Syntax Literate: Jurnal Ilmiah Indonesia p–ISSN: 2541-0849 e-ISSN:
2548-1398
Vol. 9, No. 7, Juli 2024
Universitas Indonesia, Depok, Indonesia1,2
Email: [email protected]1, [email protected]2
The development of the field
of health technology has progressed rapidly - especially in the aspect of being
connected to the Internet and also allowing the storage and monitoring of vital
signs data utilizing IoT features. However, there is no health monitoring that
contains advanced check-up analysis referring to the acquisition of vital signs
data that has been stored in the health data record. This is exactly where
additional parameters beyond the vital signs have not been integrated in health
monitoring. The purpose of this independent research is to find a gap between
industry and academia in the form of additional parameters that are not yet
available in the industrial world. The additional parameter is the percentage
value of health diagnostic risk. The selection of this parameter is based on
the need to analyze the level of diagnostic risk through the acquisition of
vital signs that have available readings in health monitoring device equipment.
The algorithm mechanism itself revolves around mapping for the diagnosis of health
conditions referring to the normal limits of vital signs utilizing the decision
tree algorithm. The goal is none other than to simplify the flow in determining
the patient's advanced health diagnosis. Regarding the diagnostic risk
percentage parameter, the diagnosis and calculation include five vital signs
that are the main indicators: heart rate, oxygen saturation (SPO2),
body temperature (TBody) supplemented with skin temperature (TSkin),
and respiratory rate.
Keywords: check-up
analysis, diagnostic risk, diagnostic risk percentage, health monitoring, medical check-up, vital signs
Introduction
In the current era of development, the field of health technology has
experienced a lot of progress. This includes the aspect of monitoring a person's/patient's
health (health monitoring). Now, health data record storage is connected to the
Internet (Da Costa et al., 2018). It can even allow storing and monitoring
vital signs data using IoT features (Swaroop et al., 2019; Chennam et al.,
2021; Lourenço, 2021). However, there is no health monitoring that includes
advanced check-up analysis referring to the acquisition of vital signs data
that has been stored in the health data record. To be precise, here it is in
the form of a diagnostic risk percentage value based on the results of vital
sign readings which have not been integrated into health monitoring.
The diagnostic risk percentage itself is a statistical measure in
percent units used in the medical world to assess the possibility or risk of someone
developing a disease or certain health condition based on the results of a
diagnostic examination referring to the acquisition of vital sign values
during a health examination (Castiglione et al., 2022). Its main role is to
make it easier for medical personnel to make appropriate clinical decisions
referring to obtaining vital signs during health examinations (Shreffler, Genova,
& Huecker, 2023).
Regarding the percentage of diagnostic risk, the role of smartwatches
has only reached the stage of providing various health indicators - including
vital signs - such as heart rate, sleep patterns and physical movements
(Masoumian et al., 2023; Kabe et al., 2020). For example, on the Xiaomi Mi Band
2 smartwatch equipment which is integrated with the Mi Fit application, apart
from checking vital signs such as heart rate, both on the smartwatch and the
application screen, it can display several additional parameters, including the
following: number of steps (step counter), total distance walked (step length/ distance),
calories burned, and sleep quality (Kang et al., 2020; Youssef et al., 2020).
Also, the Viatom CheckMe Pro equipment, apart from monitoring sleep quality,
also has analysis of crucial additional parameters in the form of perfusion
index (PI) in peripheral tissue through the reading of two basic vital signs
like oximetry equipment: heart rate and oxygen saturation (Weenk et al., 2018;
Sahu et al., 2022).
The aim of this research is to find the gap between industry and
academics in the form of additional parameters that do not yet exist in the
industrial world. The context in this case leads to a practical gap in the
industrial world where these additional parameters are not yet available in
many patient monitoring equipment in hospitals, as well as portable monitoring
equipment and smartwatches.
Research Methods
Diagnosis of Health Conditions
In order to be able
to know the patient's health condition when carrying out a diagnosis at the
advanced check-up stage, which is able to map the symptoms of abnormalities/ diseases if
they are present in the patient's body, it is necessary to set normal limits
for each vital sign (Michaud et al., 2021) in carrying out decision making using
algorithm. And this diagnosis also includes advanced check-up analysis that
combines analysis of all vital signs on health monitoring equipment: a clear
explanation of the meaning of the results, an overall summary, and a more
in-depth analysis. Even if necessary, directions for consultation or further
examination are also included (Leite, Gruber & Hodgkinson, 2020).
