Mobile technology based solutions for improving the interpretation of the 12-lead electrocardiogram and clinical decision making in cardiology

Post Details: Applications are invited for an SEUPB funded PhD studentship, tenable in Ulster University (UU), in collaboration with Altnagelvin Area Hospital (AH) at L/Derry, Letterkenny University Hospital (LUH) in Donegal and Raigmore Hospital, Inverness. This studentship is part of our €8.6 million cross-border research project, led by Ulster University’s Northern Ireland Centre for Stratified Medicine funded by the EU’s INTERREG VA programme, managed by the Special EU Programmes Body.
Background: The 12-lead electrocardiogram (ECG) remains one of the most widely used diagnostic tools in medicine and is used to rapidly detect life-threatening pathologies such as acute myocardial infarction (otherwise known as a heart attack). It is widely reported that both the human and machine interpretation of the ECG are sub-optimal and require significant improvement [1, 2]. As such, there are technology based research opportunities to improve the diagnostic accuracy of ECG interpretation. Hence, this work will involve a series of studies to design, build and trial smart phone-based solutions to improve the diagnostic accuracy of ECG interpretation in clinical practice.
Study 1 will involve the development and trial of a context aware mLearning (or mobile learning) [3] app to facilitate continuous professional development for those who interpret the ECG. The app will profile each clinician and provide a series of ECGs to interpret over a period of time to monitor their competency whilst incrementally increasing their diagnostic accuracy using targeted data-informed learning. Specifically, the app will calculate each individual’s diagnostic accuracy over a range of disease groups and provide tailored material to improve the individual’s proficiency where they show poor performance. The aim is to increase the user’s performance to an expert level of proficiency, which will be quantitatively pre-determined by assessing a cohort of expert ECG readers. The app will also feature motivational and user engagement strategies. For example, the app will learn when and how each user interacts with the app (referred to as context-awareness [4]) and use intelligent notifications and prompts to encourage continuous use of the app. This is important as it is widely known that competency in ECG interpretation is correlated with the frequency of ECG interpretation, however the number of annual ECG interpretations required to sustain proficiency is unknown. The app will present ECGs using a simulation based training approach and will require the user to provide their clinical decision or interpretation using standardized vocabulary (e.g. ICD-10, SNOMED, terminology informed by ESC/AHA). Moreover, other features that are currently under investigation in the research literature will be implemented where necessary (e.g. gamification). Like all smart phone solutions, this app will then be tested using a usability test where instruments (such as eye trackers) from Ulster’s User Experience Laboratory (www.ux-lab.org) will be used to identify the cognitive load required to use the tool and also to determine and address any usability issues that exist. Subsequently, we will conduct a trial of the app that will involve pre- and post-intervention measures to determine the impact of the app and the extent of which it can improve clinical decision-making during ECG interpretation.
Study 2 may involve the development and trial of an interactive mobile app that facilitates ‘collective intelligence’ (also known as wisdom of crowds [5]). The app will allow a clinician who is uncertain of their ECG interpretation to take a photograph of the ECG and upload it to the cloud or a private secure intranet for interpretation by other colleagues and profiled users. Since the system profiles the competency of users regarding the interpretation of ECGs from various disease groups, the system can then return a weighted visualization of the interpretations, which can then aid clinical decision-making by providing collective intelligence from peers. The app will then be tested using a usability evaluation and then trialed using a two-arm study design where a control group will interpret ECGs without collective intelligence and the experimental group will interpret ECGs with the support of the collective intelligence tool. The diagnostic accuracy of the two contrasting groups will be the outcome metric.
Study 3 will involve the development of an interactive smart phone app that can be used as a decision support tool during ECG interpretation. The app may feature an interactive form to guide the user through ECG interpretation and to provide decision support and possible diagnoses to consider. In addition, the project will explore other emerging technologies that could aid the clinician in ECG interpretation. This may include augmented reality, [6] which is a technology that can use the camera on a smart phone to capture the ECG whilst machine vision is used to overlay knowledge (in this case intervals and segments or other information about the ECG signal to aid clinical decision making/interpretation).

Project Objectives:
The primary objectives of this project will be to:
1. Develop and trial a mobile training application for clinical staff to develop and sustain their clinical decision making skills
2. Develop and trial a clinical decision making assistant application using collective intelligence
3. Development of an interactive smart phone app that can be used as a decision support tool during ECG interpretation
Deliverables:
1. An engaging adaptive mLearning mobile app for sustaining ECG interpretation skills and to reduce human error.

2. A number of mobile apps that have the potential to improve decision making that involve 12-lead ECG interpretation.
Patient Data & Samples: This project will use data collected from cardiologists and nurses as well as anonymise ECGs recorded from patients.
Data Analysis and Statistical Methods:
Data analysis will include descriptive and inferential statistics as well as predictive modelling and data mining.

Entrance Requirements:
All applicants should hold a first or upper second class honours degree in computing science or a cognate area, and be able to demonstrate strong cross-disciplinary interest. Applications will be considered on a competitive basis with regard to the candidate’s qualifications, skills experience and interests. Successful candidates will enrol as of October 2017, on a full-time programme of research studies leading to the award of the degree of Doctor of Philosophy. The studentship will comprise fees together with an annual stipend of £14,553 and will be awarded for a period of up to 3 years subject to satisfactory progress.

Additional Information & Application Process: If you wish to discuss your proposal or receive advice on this project please contact:- Dr. Aaron Peace ([Email Address Removed]). Other contacts:
Dr. Raymond Bond ([Email Address Removed])
Dr Victoria McGilligan ([Email Address Removed])

For more information on applying go to ulster.ac.uk/research and apply online ulster.ac.uk/applyonline.

The closing date for receipt of online applications is: 14th August 2017