| Title | | the title of the article |
| Authors | | the authors of the article |
| Year | | the year the article was published or submitted (for non-peer-reviewed papers) |
| Calibration Technique | | calibration technique used to develop and test model |
| | personalization | subjects used to develop model are included in testing. only first datapoints of each subject are used to initialize developed model for testing. |
| | subject split | subjects used to develop model are not included in testing |
| Evaluation Metric | | performance metric of the reported result |
| | ME | study reported bias±standard deviation |
| | MAE | study reported mean absolute error |
| Reported Result | | the claimed MAE/ME result from the article in mmHg |
| Sensor Data | | data used for BP estimtion |
| | PPG | device used photoplethysmography |
| | contact pressure | device required contact applying pressure |
| | multi-wavelength PPG | device used multiple wavelengths for photoplethysmography |
| | tonometer | device used tonometer to record arterial BP waveform |
| | ECG | device used electrocardiography |
| | biometrics | device used biometric information (i.e. age, height, gender) |
| | pressure sensor | device used pressure or force sensor |
| | ICG | device used impedance cardiography |
| | IPG | device used impedance plethysmography |
| | ultrasound | device used ultrasound |
| | PCG | device used phonocardiography |
| | SCG | device used seismocardiography |
| | accelerometer | device used accelerometer to detect motion |
| | BCG | device used ballistocardiography |
| Dataset | | dataset used for BP estimtion |
| | Internal | dataset is not available to the public |
| | MIMIC | MIMIC Waveform Database from Johnson et al., 2020. can also be MIMIC-I, II, III, or later versions. can also be a subset of the whole dataset |
| | PPG-BP | PPG-BP dataset from Liang et al., 2018 |
| | EVAL Stress Test | Non-invasive Blood Pressure Estimation from Esmaili et al., 2017 |
| | UoQ | University of Queensland Vital Signs Dataset from Liu et al., 2012 |
| | CHARIS | CHARIS dataset from physionet |
| | VitalDB | VitalDB dataset from Lee et al., 2022 |
| | HYPE | HYPE dataset from Sasso et al., 2020 |
| Algorithm | | algorithm used for BP estimation |
| | Classical ML | classical machine learning techniques such as multiple linear regression, decision trees, random forest |
| | Physiological Model | model derived from physiological principles such as Windkessel, Moens-Korteweg |
| | Deep Learning | deep learning techniques such as convolutional neural networks, long term short term memory, recurrent neural networks |
| | proprietary | algorithm is not available to the public |
| Number of Test Subjects | | number of subjects used to test the model. does not include subjects used to train the model. sometimes computed from given train-test-split percentages and total number of subjects. |
| Testing Subject Characteristics | | whether testing subjects are healthy or have medical issues |
| | healthy | testing subjects are specified as “healthy” or have no reported medical issues |
| | diseased | testing subjects have medical issues such as hypertension or are ICU/hospital patients |
| SBP STD | | Systolic Blood Pressure Standard Deviation of Error in mmHg |
| DBP STD | | Diastolic Blood Pressure Standard Deviation of Error in mmHg |
| BP Distribution SBP STD | | Systolic Blood Pressure Standard Deviation of subject used in study in mmHg. If no testing subject BP distribution is given, the provided BP distribution of the whole dataset is used. |
| BP Distribution DBP STD | | Dystolic Blood Pressure Standard Deviation of subject used in study in mmHg. If no testing subject BP distribution is given, the provided BP distribution of the whole dataset is used. |
| Time between Calibration and Test | | the time between calibration step and test step for personalization studies. |
| | s | seconds |
| | m | minutes |
| | h | hours |
| | d | days |
| | mon | months |
| Keywords | | nouns from extracted data using NLTK |