Text
PRAKIRAAN NILAI DENYUT JANTUNG PADA SERIES DATA PASIEN UNIT PERAWATAN INTENSIF MENGGUNAKAN TIME LONG SHORT TERM MEMORY
Vital sign monitoring is the most common assessment in medicine. Vital signs show how the patient's status is. Proper and accurate monitoring of vital signs is very important to ensure the progress and deterioration of the patient's health condition in the hospital. The use of vital signs as an indication of an event has long been applied in th e medical industry. However, much research is still being done to improve clinical performance. The data in this study used the ICU patient's vital sign dataset in the form of an irregular time series, where the ICU patient's vital sign dataset was obtaine d from the MIMICIII (Medical Information Mart for Intensive Care) database. The data will be processed by making the data in the form of a window with one value from a predetermined forecast range. This study uses three methods of Deep Learning, namely Lo Short Term Memory (LSTM), Bidirectional LongngShort Term Memory (BiLSTM) and Gated Reccurent Unit (GRU). Of the three methods, each of each method produces 12 models so that the total model obtained is 36 models. From the 36 models, the GRU method produc es the smallest RMSE and MAE values, which are 0.01630 and 0.01130, respectively, which are obtained in batch size 64 with epochs of 100.
Inventory Code | Barcode | Call Number | Location | Status |
---|---|---|---|---|
2207003885 | T79616 | T796162022 | Central Library (Referens) | Available but not for loan - Not for Loan |
No other version available