Deteksi Tekanan Darah Non-Invasif Tanpa Manset Secara Kontinu

Wanda, Aushaf Cintya (2024) Deteksi Tekanan Darah Non-Invasif Tanpa Manset Secara Kontinu. Diploma thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 5023201019-Undergraduate_Thesis.pdf] Text
5023201019-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2026.

Download (10MB) | Request a copy

Abstract

Tekanan darah (BP) memiliki hubungan erat dengan hipertensi, yang menjadi faktor penyebab utama berbagai penyakit kardiovaskular (CVD) setelah diabetes. Hipertensi, yang sering kali tidak disadari oleh penderita, menjadi penyebab signifikan penyakit jantung seperti serangan jantung dan stroke sehingga pengukuran BP secara real time dan kontinu sangat diperlukan untuk mendiagnosis hipertensi dengan akurat dan membantu penyesuaian pengobatan. Saat ini, metode pengukuran tekanan darah kontinu yang tersedia hanya dilakukan secara invasif yang menyakitkan, sedangkan metode non-invasif dengan manset belum memungkinkan pengukuran kontinu yang nyaman. Untuk mengatasi masalah ini, penelitian ini berfokus pada prediksi tekanan darah non-invasif tanpa manset menggunakan pengukuran Pulse Transit Time (PTT), yang mengukur delay waktu antara gelombang R-puncak pada elektrokardiogram (ECG) dan sinyal puncak photoplethysmogram (PPG). Dengan memanfaatkan machine learning, khususnya Artificial Neural Network (ANN), penelitian ini bertujuan untuk mendapatkan estimasi tekanan darah sistolik (SBP) dan diastolik (DBP) yang lebih akurat. Metode ini memungkinkan pengukuran tekanan darah tanpa tekanan eksternal pada tubuh, sehingga mengurangi risiko cedera atau infeksi. Model ANN yang dikembangkan memiliki nilai error MAE ± STD untuk SBP sebesar 13.89±13.52 mmHg dan untuk DBP sebesar 5.5±6.47 mmHg. Dilakukan pengujian model ANN yang dikembangkan kepada sejumlah subjek yang menghasilkan prediksi tekanan darah sistolik (SBP) dan diastolik (DBP). Sebelum itu, dilakukan pengukuran tekanan darah dengan tensimeter konvensional sebelum dan sesudah pengujian dengan metode yang dikembangkan. Didapatkan error rata-rata pengujian dibandingkan dengan tensimeter konvensional sebesar 11.2 mmHg untuk SBP dan 9.75 mmHg untuk DBP.
========================================================================================================================
Blood pressure (BP) has a close relationship with hypertension, which is the leading causative factor of many cardiovascular diseases (CVD) after diabetes. Hypertension, which is often unrecognized by sufferers, is a significant cause of heart diseases such as heart attacks and strokes, so real time and continuous BP measurement is necessary to accurately diagnose hypertension and assist with treatment adjustments. Currently, the available continuous blood pressure measurement methods are only performed in a painful invasive manner, while the non-invasive method with a cuff does not yet allow convenient continuous measurement. To address this issue, this study focuses on non-invasive cuffless blood pressure prediction using Pulse Transit Time (PTT) measurement, which measures the time delay between the R-peak wave on the electrocardiogram (ECG) and the photoplethysmogram (PPG) peak signal. By utilizing machine learning, specifically Artificial Neural Network (ANN), this study aims to obtain more accurate estimation of systolic (SBP) and diastolic (DBP) blood pressure. This method allows blood pressure measurement without external pressure on the body, thus reducing the risk of injury or infection. The developed ANN model has an error value of MAE ± STD for SBP of 13.89±13.52 mmHg and for DBP of 5.5±6.47 mmHg. The developed ANN model was tested on a number of subject who produced systolic (SBP) and diastolic (DBP) blood pressure predictions. Before that, blood pressure was measured with a conventional tensimeter before and after testing with the developed method. The average error of the test compared to the conventional tensimeter was 11.2 mmHg for SBP and 9.75 mmHg for DBP.

