In order to showcase the work we do in Pattern-Cog, European project funded by ERA PerMed, we caught up with Elaheh Moradi postdoctoral researcher at the University of Eastern Finland, and asked her a few questions about her work, expectations and challenges.
- Tell us a bit about you and the institution you work for
I am a data scientist currently working as a postdoctoral researcher in the Biomedical Image Analysis Research Group at the University of Eastern Finland. My primary research interest lies in the application of advanced machine learning techniques and statistical pattern recognition approaches to analyse a diverse range of data sources, including imaging, genomics, and clinical data.
Within the University of Eastern Finland (UEF), our research team focuses on the development of computational methodologies for the analysis of brain imaging data. We rigorously evaluate these methodologies through advanced simulations and openly share the software implementations of these approaches. Moreover, we are part of UEF’s multidisciplinary Neuroscience Research Community (NEURO RC). The aim of NEURO RC is to understand the disease-specific and common molecular mechanisms underlying neurodegenerative diseases and epilepsy and to identify novel biomarkers and therapeutic approaches for their prevention and cure.
Beyond my role as a data scientist and postdoctoral researcher, I am genuinely passionate about the application of data-driven solutions in healthcare. My academic background has advanced my knowledge in the biomedical research area, which has been crucial in shaping my research path. Working in the dynamic and cooperative research environment at the University of Eastern Finland has enabled me to engage closely with individuals who possess a wide range of skills and knowledge. This has not only facilitated my growth in interdisciplinary thinking but also enhanced my ability to understand complex research problems and make meaningful contributions to the pursuit of knowledge.
- What is the focus of your work within the Pattern-Cog project?
In the Pattern-Cog project, our primary objective is to harness the power of Machine Learning to develop predictive models that can identify early signs of Alzheimer’s disease in healthy individuals. My central role within this project involves integrating diverse data sources, including neuroimaging data, blood biomarkers, genetic information, cognitive assessments, and potentially other relevant factors. Our goal is to effectively combine these data sources to create a predictive model that can detect and evaluate an individual’s risk of future cognitive decline.
One notable aspect of my work is the utilisation of longitudinal data. This involves incorporating data collected by tracking individuals over time, which enables us to capture changes and patterns that may indicate the development of Alzheimer’s disease long before clinical symptoms become apparent. By integrating this longitudinal data into our models, we enhance their performance and increase their ability to make early and accurate predictions.
In essence, my role revolves around integrating and analysing longitudinal and multimodal data, all with the aim of contributing to the project’s overarching goal: the early detection of the earliest signs of impending cognitive decline and the identification of markers that will enable early and personalised multidomain interventions.
- What do you enjoy the most about your work on the Pattern-Cog project? What do you find most challenging?
What I like most about working on this project is the possibility of making a real-world impact. Knowing that our work could lead to early identification and better outcomes for people at risk of Alzheimer’s disease is extremely rewarding. I particularly appreciate the Pattern-Cog project’s collaborative aspect, since it allows me to interact closely with experts from other disciplines, which influences my own learning and skill development. Furthermore, staying updated with the latest developments in machine learning and data analysis tools keeps me motivated to investigate novel solutions for challenging healthcare problems.
One of the most significant challenges I encountered in this project is the complexity of the data we are dealing with. Combining and harmonising diverse data sources, particularly when dealing with longitudinal data, can be difficult and time-consuming. Additionally, the limited availability of data, especially in certain data modalities and particularly for longitudinal data analysis, presents a significant challenge.
- What are your expectations and what do you think is the importance of the project for the wider field?
I have two primary expectations for this project. First, I see it as a valuable opportunity for personal growth. I am eager to expand my expertise in predictive modeling, data analysis, and healthcare research through practical experience in this dynamic and challenging field.
Second, I believe that the Pattern-Cog project has a positive impact on the field of neuroscience. Detecting Alzheimer’s disease in healthy individuals is crucial as it can lead to more effective interventions and treatments, potentially slowing down the disease’s progression. This project has the potential to provide valuable insights and tools that not only improve patient outcomes but also help us to better understand Alzheimer’s disease. Furthermore, it aligns with the larger goal of implementing data-driven solutions in healthcare, which is becoming increasingly important.
In summary, I’m excited about both personal and professional growth through this project. At the same time, I believe the Pattern-Cog project has the potential to make a significant impact in healthcare and contribute to our ongoing fight against Alzheimer’s disease.
