Research Blog

Revolutionizing Product Development with Generative AI: The Next Frontier

bruttygates
Dec 21, 2023
2 min read

Abstract: In the dynamic landscape of technological innovation, generative artificial intelligence (AI) emerges as a transformative force in product development and management. This document elucidates the groundbreaking research and application strategies that harness generative AI to redefine creativity, efficiency, and decision-making in product lifecycles.

Introduction: The advent of generative AI has unlocked unprecedented potential in product development. By automating and enhancing creative processes, generative AI is not just a tool but a paradigm shift, enabling a future where human ingenuity is augmented by machine intelligence.

Innovation Through Automation: Our research centers on automating the iterative design processes. Generative AI algorithms can rapidly prototype designs, test variations, and optimize features, significantly reducing the time from concept to market. This automation extends to coding, where AI systems generate and test new software features autonomously.

Enhanced Creativity and Customization: Generative AI transcends traditional boundaries of creativity. By analyzing market data and consumer trends, our AI models can suggest product designs tailored to evolving consumer preferences, delivering personalized experiences at scale.

Data-Driven Decision Making: Product management benefits immensely from generative AI's predictive capabilities. By simulating market responses and analyzing the competitive landscape, generative AI aids in strategic planning and resource allocation, ensuring that product decisions are data-driven and aligned with market dynamics.

Streamlined Operations and Efficiency: Generative AI optimizes operations by predicting and mitigating supply chain disruptions. Its ability to model and adapt to various scenarios results in resilient and efficient production workflows.

Sustainability and Ethical Considerations: Our commitment to sustainability is integral to our use of generative AI. By optimizing materials and production processes, we not only reduce waste but also advance eco-friendly product development. Ethical AI use is also a cornerstone of our research, ensuring that our advancements are beneficial and equitable.

Conclusion: The application of generative AI in product development and management marks the beginning of a new era. Our groundbreaking research and practical applications demonstrate a roadmap for harnessing AI's potential responsibly and creatively. The future of product innovation is here, and it is inherently intelligent, adaptive, and boundlessly creative.

Research World - Fall 2023

bruttygates
Sep 13, 2023
2 min read

Updated: Dec 21, 2023

I invite you to join me on an exciting journey through the realms of intelligence, technology, and their transformative impact on healthcare. With a background spanning geophysics, business intelligence, and data science, I've embarked on a new adventure as a Ph.D. researcher in the field of Intelligent Systems, with a particular focus on Imaging, Human-Computer Interaction (HCI), and Natural Language Processing (NLP).

In this blog, I aim to share my insights, discoveries, and ongoing projects as I delve into the fascinating world of artificial intelligence, machine learning, and statistical analysis in the context of healthcare and neuroimaging. Here's a sneak peek of what you can expect:

Bridging the Gap Between Data and Health: I'll be exploring innovative approaches to deciphering complex neurological data using advanced AI algorithms. Together, we'll unravel the mysteries of the human brain and its implications for healthcare.

Harmonizing Data for Progress: My research extends to data harmonization in multi-scanner, multi-center neuroimaging studies. We'll delve into the intricacies of ensuring data consistency and reproducibility across diverse sources, a critical component in advancing our understanding of neuro diseases.

Unleashing the Potential of Biomarkers: Join me as I uncover potential biomarkers for various neuro conditions. I'll walk you through the intricacies of machine learning models and statistical techniques employed in this quest for valuable insights.

Collaborations and Beyond: I believe in the power of collaboration. I'll share my experiences working with colleagues from diverse backgrounds and disciplines, aiming to bridge the gap between academia and industry.

My ultimate goal is to utilize technology and intelligence to revolutionize precision. I invite potential collaborators, academicians, and employers to join me on this exciting research expedition. Together, we can make a profound impact on the future of healthcare, pushing the boundaries of what's possible in the realm of intelligent research.

Feel free to reach out, share your thoughts, or explore potential collaborations. Let's embark on this journey together and transform the landscape of healthcare through intelligence, innovation, and research excellence. Thank you for visiting, and I look forward to our shared exploration of the limitless possibilities that lie ahead.

Title: Advanced Neuroimaging Biomarkers and Multi-Modal Data Fusion for Dynamic Disease Progression Modeling

Screenshot 2023-09-12 at 12.51.54 PM.png

The field of neuroinformatics is on the verge of a paradigm shift, thanks to the groundbreaking research project "Advanced Neuroimaging Biomarkers and Multi-Modal Data Fusion for Dynamic Disease Progression Modeling." This ambitious undertaking, driven by the power of data, machine learning, and cutting-edge neuroimaging techniques, is poised to revolutionize our understanding of neurodegenerative diseases, particularly Alzheimer's Disease (AD). In this comprehensive analysis, we delve into the core aspects of this project and explore the potential state-of-the-art discoveries it holds.