Vital Signs Risk
Scoring Mechanism
Referring to the
early detection scoring system, equipped with reading limit values for
mapping abnormalities for each vital sign in Table 1, along with the overall
risk scoring mechanism for each vital sign, as well as markings for the overall
total diagnostic risk score (RD).
Table 1. Overall Mechanism of Vital Sign Risk Score Scoring and Final
Diagnostic Risk (RD) Percentage Scale
|
Risk score scale (R) |
3 |
2 |
1 |
0 |
1 |
2 |
3 |
|
heart rate (BPM) |
<40 |
40-50 |
51-59 |
60-100 |
101-110 |
111-130 |
<130 |
|
respiratory frequency (RR) |
|
<9 |
9-11 |
12-20 |
21-30 |
31-39 |
<40 |
|
Body Temperature (oC) |
< 34 |
34.1-35 |
35-35.9 |
36-37.4 |
37.5-37.9 |
38-39.9 |
<40 |
|
oxygen saturation (%) |
<90 |
90-92 |
92-94 |
95-100 |
|
|
|
|
Diagnostic
risk score scale (RD) |
0-20% |
21-40% |
41-60% |
61-80% |
81-100% |
|
|
Research Methodology Flow
The research began by
conducting a literature study to determine the most significant vital signs in
measuring diagnostic risk. After analyzing various scientific sources to
identify the main vital signs in order to determine the percentage of
diagnostic risk, four vital signs were finally selected sequentially according
to the priority scale as a reference regarding diagnostic risk: BPM (heart
rate), RR (respiratory frequency), and TBody (Body Temperature), and SPO₂
(oxygen saturation).
The next step is to understand the value
limits and the scoring calculation mechanism for each vital sign that has been
identified. This is done through in-depth literature study to ensure that each
vital sign is assigned an appropriate value based on the clinical condition at
hand.
The research then continued by studying
various existing regression and classification algorithms. The goal is to find
the most suitable and effective algorithm for predicting diagnostic risk based
on the vital signs that have been collected.
After evaluating various algorithms, the
decision tree became the preferred algorithm used in this research. This is
because the decision tree approach refers to the limits of obtaining values
for each vital sign.
The next step is to prepare a dataset
consisting of the vital signs that have been identified. This dataset includes
input in the form of four vital signs, namely BPM, SPO₂, TSkin, and TBody, as
well as output in the form of the RR vital sign that you want to predict.
Datasetsthat
has been prepared is then used for the training, validation and testing
processes with the regression algorithm available in MATLAB R2021b software.
The goal is to produce a model that is able to predict RR with high accuracy
based on vital signs input.
The prediction model
that has been successfully validated is then implemented in coding form on the
Arduino IDE platform. This allows RR predictions to be carried out in real-time
into health monitoring device equipment.
The final step is to
create coding for the Decision Tree classification which can determine the
percentage of diagnostic risk. This coding is also implemented in the Arduino
IDE to enable automatic diagnostic risk assessment based on vital signs
contained in the health monitoring device equipment.
When summarized, this
research follows a systematic methodological flow starting from literature
study to determine important vital signs, understanding the scoring mechanism,
choosing the right algorithm, to implementing coding on the Arduino IEE
platform for prediction and classification using health monitoring device
equipment. With these steps, the study aims to produce a system that can
accurately predict RR and assess diagnostic risk based on patient vital signs.
Combination of Regression and Classification
In general, this research combines regression
and classification approaches in the use of its algorithm. A regression
approach was used to obtain predictions of respiratory frequency (RR) vital
sign readings. Meanwhile, a classification approach, using a decision tree
algorithm, is used to obtain a diagnostic risk percentage.
Data
retrieval
The following data is collected on patients, namely as
follows:
1)
Collecting data on vital signs on patients who have
previously been taken by health monitoring devices made by researchers (Juan
Karnadi et al, 2021) where all vital signs (Juan Karnadi et al, 2020) have been
validated against a tool that is trusted to be a validator (Juan Karnadi et al,
2024) as input.
2)
Using a special respiratory frequency sensor which is
not part of the health monitoring device equipment) as an output reference base
in predicting RR vital sign values.
3)
Requires training on overall vital sign data at points
1 and 2 with MATLAB R2021b software to then obtain predicted results for RR
vital sign values.
4)
Perform vital sign scoring calculations and diagnostic
risk percentages using the vital sign data that has been obtained.
