An AI research paper co-authored by Digital Planning will be presented at a leading science and technology conference in Morocco.
The conference paper, Recognizing Cardiovascular Risk Patterns using Ensemble Learning Algorithms, demonstrates how combining multiple artificial intelligence models can improve the predictive accuracy of heart attack risk.
This approach of combining multiple AI models is known as ensemble learning.
Designed as an early stage decision-support, the model aims to help healthcare professionals identify early detection of cardiovascular risk and target interventions using patient lifestyle and health data.
Digital Planning Chief Science Officer Ben Hutchings said: “The paper examines how ensemble learning can reveal relationships between patient data and cardiovascular health outcomes that traditional single-model approaches might overlook.”
The paper, which will appear in the proceedings of the Recent Trends in Image Processing & Pattern Recognition conference, will be presented by PhD researcher Aman Wakade in December.
Working in partnership with colleagues from the University of Derby and the Indian Institute of Information Technology, Aman produced the paper with Ben contributing as a technical co-author.
Alongside, Aaisha Makkar, an accomplished AI academic and researcher with 60+ published papers and experience supporting industrial collaboration projects, and supervisor to Aman Wakade’s research.
The collaboration brings together technical expertise and academic depth, combining Digital Planning’s experience in applied AI with research leadership from universities.
By integrating data science, medical informatics, and clinical research, the paper turns complex patient data into actionable insights for early cardiovascular risk detection.
Digital Planning co-founder Mark Underwood said: “ We’re incredibly proud to see Digital Planning’s research and ideas being shared at a leading science and technology conference in Morocco. It’s a real testament to the creativity and hard work of our team.
“Our partnership with leading universities plays a key role in bringing together academic insight and real-world innovation, helping us deliver practical solutions that create real impact.”
Findings show that ensemble learning algorithms, evaluated on the Kaggle dataset of 8,763 patient records, deliver 74% accuracy.
CatBoost, a gradient boosting method optimised for categorical data, and Random Forest, a tree-based ensemble model that reduces overfitting, both outperformed other classifiers. Both achieving the highest scores for accuracy, precision, recall and F1-score, a metric that balances both precision and recall.
The international conference brings leading minds together to explore the latest breakthroughs in image processing, pattern recognition, computer vision, and machine learning.
It attracts researchers, academics, and innovators internationally to share new thinking in AI-driven image analysis and its real-world applications.
As AI continues to shape the future of healthcare, this work highlights how collaboration between industry and academia can accelerate innovation and improve patient outcomes.
Digital Planning is proud to be part of this important step forward.
