Dr Balbir Singh

Childhood obesity is a growing public health concern worldwide, with significant implications for long-term health outcomes and healthcare costs. Early identification of children at risk of obesity is essential for implementing effective prevention and intervention strategies. In recent years, machine learning techniques have emerged as powerful tools for predicting childhood obesity risk based on various predictors such as demographic factors, dietary habits, physical activity levels, and genetic predisposition. This paper explores the use of temporal deep-learning models, specifically recurrent neural networks and long short-term memory, to predict obesity childhood stages using longitudinal datasets. The long short-term memory model emerges as the top performer, achieving an accuracy of 85% and an F1-score of 0.85, surpassing existing models. To enhance model transparency, model-agnostic methods are employed, revealing critical insights into model predictions. Results indicate that BMI at age 11 is the most influential feature in predicting obesity. These findings underscore the importance of applicability models and explainable machine learning methods in understanding and predicting childhood obesity progression.

Obesity is a major global concern with more than 2.1 billion people overweight or obese worldwide which amounts to almost 30% of the global population. If the current trend continues, the overweight and obese population is likely to increase to 41% by 2030. Individuals developing signs of weight gain or obesity are also at the risk of developing serious illnesses such as type 2 diabetes, respiratory problems, heart disease, stroke, and even death. Some intervention measures such as physical activity and healthy eating can be a fundamental component to maintain a healthy lifestyle and help in tackling obesity. Therefore, it is essential to detect childhood obesity as early as possible since children who are either overweight or obese around their adiposity rebound age tend to stay obese in their adult lives. This research utilises the vast amount of data available via UK’s millennium cohort study to construct machine learning driven frameworks to predict young people at the risk of becoming overweight or obese. Although there are several research examples globally in predicting childhood obesity, the use of Millennium Cohort Study (MCS) data remains underutilised. Furthermore, attempts have only been made to predict obesity with a small timelapse between the observed data and target prediction age. This research focuses on predicting obesity at two major milestones in children’s growth using data collected from birth, and subsequent surveys conducted at ages 3, 5, 7 and 11. The first milestone is the adiposity rebound age which occurs around the age of 6 or 7. This milestone is important since children tend to stay on the obesity centile in their later growth years whichever obesity centile they are on during their adiposity rebound age period. The second important milestone is when children enter their teen years and hormones start to change. Having a healthier weight when entering teen years will help them not having to worry about problems caused by overweight and obesity. The MCS survey took place 6 times and data from six survey waves were combined to form one dataset containing longitudinal and cross-sectional features relating to children’s growth. There is an inherent imbalance in the dataset of individuals with normal BMI and the ones at risk. Data balancing was carried out using the random under-sampling algorithm to under sample the majority class to create subsets of data to match the at risk class size. The results obtained from each subset were averaged to arrive at a final classification accuracy. This approach makes use of the complete dataset and does not involve generation of any synthetic data to over sample the minority class. Several frameworks are proposed including classification from regression to predict BMI and then determining obesity flags using age and sex and also to identify key factors that influence obesity. These proposed frameworks allow to maximise all classification metrics and predict the adiposity rebound age obesity with an accuracy of 83% using the age 5 BMI values. One of the frameworks was used to predict the adiposity rebound age obesity when using the data from age 3 survey. The use of data balancing and additional relevant features helped in improving prediction accuracy from 64.55% to over 73% and the F1 score from 45.93% to over 71%. Combining the other relevant features with the BMI data allows to predict the age 14 obesity status as early as age 3 with an accuracy of over 70%. There are hundreds of additional features but only the easily obtainable ones were considered so that even parents or caregivers can make use of them in predicting the obesity status of a child. The focus has been not only to maximise the average accuracies but to also to enhance the specificity and precision values to minimise the prediction of false positives. The suitability of each framework for clinical assessment and population monitoring is clearly identified.

Artificial Intelligence is revolutionizing the healthcare business, thanks to the rising availability of structured and unstructured data and the rapid advancement of analytical methodologies. With artificial intelligence's expanding importance in healthcare, there are growing issues about a lack of transparency and explainability, as well as potential bias in model predictions. The goal of this paper is to use interpretable ML to provide a better understanding of the lifestyle factors that influence the model's predictions of the weight levels, as well as to identify the most critical features for the classification task. We aim to create more accurate and effective predictive models for weight management, which could eventually help individuals make more informed decisions about their diet and lifestyle choices. Our machine learning model got an accuracy of 76% using XG-boost and Random Forest and we tried to interpret the result using a model-agnostic method including, the permutation importance method and partial dependence plot. The results showed that the number of main meals, frequent consumption of vegetables, and time using technology are the most important features. Further explanation was performed using a partial dependence plot which interprets the relationship between these important features and the model behavior.

In recent years, the prevalence of obesity and its related co-morbidities have been increasing significantly. Therefore, it is an important challenge to pursue an early prediction of obesity risk that could help in reducing the pace of obesity rise when appropriate interventions are placed, accordingly. The prediction and classification of obesity depend on different factors such as body mass index (BMI) and lifestyle aspects, including eating habits. By focusing on these lifestyles and eating habit factors, we can develop a more holistic approach to weight management and prevention of obesity. The aim of this paper is to propose a machine-learning model that can classify weight levels using lifestyle variables without relying on BMI which enables us to investigate how lifestyle factors affect different levels of weight categorization. Although BMI is the most widely used estimation of obesity, there are other factors that can contribute to gaining weight such as lifestyle factors. The accuracy of our lifestyle-based model reached 75% excluding weight, height, and family history. Our model could serve as a starting point for using an interpretable machine learning model to better understand the effect of lifestyle factors on obesity levels.