Results and Discussion
Comparison of Vital Signs
Availability and Diagnostic Risk Percentage Parameter Output
The following is a comparison table
between smartwatch equipment, portable monitoring, and health monitoring
devices made by researchers (see Table 2)
regarding the availability of vital signs, efforts to add new vital signs, and
additional parameters that exist or are the target output/goal of this
research.
Table 2. Comparison of Vital Signs Availability, Efforts to Add New
Vital Signs, and Additional Parameters in Equipment
|
Tool/Device Name |
Availability of Vital Signs |
Efforts to Add New Vital Signs |
Additional Parameters |
|
Xiaomi Mi Band 2
Smart Watch |
Heart Rate &
Oxygen Saturation |
- |
Sleep Quality |
|
Viatom CheckMe Pro |
Heart Rate, Oxygen
Saturation & Temperature |
- |
Perfusion Index |
|
Health Monitoring
Heart Rate, Saturation Device (Made by Researchers) |
Heart Rate, Oxygen
Saturation, Body Temperature & Skin Temperature |
Respiratory
Frequency |
Diagnostic Risk
Percentage (Research Objective)) |
Based on the table above, the diagnostic
risk percentage parameter output is not yet available on the comparable
smartwatch or portable monitoring equipment. So, in this study, it is necessary
to include a diagnostic risk percentage. The purpose of efforts to complete
diagnostic risk percentages into health monitoring device equipment is to make
it easier for medical parties to make decisions based on the condition of the
patient's vital signs.
Basis for Establishing the Diagnostic Risk Percentage Formula
The main consideration that is the
basis for forming a diagnostic risk percentage formula - apart from continuous
and non-continuous monitoring which has been mentioned in the literature review
- is the high cost and mortality caused by one non-communicable disease:
cardiovascular disease. Another consideration is also looking at the prevalence
of sufferers of non-communicable diseases in Indonesia - specifically
cardiovascular diseases. The two cardiovascular diseases that are priority
attention here are stroke and heart disease.
Stroke is closely related to the
vital sign of respiratory frequency. Meanwhile, diseases that lead to heart
disease are closely related to the vital sign heart rate. Based on data on the
prevalence of non-communicable diseases from Riskesdas (2018), 109 out of 1000
people suffer from stroke; while 15 out of 1000 people suffer from heart
disease. Then other non-communicable diseases, namely asthma, correlate with
the vital signs of respiratory frequency and oxygen saturation. 24 out of 1000
people suffer from asthma (Indonesian Ministry of Health, 2018).
Priority Determination of Vital Signs in Diagnostic Risk
From a medical/clinical aspect, the
vital signs mentioned in the literature review section are able to provide an
actual picture of a person's physical health condition and are considered to
have the most essential information in mapping a patient's health for medical
parties - especially health workers. This was confirmed after thoroughly
observing the medical needs. There was also a post-observation discussion with
medical personnel who were skilled in using patient monitoring equipment at the
hospital.
On the other hand, regarding the
availability of vital signs, health monitoring devices made by researchers with
available vital signs which have been discussed in Table 2
previously, are the preference in determining the priority of vital signs in
diagnostic risk. Heart rate, oxygen saturation, skin temperature and available
body temperature - as well as the addition of the vital sign respiratory
frequency - have covered almost all the most crucial vital signs in medical
personnel. Only the vital sign blood pressure - because it is included in the
most crucial vital sign - is not included in this health monitoring device (Levental et al., 2018; Kumar &
Krishnamoorthi, 2021).
Regarding the implementation of
health checks, the health monitoring device used - with reading results that
can also be accounted for - makes it easy to carry out short assessments. This
convenience is another factor when determining the priority of vital signs
which are closely related to diagnostic risk. The three aspects mentioned in
this subchapter are taken into consideration in determining the priority of
vital signs regarding diagnostic risk using health monitoring devices made by
researchers in research regarding the percentage of diagnostic risk.
_3836%20-%203847.fld/image001.jpg)
Establishment of Scoring Mechanism and Diagnostic Risk Percentage
The formation of the scoring
mechanism and diagnostic risk percentage combines existing domain knowledge in
health scientific aspects, as well as aspects of algorithm design and use. The
health scientific aspects referred to here include vital sign diagnostic risk
mapping and early detection systems - through early warning scores. Meanwhile,
aspects of design and use of the algorithm utilize a classification algorithm
approach where the decision tree is the main reference and reference algorithm.
