoming the challenges, in the process, I grew up as an independent researcher, when I became able to prioritize my professional character over my personal/emotional feelings. I was really fortunate to have an excellent, supportive supervisor and I am truly grateful for that.

Where do you see yourself 5 years after completing your PhD?

I hope to continue working in academia! I just love doing research! I want to get to PI at some point down the line with my own grant, and would love to set up a new research lab to explore science for solving unsolved, important healthcare problems working with my students. My long-term goal is to develop advanced, affordable, and accessible AI-powered healthcare tools, especially to benefit the most vulnerable populations.

What’s one piece of advice that you’d offer students that are thinking of doing a PhD?

Give some thought if you really wanna go for PhD. Do your own research and talk to the experienced people around. Pick some topic/problem that you are passionate about and would love to make contributions in. Remember, PhD is long journey. Set your objectives and what you wanna achieve in your PhD and afterwards. Be prepared to accept failures and learn from your mistakes. During the PhD everyone is challenged technically but is also challenged personally. You will face and have to grow in a personal sense beyond just technical.

What makes your university a good place to study?

Definitely the location and the beautiful weather of LA. One of the world’s most ethnically and culturally diverse communities, UCLA has representation of students from all 50 states and more than 120 foreign countries. The quality of UCLA’s academic program is supported by the exceptional resources and facilities. You can enjoy all different things, the amazing city of Los Angeles, beaches, mountains, forests, Hollywood as all are close by. You will never feel bored as you can always find something for fun. Moreover, the vibrant and beautiful UCLA campus makes it really a unique place to study.

Lastly, what do you enjoy doing in your free time?

I don’t have anything consistent though. I love to do some free-hand exercise and enjoy sports (mostly watching). I enjoy sharing my experience and knowledge mostly through my social media profile, especially for the students in Bangladesh. I like meeting friends, traveling, and driving. I also like enjoying my time sitting in a coffee shop or a park bench. I enjoy reading and running too.

Want to learn more about Abdullah?

Connect with him on Twitter and LinkedIn and check out his website using the links below:

Website | Twitter | LinkedIn