Item Type: Thesis (Diploma)
Uncontrolled Keywords: Tekanan darah (BP) memiliki hubungan erat dengan hipertensi, yang menjadi faktor penyebab utama berbagai penyakit kardiovaskular (CVD) setelah diabetes. Hipertensi, yang sering kali tidak disadari oleh penderita, menjadi penyebab signifikan penyakit jantung seperti serangan jantung dan stroke sehingga pengukuran BP secara real time dan kontinu sangat diperlukan untuk mendiagnosis hipertensi dengan akurat dan membantu penyesuaian pengobatan. Saat ini, metode pengukuran tekanan darah kontinu yang tersedia hanya dilakukan secara invasif yang menyakitkan, sedangkan metode non-invasif dengan manset belum memungkinkan pengukuran kontinu yang nyaman. Untuk mengatasi masalah ini, penelitian ini berfokus pada prediksi tekanan darah non-invasif tanpa manset menggunakan pengukuran Pulse Transit Time (PTT), yang mengukur delay waktu antara gelombang R-puncak pada elektrokardiogram (ECG) dan sinyal puncak photoplethysmogram (PPG). Dengan memanfaatkan machine learning, khususnya Artificial Neural Network (ANN), penelitian ini bertujuan untuk mendapatkan estimasi tekanan darah sistolik (SBP) dan diastolik (DBP) yang lebih akurat. Metode ini memungkinkan pengukuran tekanan darah tanpa tekanan eksternal pada tubuh, sehingga mengurangi risiko cedera atau infeksi. Model ANN yang dikembangkan memiliki nilai error MAE ± STD untuk SBP sebesar 13.89±13.52 mmHg dan untuk DBP sebesar 5.5±6.47 mmHg. Dilakukan pengujian model ANN yang dikembangkan kepada sejumlah subjek yang menghasilkan prediksi tekanan darah sistolik (SBP) dan diastolik (DBP). Sebelum itu, dilakukan pengukuran tekanan darah dengan tensimeter konvensional sebelum dan sesudah pengujian dengan metode yang dikembangkan. Didapatkan error rata-rata pengujian dibandingkan dengan tensimeter konvensional sebesar 11.2 mmHg untuk SBP dan 9.75 mmHg untuk DBP. Blood pressure (BP) has a close relationship with hypertension, which is the leading causative factor of many cardiovascular diseases (CVD) after diabetes. Hypertension, which is often unrecognized by sufferers, is a significant cause of heart diseases such as heart attacks and strokes, so real time and continuous BP measurement is necessary to accurately diagnose hypertension and assist with treatment adjustments. Currently, the available continuous blood pressure measurement methods are only performed in a painful invasive manner, while the non-invasive method with a cuff does not yet allow convenient continuous measurement. To address this issue, this study focuses on non-invasive cuffless blood pressure prediction using Pulse Transit Time (PTT) measurement, which measures the time delay between the R-peak wave on the electrocardiogram (ECG) and the photoplethysmogram (PPG) peak signal. By utilizing machine learning, specifically Artificial Neural Network (ANN), this study aims to obtain more accurate estimation of systolic (SBP) and diastolic (DBP) blood pressure. This method allows blood pressure measurement without external pressure on the body, thus reducing the risk of injury or infection. The developed ANN model has an error value of MAE ± STD for SBP of 13.89±13.52 mmHg and for DBP of 5.5±6.47 mmHg. The developed ANN model was tested on a number of subject who produced systolic (SBP) and diastolic (DBP) blood pressure predictions. Before that, blood pressure was measured with a conventional tensimeter before and after testing with the developed method. The average error of the test compared to the conventional tensimeter was 11.2 mmHg for SBP and 9.75 mmHg for DBP.
Subjects: Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines.
Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression
Q Science > QA Mathematics > QA336 Artificial Intelligence
Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Q Science > QA Mathematics > QA76.9.U83 Graphical user interfaces. User interfaces (Computer systems)--Design.
Q Science > QA Mathematics > QA76 Computer software > QA76.8 Microprocessor
R Medicine > R Medicine (General) > R856.2 Medical instruments and apparatus.
R Medicine > RC Internal medicine > RC683.5.E5 Electrocardiography
T Technology > T Technology (General)
T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5102.9 Signal processing.
T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7878 Electronic instruments
Divisions: Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Biomedical Engineering > 11410-(S1) Undergraduate Thesis
Depositing User: Aushaf Cintya Wanda
Date Deposited: 12 Aug 2024 02:53
Last Modified: 12 Aug 2024 02:54
URI: http://repository.its.ac.id/id/eprint/112075

Actions (login required)

View Item View Item