Outputsfrom
the formation of a mechanism which then becomes a framework, it is called a
framework because it can comprehensively summarize the mapping of a person's
health condition and the potential diagnostic risks he or she experiences
simply by referring to the limits of the output values of existing vital
signs. Even further, from this quantitative approach, the health condition of
each vital sign that has been mapped earlier, is summarized again through the
classification of the range of diagnostic risk percentage values, which
provides a comprehensive picture of the overall health condition of the
person/patient being examined.
Data Collection Approach
Regarding data collection on patient
signs that was carried out, with a total of 100 respondents obtained, the data
collection technique was carried out using an experimental approach using
health monitoring devices made by researchers. Meanwhile, conceptually, the
method used is the volunteer lab test method. To be precise, this was carried
out for 3 months starting from February to April 2024 at the FT UI
Manufacturing Research Center (MRC) Building.
Prediction of Respiratory Frequency Readings
_3836%20-%203847.fld/image002.png)
Utilizing the neural network feature
contained in MATLAB R2021b software, crucial steps to obtain respiratory
frequency reading values are taken so that they can be implemented into
health monitoring device equipment. The input uses the readings of the following
four vital signs: heart rate, oxygen saturation, skin temperature and body
temperature. And there are 8 layers used (see Figure 2)
to carry out predictions of respiratory frequency readings. What has been
executed by MATLAB, will be converted into the Arduino IDE programming
language.
As a result, the prediction of
respiratory frequency readings obtained from MATLAB software is highly
correlated with the respiratory frequency values obtained from other sensor
tests where data collection is carried out with separate equipment outside the
health monitoring device and this is the output target in training. The error
obtained in the respiratory frequency prediction results obtained during
training, validation, and testing for the respiratory frequency value used as
the output reference is close to 0 as in Figure 4.2 below. However, the range
of input values for heart rate and oxygen saturation still does not vary:
heart rate is in the range of 62 to 77 BPM; oxygen saturation in the range of
95 to 97%.
_3836%20-%203847.fld/image003.png)
Interpretation of the Diagnostic Risk
Percentage Scale
From the diagnostic risk percentage value scale that has been described, the following is an explanation of the classification. A value of 0-20% describes the condition of the patient who is still in a normal/healthy condition. Meanwhile, a value of 21-40% indicates that the patient's health condition is starting to require attention. Then a value of 41-60% means that the patient/person needs to undergo further examination. Meanwhile, a value of 61-80% means that the person has begun to need to have their health monitored regularly. And a value in the range of 81-100% gives an interpretation that the patient is in a very critical condition.
