During my time as a Research Assistant in the AIDAK research group at Linnaeus University, Växjö, Sweden, I had the privilege to work on cutting-edge projects focused on the application of artificial intelligence (AI) in medical imaging. My primary research centered on utilizing deep learning techniques to enhance diagnostic capabilities, specifically in the detection and classification of Acanthamoeba Keratitis cysts. This experience allowed me to merge my academic knowledge with practical research, contributing significantly to advancements in the field of medical AI.
One of my key contributions was the design and implementation of a replacement system for a previously published paper in a reputable journal. This project involved a comprehensive understanding of the existing methodologies and a critical assessment of their limitations. By leveraging state-of-the-art deep learning models, I developed a more efficient and accurate system that significantly improved the performance of the existing solutions. This replacement system not only enhanced the accuracy of cyst detection but also streamlined the process, making it more robust and reliable for clinical applications.
In addition to my primary research responsibilities, parts of my work were published as my bachelor thesis titled “Deep Learning-Based Pipeline for Acanthamoeba Keratitis Cyst Detection: Image Processing and Classification Utilizing In Vivo Confocal Microscopy Images.” This thesis encapsulated the essence of my research, detailing the innovative pipeline I developed for image processing and classification. The use of in vivo confocal microscopy images was particularly noteworthy as it provided a non-invasive method for obtaining high-resolution images crucial for accurate diagnosis. My thesis was well-received, reflecting the approval of quality of my work by field professionals.
Throughout my time with the AIDAK research group, I collaborated with a team of experienced researchers and clinicians, which enriched my understanding of the interdisciplinary nature of medical AI. This collaboration was instrumental in ensuring that the solutions we developed were not only technically sound but also clinically relevant. It was a rewarding experience to see how our work could directly impact patient care by providing tools that aid in the early and accurate diagnosis of eye infections.
Furthermore, I am excited to announce that our team’s research will be presented in an upcoming scientific paper at a prestigious conference. This publication will highlight our advancements in AI-based medical imaging and showcase the innovative techniques we developed. Being part of this conference is a significant milestone, reflecting the impact and relevance of our work in the scientific community.
Overall, my experience as a Research Assistant at Linnaeus University has been profoundly impactful. It has equipped me with advanced skills in AI and medical imaging, honed my research abilities, and underscored the importance of interdisciplinary collaboration in addressing complex healthcare challenges. As I continue my career, I am motivated by the potential of AI to transform medical practices and improve patient outcomes, and I am eager to contribute further to this exciting and vital field.