The College of Education for Pure Sciences, Department of Computer Science, discussed a master’s thesis on ECG image diagnosis using vision transformers (ViT).
The message submitted by the researcher (Fatima Maalak Hanoun) included:
The vision transformer (VIT) paradigm has seen significant use in computer vision applications recently, especially in image classification tasks. The use of VIT in electrical image (ECG) classification problems is presented.
The proposed model can help in early detection and diagnosis of cardiovascular diseases
As for the purpose of the model, the main purpose of this work is to design a deep learning system that we can benefit from in predicting a suspected heart condition, whether normal or abnormal, through ECG images.
The goal of the message
The high accuracy of the proposed method demonstrates the extent to which the use of computer vision can replace the tedious and time-consuming work of image interpretation in different hospitals. This results with an impressive accuracy rate of 98.23% in detecting abnormal heartbeats. This enabled the model to better capture complex patterns and features in ECG heartbeat images, ultimately leading to more accurate and earlier diagnosis of heart irregularities in arrhythmias. These findings have great potential for the healthcare field, offering the prospect of improving diagnostic tools to identify cardiac health problems at an early stage, thus facilitating timely interventions and saving lives.