Stuart E Lacy

• PhD, MEng
• Developer at University of York

Parkinson's disease is a debilitating neurological condition that affects approximately 1 in 500 people and often leads to severe disability. To improve clinical care, better assessment tools are needed that increase the accuracy of differential diagnosis and disease monitoring. In this paper, we report how we have used evolutionary algorithms to i...

Ensemble classifiers have become popular in recent years owing to their ability to produce robust predictive models that generalise well to previously unseen data. In principle, Evolutionary Algorithms (EAs) are well suited to ensemble generation since they result in a pool of trained classifiers. However, in practice they are infrequently used for...

Follicular lymphoma is morphologically and clinically diverse; with mutations in epigenetic regulators alongside t(14;18) identified as disease initiating events. Identification of additional mutational entities confirm this cancer's heterogeneity, but whether mutational data can be resolved into mechanistically distinct subsets remains an open que...

Based on the profile of genetic alterations occurring in tumor samples from selected diffuse-large-B-cell-lymphoma (DLBCL) patients, two recent whole exome sequencing studies proposed partially overlapping classification systems. Using clustering techniques applied to targeted sequencing data derived from a large unselected population-based patient...

Follicular lymphoma is an incurable haematological cancer that tends to follow a remitting relapsing course; with treatment options ranging from “watch and wait” (active monitoring), chemotherapy, and radiotherapy. The treatment decision is typically dependent upon patient characteristics and disease stage. Using routine data collected by the Haema...

Despite having notable advantages over established machine learning methods for time series analysis, reservoir computing methods, such as echo state networks (ESNs), have yet to be widely used for practical data mining applications. In this paper, we address this deficit with a case study that demonstrates how ESNs can be trained to predict diseas...

This work presented an interactive web application for building multi-state models of disease pathways. The app is flexible, allowing for both parametric and semi-parametric models, with transition-specific distributions. The presentation won the award for Best Presentation.

A survey of possible ways to evaluate survival models that are intended for prognostic, rather than inferential aims. The work was demonstrated on a clinically motivated data set of Follicular Lymphoma. This presentation won the Best in Session Award.

Objective To investigate separable components of finger tapping (FT) of the thumb and index finger in Parkinson's (PD) with normal (PD-NC) and impaired cognition (PD-CI) and in healthy controls (HC). Methods 58 PD and 29 HC performed FT for 30 seconds whilst attached to electromagnetic movement sensors sampling at 60 Hz. All subjects completed the...

Ensemble classifiers have become a widely researched area in machine learning because they are able to generalise well to unseen data, making them suitable for real world applications. Many approaches implement simple voting techniques, such as majority voting or averaging, to form the overall output. Alternatively, Genetic Algorithms (GAs) can be...

Ensembles are groups of classifiers which cooperate in order to reach a decision. Conventionally, the members of an ensemble are trained sequentially, and typically independently, and are not brought together until the final stages of ensemble generation. In this paper, we discuss the potential benefits of training classifiers together, so that the...

The treatment of Parkinson’s Disease can prove extremely costly and as of yet there are no available cures. Rapid and accurate diagnoses are essential for managing the condition as best as possible, however even among trained medical professionals there still remains a relatively high misdiagnosis rate of 25%. Ensemble classification is a popular f...

Objective: To assess whether an accurate objective measurement of ‘slight’ (MDS-UPDRS grade one) bradykinesia can be obtained from a novel device employing electromagnetic (EM) tracking sensors and evolutionary algorithm analysis. Background: Bradykinesia is the fundamental clinical feature of Parkinson's disease (PD) but when very mild is diffic...

Parkinson’s Disease is a devastating illness with no currently available cure. As the population ages, the disease becomes more common with a large financial cost to society. A rapid and accurate diagnosis, as well as practical monitoring meth- ods are essential for managing the disease as best as possible. This paper discusses two approaches...

We describe the use of a genetic programming system to induce classifiers that can discriminate between Parkinson's disease patients and healthy age-matched controls. The best evolved classifer achieved an AUC of 0.92, which is comparable with clinical diagnosis rates. Compared to previous studies of this nature, we used a relatively large sample o...