Online communities have been proactive in producing collaborative creative content such as music, games and other social interactions. Online collaboration has enabled contributors to peer produce and share masses of creative content. Examples range from information sharing such as Wikipedia to open source software and other specific art projects. Software vendors have recently introduced low cost 2D and 3D content authoring tools allowing user communities to generate and share creative content. Emerging networking programming interfaces available inside modern game engines allow contributors to implement multiplayer or multiuser interaction relatively easily. This paper presents a 3D art creation framework to be used over networked infrastructure in a multiuser environment. Contributors will be able to create 3D sculptures at runtime, share with other users in a common networked working environment and critique each other’s work.

Obesity is a critical health issue associated with severe medical conditions. To enhance public health and well-being, early prediction of obesity risk is crucial. This study introduces an innovative approach to predicting obesity levels using explainable artificial intelligence, focusing on lifestyle factors rather than traditional BMI measures. Our best-performing machine learning model, free from BMI parameters, achieved 86.5% accuracy using the Random Forest algorithm. Explainability techniques, including SHAP, PDP and feature importance are employed to gain insights into lifestyle factors’ impact on obesity. Key findings indicate the importance of meal frequency and technology usage. This work demonstrates the significance of lifestyle factors in obesity risk and the power of model-agnostic methods to uncover these relationships.

Online communities have been proactive in producing collaborative creative content such as music, games and other social interactions. Online collaboration has enabled contributors to peer produce and share masses of creative content. Examples range from information sharing such as Wikipedia to open source software and other specific art projects. Software vendors have recently introduced low cost 2D and 3D content authoring tools allowing user communities to generate and share creative content. Emerging networking programming interfaces available inside modern game engines allow contributors to implement multiplayer or multiuser interaction relatively easily. This paper presents a 3D art creation framework to be used over networked infrastructure in a multiuser environment. Contributors will be able to create 3D sculptures at runtime, share with other users in a common networked working environment and critique each other’s work. Experimental work also involved evaluating procedurally generated meshes versus instantiation of primitive mesh objects. Saving and loading mesh information in an optimum way is also explored.

Obesity is a major global concern with more than 2.1 billion people overweight or obese worldwide which amounts to almost 30% of the global population. If the current trend continues, the overweight and obese population is likely to increase to 41% by 2030. Individuals developing signs of weight gain or obesity are also at a risk of developing serious illnesses such as type 2 diabetes, respiratory problems, heart disease and stroke. Some intervention measures such as physical activity and healthy eating can be a fundamental component to maintain a healthy lifestyle. Therefore, it is absolutely essential to detect childhood obesity as early as possible. This paper utilises the vast amount of data available via UK’s millennium cohort study in order to construct a machine learning driven model to predict young people at the risk of becoming overweight or obese. The childhood BMI values from the ages 3, 5, 7 and 11 are used to predict adolescents of age 14 at the risk of becoming overweight or obese. There is an inherent imbalance in the dataset of individuals with normal BMI and the ones at risk. The results obtained are encouraging and a prediction accuracy of over 90% for the target class has been achieved. Various issues relating to data preprocessing and prediction accuracy are addressed and discussed.

This report outlines the process and findings from an innovative project for students. This work was part of the curriculum and involved students working with West Yorkshire Police as part of the safer Leeds project in designing and making a film for students n crime prevention and personal safety in Leeds

Individuals developing signs of weight gain or obesity are at a risk of developing serious illnesses such as type 2 diabetes, respiratory problems, coronary heart disease and stroke. Physical activity and healthy eating can be a fundamental component to maintain a healthy lifestyle. Therefore, detecting childhood obesity is of paramount importance. This paper utilises the vast amount of data available via the millennium cohort study. Various regression methods and artificial neural network models have been evaluated to predict the teenager BMI from earlier BMI values. The results obtained are encouraging and a prediction accuracy of over 90% has been achieved. Various issues relating to data mining and prediction accuracy are discussed.

Obesity is a major global concern with more than 2.1 billion people overweight or obese worldwide, which amounts to almost 30% of the global population. If the current trend continues, the overweight and obese population is likely to increase to 41% by 2030. Individuals developing signs of weight gain or obesity are also at the risk of developing serious illnesses such as type 2 diabetes, respiratory problems, heart disease, stroke, and even death. It is essential to detect childhood obesity as early as possible since children who are either overweight or obese in their younger age tend to stay obese in their adult lives. This research utilises the vast amount of data available via UK's millennium cohort study to construct machine learning driven framework to predict young people at the risk of becoming overweight or obese. The focus of this paper is to develop a framework to predict childhood obesity using earlier childhood data and other relevant features. The use of novel data balancing technique and inclusion of additional relevant features resulted in sensitivity, specificity, and F1-score of 77.32%, 76.81%, and 77.02% respectively. The proposed technique utilises easily obtainable features making it suitable to be used in a clinical and non-clinical environment.