Skripsi
DETEKSI ST ELEVATION MYOCARDIAL INFARCTION PADA SINYAL ELEKTROKARDIOGRAM SINGLE LEAD MENGGUNAKAN KECERDASAN BUATAN
Artificial Intelligence (AI) refers to efforts to simulate human intelligence in machines such as computers or robots. AI is a more general concept than Machine Learning, and Deep Learning. Deep learning enables computational models consisting of multiple processing layers to learn data representations with various levels of abstraction in stages. The electrocardiogram signal consists of P, Q, R, S, T and U waves. To identify the waves in the EGK signal, a delineation process was carried out where in this study the data was delineated into eight classes, namely Pwave, PR Segment, Qon-Rpeak, Rpeak-Qoff, ST, Twave, Toff-Pon and Zeropad segments. The use of deep learning methods in the delineation process aims to reduce misinterpretation. In this study, a computer-based delineation system will use deep learning methods. The deep learning method used is LSTM and a combination of CNN-BiLSTM. Signal delineation is carried out on eight wave classes with a total of 24 models designed for each which will be trained and tested with LUDB data. Each model is designed with the best combination of hidden layer, batch size, learning rate, and epoch parameters. The best model obtained is the 4th CNN-BiLSTM model with 13 hidden layers of CNN, and 1 layer of BiLSTM. This model produces the best evaluation results with a value of 0.0001, epochs 400, batch size 8, with an average sensitifity value of 95.57%, 95.54% precision, 99.68% specificity, 99.43% accuracy, 0.57% error, and 95.55% f1-score. Then, from the results of the delineation process for 8 classes, it will be followed by an ST Elevation detection process that focuses on the ST Segment and the PR Segment.
Inventory Code | Barcode | Call Number | Location | Status |
---|---|---|---|---|
2307004559 | T122266 | T1222662023 | Central Library (REFERENS) | Available but not for loan - Not for Loan |
No other version available