Project Overview:

At its core, this project harnesses the extensive and invaluable dataset provided by the Alzheimer's Disease Neuroimaging Initiative (ADNI) to pioneer advancements in neuroimaging biomarkers and dynamic disease progression modeling. ADNI offers a treasure trove of longitudinal data encompassing neuroimaging scans, clinical assessments, genetics, and more. This data serves as the bedrock upon which this transformative research is built.

Key Objectives:

Unearthing Novel Biomarkers: A primary mission of this project is to uncover novel biomarkers associated with neurodegenerative diseases. These biomarkers may manifest as structural brain changes, functional variations, or genetic indicators that underlie the progression of these diseases. By identifying these markers, we can gain deeper insights into the pathological processes at play.

Integration of Multi-Modal Data: This research pioneers the integration of multi-modal data sources. By harmonizing neuroimaging data with clinical evaluations, genetic profiles, and other pertinent information, it aspires to create a comprehensive and holistic view of disease progression. This synergistic approach promises to provide a more nuanced understanding of neurodegenerative disorders.

Dynamic Disease Progression Modeling: The project's dynamic disease progression models represent a significant leap forward. Conventional models often overlook the ever-evolving nature of neurodegenerative diseases. These models will be tailored to individual patients, capturing the intricate and personalized trajectories of disease progression. This allows for the development of more precise and personalized treatment strategies.

Potential State-of-the-Art Discoveries:

Early Detection Revolution: The project is poised to uncover new biomarkers that facilitate early disease detection. Early diagnosis is a game-changer in neurodegenerative diseases, offering the possibility of interventions before irreversible damage occurs.

Precision Medicine: The dynamic disease progression models will open the door to personalized treatment approaches. Tailored interventions based on an individual's disease trajectory have the potential to significantly enhance patient outcomes and quality of life.

Trailblazing Data Fusion: This project's innovative approach to integrating multi-modal data is likely to set new standards in the field. It could lead to the development of comprehensive diagnostic tools that offer a more holistic view of neurodegenerative diseases, transforming diagnosis and treatment.

Scientific Advancement: The methodologies and findings of this project will be invaluable contributions to the scientific community. They will serve as foundational knowledge for further research and collaboration, propelling the field of neuroinformatics into uncharted territory.

Interdisciplinary Potential: The project bridges diverse fields, from neuroscience to data science and beyond. Collaborators from various backgrounds can bring their unique expertise to the table, fostering innovation through diverse perspectives.

Therefore, "Advanced Neuroimaging Biomarkers and Multi-Modal Data Fusion for Dynamic Disease Progression Modeling" represents a watershed moment in neuroinformatics research. With the potential to make state-of-the-art discoveries, this project invites collaborations from researchers, academicians, and institutions eager to push the boundaries of knowledge. Together, we can unlock new insights, revolutionize diagnostics and treatments, and profoundly impact the lives of individuals affected by neurodegenerative diseases. Join us in this collective endeavor to redefine the future of healthcare and neuroscience.

Biomarker Discovery from Rat Gene Expression for Intervertebral Disc Degeneration -
Bamidele Ajisogun

Screenshot 2023-09-14 at 12.17.20 AM.png

Screenshot 2023-09-14 at 12.17.33 AM.png

ABSTRACT

Intervertebral disc degeneration (IDD) is a common musculoskeletal disorder that can cause back, or neck discomfort and chronic pain associated with aging. The degeneration of the nucleus pulposus (NP) cells, the central component of the intervertebral disc, leads to dehydration, loss of disc height, disc distortion, and segmental instability. Identifying biomarkers for IDD can aid in diagnosis, monitoring, and developing precise treatments for the condition. Gene expression data from young and old, male, and female rat intervertebral disc (IDD) tissue types, along with known extracellular matrix-related genes from related human tissues were analyzed to discover potential biomarkers. Machine learning techniques of Logistic Regression, Support Vector Machines, Random Forest, Naïve Bayes, and Rule Learner were utilized to analyze the genes and discriminate between the nucleus pulposus (NP) and annulus fibrosus (AF) tissue types. This study presents the feasibility of a knowledge Augmented Rule Learner (KARL) to provide accurate and interpretable models, which can be useful as an efficient integrative biomarker discovery tool for diagnosing and treating IDD in precision medicine. The dataset contains 16,378 genes and 38 samples distributed across tissue types, gender, and age. Further research is necessary to validate the identified biomarkers and understand their role in the disease process.

