KAUST Application: Craft an Impactful Statement of Purpose

Statement of Objectives

How can we utilize artificial general intelligence to benefit all of humanity? Inspired by this question, my current research focuses on facilitating personalized, patient-centric healthcare. Specifically, I apply machine learning algorithms to move beyond one-size-fits-all treatments and tailor medical interventions to each patient's unique needs. This personalized approach aims to predict how each person will respond to different therapies, leading to more effective treatments and improved healthcare outcomes. It is with these broad visions in mind that I am applying for the KAUST Visiting Student Research Program (VSRP) Internship 2024.

Finding My Research Interests. With a focus on computer vision, my research journey began under Prof. Mohamed Uvaze Ahamed Ayoobkhan at New Uzbekistan University. I was particularly interested in image classification, a fundamental area within computer vision with vast applications across various sectors. The recent advancements in deep learning have significantly enhanced the performance of image classification systems. This work led to the very first research paper I co-authored, Machine Learning-Based Image Classification with COREL 1K Dataset (ICSDI 2024). My main contributions include identifying effectiveness of various machine learning (ML) algorithms for image classification on the COREL 1K dataset. We investigated a range of techniques, including decision trees, k-Nearest Neighbors, and Support Vector Machines (SVM), to determine their respective strengths and limitations. We explored the influence of different feature extraction techniques, such as Local Binary Patterns (LBP), color histograms, and Gray-Level Co-occurrence Matrix, on classification accuracy. Our findings revealed that SVM, coupled with LBP feature extraction, achieved the highest accuracy at 87.5%, highlighting the crucial role of selecting appropriate feature extraction methods and ML algorithms in optimizing image classification on this dataset. I truly enjoyed my experience in research more than the industry for the ownership over my work. But, I also had some burning questions regarding my future research interests. Meanwhile I was engaged by the technical aspects of solving real-world problems, I wanted to figure out something that would really excite me - what is the thing that would get me out of bed every morning? And how could I find it?

My next project, Developing Medical Software that helps with making diagnostics decisions based on Brain MRI scan results, helped me answer those questions. Working under the supervision of Prof. Mohamed Uvaze Ahamed Ayoobkhan, we set out to address the challenges that radiologists are facing today. Prior to our work, the approach to make diagnostic decisions was solely based on expert opinions. To help radiologists make proper diagnostic decisions informed by Brian MRI scans, I designed and developed a mobile application to give them a user-friendly way to use AI and make sense of observation. In February 2024, the ML Community hosted a AI hackathon at the New Uzbekistan University, and I was proud to present this work. Through zooming in and out on a pressing, real-world issue, I realized what I should be looking for in the research I pursue: the possibility of helping others and the insight into real-world issues that would spark that possibility. I started to envision making an impact on the real world through my research. The value of our work in the scientific community can only be actualized when our tools are adopted in hospitals and clinics.

While we discovered numerous applications designed to treat patients or expedite processes within the healthcare system, we observed a significant gap in patient support and community building. Existing solutions often lacked features for direct patient-to-patient & patient-to-professional online communication, personalized support networks, and ongoing engagement beyond episodic care. With that overarching goal in mind, I initiated a project called CoMed. CoMed will research and develop groundbreaking medical software solutions, including CoMedAI, CoMedAI 2.0, and Medicord, that promise to revolutionize the way Brain MRI examinations are conducted, analyzed, and managed across the healthcare industry. I presented a new approach to developing advanced software that is designed to seamlessly integrate into the workflows of hospitals and clinics, enhancing the accuracy of diagnoses while fostering a supportive community among healthcare professionals and patients all over the globe. I designed a comprehensive business plan with the mission of growing it into a successful startup. Through this project, I found myself enjoying both scoping and solving open-ended problems and hope to further improve with additional formal training in research programs.

Future Work. All my experiences collectively shaped my research interests and motivated me to pursue The Visiting Student Research Program (VSRP). Today, personalized medicine is revolutionizing healthcare by tailoring treatments to individual patients. Seeing the critical challenge of radiologist shortages, especially in rural areas or underserved communities and the potential of data-driven decision-making in areas outside of CS (e.g., healthcare industry) incites my urge to build my work around the theme of streamlining the analysis workflow and empowering radiologists with valuable quantitative data for more precise and nuanced diagnoses. Specifically, I would be excited to work with Professor Raul Tempone. Professor Tempone is a leading expert in numerical analysis and uncertainty quantification, areas crucial for developing robust and efficient algorithms for data analysis. His work on enabling data analytics for individuals outside of CS using ML-inspired techniques aligns perfectly with my research interest in applying AI to personalized medicine. While his focus is on broader data analytics, I believe his expertise in developing numerical methods for stochastic models could be applied to the complex challenges of analyzing medical data. For example, his work on Bayesian model calibration and validation could be valuable in developing and refining personalized treatment plans based on individual patient responses and outcomes. I would be excited to work with Researcher Raul Tempone.

Where I See Myself. As a research-oriented undergraduate student, participating in research programs is the most effective way to explore new opportunities and shape the world. Although I spent a year learning Artificial Intelligence and Machine Learning at Academy. While studying full-time, I have also worked part-time in a startup, as a research assistant, writing tutorials, mentoring, and facilitating. Through these valuable experiences, I not only learned about the many real-world challenges that made me more resilient, but also discovered research interests that would allow me to address some of those challenges. After the research program, I aim to pursue a graduate degree, so that I can develop the research and tools to address these challenges and more. Furthering my research at KAUST would bring me one step closer to my goal of advancing data-driven decision-making in a wide range of fields and ultimately improving patient outcomes and making a meaningful impact on global healthcare systems.