Javascript must be enabled for the correct page display

Internship: Development of a set of concept rules for objective ECG assessment

Chikovskii, N. (2014) Internship: Development of a set of concept rules for objective ECG assessment. Master's Thesis / Essay, Biology.

MasterLS_BME_2015_NChikovskii.pdf - Published Version

Download (1MB) | Preview
[img] Text
toestemming.pdf - Other
Restricted to Backend only

Download (856kB)


Current techniques used by cardiologists to assess ECGs are highly subjective. Even doctors who graduated from the same department and university have different approaches when assessing ECG of the patients. The problem is even worse for the patients with different heart diseases and abnormalities. All this leads to high variation of diagnosis and consequent high number of false prescription of certain therapies. The goal of the project was to objectify the procedure of ECG assessment for patients with problems in heart conduction system. To address the issue a team of the department of cardiology of UZ Gent was formed by Prof. Ronald Stroobandt, MD, PhD, Milad El Haddad,MSc, and Jan De Pooter, MD. Their goal was to introduce an objective method of ECG assessment with its further development into an algorithm for Implantable Cardiac Defibrillators. They initially acquired a set of 12-lead ECG from 52 patients with different problems of heart conduction system. I performed evaluation of all 52 ECGs in order to identify patterns and specific objective features, which would allow for development of rules for objective assessment of ECGs. After first set of rules was developed it was tested against results of ECG assessment by 5 different cardiologists. The mean and standard deviation of the results of the cardiologists was taken as a reference for evaluation of performance of the rules. Accuracy of applying the rules was evaluated and determined for each of 12 ECG used for rules testing. Approximately 75% of ECGs were inaccurately assessed, i.e. the values acquired by applying the rules were outside the confidence interval (i.e. reference standard deviation). Next, after several consultations with the team and external cardiologists rules were adjusted and tested again. That time, the rules performed significantly better – all ECGs were assessed with at least 75% accuracy relative to the reference (0% accuracy is when the rules' value deviates by the value of standard deviation from the mean taken as a reference value). At the end of the project a presentation was made showing the process and the results of the project: a final set of 12 rules was developed and its performance assessed by cardiologists. Outcomes of this project were: • Development of objective rules for ECG assessement with further development of a possible gold standard for ECG assessemtn • Development of an algorithm for ICD devices for patients with complex ECGs

Item Type: Thesis (Master's Thesis / Essay)
Degree programme: Biology
Thesis type: Master's Thesis / Essay
Language: English
Date Deposited: 15 Feb 2018 08:08
Last Modified: 15 Feb 2018 08:08

Actions (login required)

View Item View Item