Research Statement

I still find it hard to imagine how people got around before GPS was a thing or how anyone could wait for days, if not weeks, to receive a response before telephones - in the not too distant past. Couple years later, would people wonder "How did people write and/or code before LLMs"?

Back towards the end of 2020 as I was wrapping up my internship at Biogen (perhaps the best 6 months of my life, as of September 2024, will probably write a blog on it next time I want to procrastinate), I decided that I wanted to pursue a PhD developing interpretable, handcrafted imaging biomarkers. The struggle of writing the statement of purpose (SoP) was REAL (and I still do struggle with writing a lot)! Back then, Prof. Priya Narasimhan's tweetorials helped a lot - Link 1, Link 2. Perhaps the one tweet that really helped me get started was

You can always edit a bad page. You can’t edit a blank page.
— @jodipicoult

— Prof. Priya Narasimhan (@YinzcamPriya) November 14, 2020

I spent several hours sharpening the axe - reading SoPs and guides on writing SoPs. Nicholas Matiasz's SoP stood out, and I spent a decent amount of time in analyzing - not the contents per se but the structure and organization. To that end, I feel it is time to pass along and share my one page research statement - a condensed, to the point version of my SoP. I joined the Biomedical Engineering PhD program at Case Western Reserve University with Dr. Anant Madabhushi in Fall 2021 (I moved with him to Georgia Tech and Emory University a year later). Anyway here it goes:

Research Statement

Medical imaging is widely used in clinical practice for diagnosis, prognosis and treatment evaluation of several diseases, yet its interpretation is restricted to visual inspection by radiologists. Quantitative imaging biomarkers currently have limited use as surrogate endpoints in clinical trials. The clinical translation of these biomarkers can provide accurate and valuable information to physicians and enhance clinical decision-making. The development, validation, and clinical translation of imaging biomarkers presents a remarkable opportunity for enabling personalized medicine and transforming patient care. A Ph.D. in Biomedical Engineering from Case Western Reserve University will help me develop the technical expertise, analytical mindset, and scientific rigor required to pursue a successful career in the development of such novel biomarkers in the healthcare industry.

During my undergrad, I focused on developing a strong understanding of signal processing fundamentals. This helped me easily grasp and apply the concepts of speech processing for extracting vocal biomarkers of depression for my senior year project "Depression Detection using Speech". In my project consultation with the college’s counseling psychologist, I learned that an objective screening tool for depression could be an invaluable addition at the point of care. Driven by the idea of developing technology to assist doctors in diagnosing diseases, I decided to pursue a Master’s with a specialization in Biomedical Signal Processing and Data Science.

In my Master’s, I became interested in Dr. Olaf Sporns’ work in brain connectomics to model the connectivity and interactions between different brain regions using network science based measures. I will be working with Dr. Liang Zhan on investigating Alzheimer’s Disease related changes in diffusion MRI derived brain connectome networks for my Master’s thesis. Over the course of my Master’s, I have also realized that translational research demands both accuracy and interpretability. Dr. Leo Breiman’s landmark paper “Statistical Modeling: The Two Cultures”[1] has further influenced me, and in my Ph.D. I aspire to broaden my skill set with the knowledge of statistics as well as algorithmic methods and hone my judgement of selecting appropriate tools for developing practical healthcare solutions.

I joined Biogen as a co-op in the Personalized Health, Research, Analytics and Solutions group where I developed and validated an algorithm for automated assessment of image quality for brain magnetic resonance images (MRIs). With iterative designs, I devised the algorithm to be scanner independent and robust to a multitude of image quality issues. At Biogen, I have closely witnessed the efforts involved in the development of quantitative imaging biomarkers for Multiple Sclerosis. A profound understanding of biology, engineering, computer science, and statistics is necessary to hypothesize, develop and validate these biomarkers. These biomarkers must also be compatible with hospital infrastructures, easily integrable in clinical workflows, as well as reliable, interpretable and intuitive to the clinicians for successful adoption in clinical decision-making. The ethical responsibility and regulatory guidelines are important at every step of the process. Above all, I learned the intricacies of working in a collaborative environment, being focused in research and setting pragmatic goals towards solving a problem. My co-op at Biogen augmented my desire to pursue a Ph.D. and fortified my intentions of leveraging technology to enhance clinical-care.

I am interested in developing statistical models & algorithms for automated detection, quantification and assessing the clinical correlations of disease related imaging abnormalities. I find Dr. Anant Madabhushi, Dr. Satish Viswanath, and Dr. Pallavi Tiwari’s work at Center of Computational Imaging and Personalized Diagnostics (CCIPD) in quantitative imaging to monitor disease progression highly relevant to my research interests and career goals. The opportunity to participate in collaborative research with Cleveland Clinic will help me understand clinical workflows and physicians' perspectives, which would be vital to incorporate in healthcare solutions I develop. CCIPD will provide me with a platform to synthesize my technical and clinical knowledge to enable new insights in medical practice. With a Ph.D. in Biomedical Engineering from Case, I envision myself in the healthcare industry; building solutions for unmet clinical needs, ultimately improving patients’ quality of life.

References: 

[1] Breiman, L. (2001). Statistical modeling: The two cultures (with comments and a rejoinder by the author). Statistical science, 16(3), 199-231. 

PS: No ChatGPT/LLM was used while writing this blog.