We analyzed the results of the machine learning algorithms from which logistic regression performed the best and compare them with the framework of the Knowledge Augmented Rule Learning (KARL), which incorporates two sources of knowledge, domain, and data, for pattern discovery from small and high-dimensional datasets. We examined the effectiveness of KARL as a transfer rule learning framework in which knowledge of the domain is transferred to the learning process on data to 1) improve the reliability of the discovered patterns, and 2) study the knowledge of the domain when compared with the results of the machine learning algorithms for modeling. In this work, we generated KARL models on gene expression datasets for six data tissue types of the rat data. As our knowledge of the domain, we used the Ingenuity Knowledge Base (IKB) to extract genes related to hallmarks of IDD from human extracellular matrix data and annotated these prior relationships before learning classifiers from these datasets.

Our results revealed that the Machine learning model of Logistic Regression was effective in identifying the biomarkers responsible for the IDD disease. However, KARL produces, on average, rule models that are more robust classifiers than the baseline without such background knowledge, for our tasks of IDD prediction using the gene expression datasets. Moreover, KARL served as an integral approach and helped us learn insights about previously known relationships in these gene expression datasets, along with new relationships not input as known, to enable informed biomarker discovery for IDD prediction tasks. KARL can be applied to modeling similar data from any other domain and classification task. Future work would involve extensions to KARL to handle ranked knowledge to derive more general hypotheses to drive biomedicine.

Building an Advanced Recommendation System using Neural Collaborative Filtering (NCF) and Variational Autoencoders (VAEs)

ABSTRACT

Recommender systems are widely used to offer personalized recommendations for goods or services. Machine learning algorithms have been increasingly adopted by these systems in recent years, and selecting the appropriate algorithm is crucial. However, little guidance is available regarding the current usage of algorithms in recommender systems, and there are many challenges to creating effective recommender systems using machine learning algorithms. In this project, we build an advanced recommendation system using Neural Collaborative Filtering (NCF) and Variational Autoencoders (VAEs). We leverage a large-scale rating dataset containing user ID, movie ID, rating, and timestamp attributes to train and test our recommendation system. Our investigation focuses on combining traditional collaborative filtering and matrix factorization techniques with advanced deep learning-based techniques, such as NCF and VAEs, to build our recommendation system. We evaluate the performance of each model using multiple evaluation metrics, including precision, recall, and F1-score and compare their performance to decide on the best model based on recommendation quality, evaluation metrics, and time efficiency.

FUTURE WORK

Our study presents various opportunities for future work in movie recommendation systems. First, we could explore the combination of collaborative filtering and content-based recommendation systems to create a hybrid model that leverages the strengths of both approaches. This could potentially improve the accuracy of recommendations by incorporating additional features such as movie genre, director, and actors, which can be extracted from the movie metadata.

Incorporating contextual information: In this project, we only used user-item interaction data for the recommendation. However, incorporating contextual information such as time, location, and user demographics could improve the quality of recommendations.

Exploration of Deep Learning models: Although we compared the performance of two deep learning models in this project, there are many other deep learning models that could be explored, such as Convolutional Neural Networks (CNNs) and Transformer-based models.

Explainability and transparency: Recommendation systems can sometimes be seen as a "black box" where users do not know how the system arrives at its recommendations. Future work could focus on developing more transparent and explainable models to increase user trust and confidence in the recommendations.

Active Learning: The project evaluated different recommendation algorithms based on the accuracy of their predictions. However, accuracy alone may not always be sufficient for some applications. Future work could investigate active learning techniques to optimize the trade-off between recommendation accuracy and diversity.

Privacy and Security: With increasing concerns about data privacy and security, future work could focus on developing more privacy-preserving and secure recommendation systems, such as Federated Learning, Differential Privacy, and Secure Multi-Party Computation.

Multi-Objective Optimization: In this project, we mainly focused on optimizing for the accuracy of recommendations. However, there are often multiple objectives that need to be considered, such as diversity, novelty, and serendipity. Future work could explore multi-objective optimization techniques to balance these competing objectives with the system.

Research Blog

Title: Advanced Neuroimaging Biomarkers and Multi-Modal Data Fusion for Dynamic Disease Progression Modeling

Biomarker Discovery from Rat Gene Expression for Intervertebral Disc Degeneration - Bamidele Ajisogun

Building an Advanced Recommendation System using Neural Collaborative Filtering (NCF) and Variational Autoencoders (VAEs)

Contact Information

Biomarker Discovery from Rat Gene Expression for Intervertebral Disc Degeneration -
Bamidele Ajisogun