Cardiac Measurement and Analysis
CSEF · 2005 Mathematics & Software
Overview
Objectives/Goals Purpose of the Experiment: The goal of this project was to develop an expert system to screen for cardiac anomalies. The specific objectives were to 1) construct an electrocardiograph 2) create an expert system which will differentiate among different heart anomalies and 3) record and accurately analyze cardiac waveforms. Methods/Materials Methods: A one channel ECG was constructed to measure cardiac electric potential. Data was digitized using an analog to digital converter. An expert system was created using Java. It was designed to identify heart potential waveforms by using a set of trained neural networks. The training was based on digitized heart anomaly waveforms available over the WEB from MIT/BIH. In addition variants of these neural networks were designed to find distinct segments (P, QRS, PR complexes) of the cardiac waveform, allowing analysis of waveform features (peaks, duration and sequence). Actual ECGs were taken and analyzed for a number of subjects. Results Results/Conclusions: Initial attempts were made to train the system to recognize entire beats of various anomalies. Because of significant waveform variation even within a particular anomaly, this approach was abandoned in favor of training the networks to recognize various waveform complexes within each beat and then using their shape, duration and relative timing to diagnose anomalies. On this basis the system was able to distinguish a limited set of heart anomalies discernable on a single modified lead 2 channel with a high rate of success. These anomalies included tachycardias, bradycardias, escape and premature beats, and 1 deg AV and bundle blocks.
Summary statement
Built an electrocardiograph and developed expert system software to diagnose heart anomalies.
Help received
Ms Helms (teacher) advised on software issues, mother helped with board, father helped edit text
Awards (1)
- Category Award
Competition history
- CSEF 2005
Resources
Related projects
CSEF · 2017
CordisX: Personalized Cardiac Monitor
ISEF · 2016
Investigations into Cardiology: Categorizing Heart Rhythms using Machine Learning
CSEF · 2018
iCordisX: SmartPhone-Based Personalized Cardiac Monitoring Using Computer Vision and Bluetooth Low Energy
AJAS · 2018
CordisX: Personalized Cardiac Monitoring Through Low-Cost Sensor Systems and Real-Time Statistical Analysis Algorithms
CSEF · 2016
Project POE: Tri-Sensor Cardiac Monitoring System
CSEF · 2019
Accurate Identification of Cardiac Anomalies through Deep Learning
CSEF · 2017
Combatting Cardiac Arrest through a Compact and Portable Heart Rate Monitor and Analyzer
CSEF · 2017
CardioWatch
Closest projects by meaning, across every fair and year in the corpus.
Browse more like this
Source: California Science & Engineering Fair public projects