From a total of 100 respondents/patients whose data were taken, the values obtained for heart rate, oxygen saturation or body temperature did not indicate any significant risk (R = 0). Meanwhile, for respiratory frequency risk scoring, 48 of them had no risk related to respiratory frequency. The remaining 52 people showed a mild risk level for experiencing non-communicable diseases involving respiratory problems. Even so, the overall RD percentage value gives an idea that all respondents/patients whose data were taken are in a healthy condition. The summary can be seen in Table 3 below.
Table 3. Summary of Scoring Results and Percentage of
Health Diagnosis Risk
|
Number of subjects |
RBPM |
RSPO2 |
RTBODY |
RRR |
RD |
|
48 peoples |
0 (healthy) |
0 (healthy) |
0 (healthy) |
0 (healthy) |
0% (Normal) |
|
52 peoples |
1 (low risk) |
10% (Normal) |
Results and Analysis of Obtaining Diagnostic Risk Percentage Values
Table 4. Diagnostic Risk
Percentage Results Based on Obtaining Vital Signs, Heart Rate (BPM), Oxygen
Saturation (SPO2), and Body Temperature (TBODY)
|
BPM |
SPO2 (%) |
TBODY (°C) |
RR/min |
Diagnostic Risk |
|
69 |
97 |
36.3 |
19 |
0% |
|
62 |
97 |
36.3 |
13 |
0% |
|
77 |
97 |
36.2 |
25 |
10% |
|
67 |
97 |
36.1 |
18 |
0% |
|
70 |
96 |
36.4 |
15 |
0% |
|
74 |
96 |
36.4 |
22 |
10% |
|
67 |
97 |
36.3 |
13 |
0% |
|
73 |
97 |
36.1 |
23 |
10% |
|
73 |
97 |
36.2 |
19 |
0% |
|
71 |
97 |
36.2 |
20 |
0% |
|
64 |
97 |
36.3 |
13 |
0% |
|
67 |
96 |
36.1 |
15 |
0% |
|
68 |
96 |
36.1 |
16 |
0% |
|
72 |
96 |
36.5 |
18 |
0% |
|
76 |
95 |
36.4 |
21 |
10% |
|
72 |
97 |
36.2 |
20 |
0% |
|
77 |
97 |
36.2 |
25 |
10% |
|
73 |
96 |
36.3 |
25 |
10% |
|
71 |
97 |
36.3 |
17 |
0% |
|
71 |
95 |
36.3 |
16 |
0% |
|
78 |
96 |
36.1 |
27 |
10% |
|
75 |
96 |
36.3 |
28 |
10% |
|
81 |
96 |
36.2 |
29 |
10% |
|
70 |
96 |
36.2 |
18 |
0% |
|
78 |
96 |
36.2 |
25 |
10% |
|
80 |
96 |
36.2 |
26 |
10% |
|
70 |
96 |
36.1 |
18 |
0% |
|
77 |
96 |
36.3 |
27 |
10% |
|
72 |
96 |
36.5 |
18 |
0% |
|
73 |
96 |
36.2 |
21 |
10% |
|
76 |
97 |
36.1 |
27 |
10% |
|
81 |
96 |
36.1 |
28 |
10% |
|
72 |
96 |
36.4 |
16 |
0% |
|
73 |
96 |
36.3 |
29 |
10% |
|
72 |
96 |
36.3 |
28 |
10% |
|
74 |
96 |
36.4 |
22 |
10% |
|
68 |
96 |
36.3 |
10 |
0% |
|
75 |
97 |
36.2 |
21 |
10% |
|
78 |
96 |
36.1 |
26 |
10% |
|
73 |
96 |
36.5 |
18 |
0% |
|
76 |
96 |
36.1 |
24 |
10% |
|
67 |
97 |
36.4 |
16 |
0% |
|
81 |
96 |
36.1 |
28 |
10% |
|
77 |
96 |
36 |
24 |
10% |
|
78 |
96 |
36.3 |
26 |
10% |
|
79 |
97 |
36.1 |
28 |
10% |
|
75 |
97 |
36.2 |
22 |
10% |
|
73 |
97 |
36.3 |
23 |
10% |
|
66 |
97 |
36.1 |
18 |
0% |
|
80 |
95 |
36.3 |
21 |
10% |
|
73 |
97 |
36.5 |
21 |
10% |
|
71 |
97 |
36.4 |
17 |
0% |
|
78 |
96 |
36.2 |
26 |
10% |
|
67 |
96 |
36.1 |
15 |
0% |
|
75 |
97 |
36.2 |
21 |
10% |
|
78 |
96 |
36.1 |
26 |
10% |
|
68 |
97 |
36.5 |
20 |
0% |
|
69 |
96 |
36.5 |
17 |
0% |
|
69 |
96 |
36.8 |
13 |
0% |
|
69 |
96 |
36 |
22 |
10% |
|
73 |
96 |
36.7 |
16 |
0% |
|
72 |
96 |
36.5 |
18 |
0% |
|
70 |
96 |
36 |
23 |
10% |
|
71 |
96 |
36.3 |
25 |
10% |
|
69 |
96 |
36.3 |
15 |
0% |
|
69 |
96 |
36.3 |
15 |
0% |
|
69 |
96 |
36.3 |
15 |
0% |
|
69 |
97 |
36.2 |
19 |
0% |
|
72 |
96 |
36.3 |
28 |
10% |
|
69 |
97 |
36.2 |
19 |
0% |
|
69 |
96 |
36.2 |
17 |
0% |
|
70 |
96 |
36.4 |
16 |
0% |
|
69 |
96 |
36.1 |
17 |
0% |
|
69 |
96 |
36.3 |
15 |
0% |
|
69 |
96 |
36.3 |
15 |
0% |
|
74 |
96 |
36.2 |
21 |
10% |
|
68 |
96 |
36.4 |
16 |
0% |
|
69 |
97 |
36.1 |
17 |
0% |
|
69 |
96 |
36.2 |
17 |
0% |
|
69 |
96 |
36.6 |
17 |
0% |
|
69 |
96 |
36.6 |
17 |
0% |
|
69 |
96 |
36.3 |
15 |
0% |
|
70 |
96 |
36.5 |
18 |
0% |
|
72 |
96 |
36.3 |
28 |
10% |
|
73 |
96 |
36.3 |
29 |
10% |
|
75 |
97 |
36.8 |
21 |
10% |
|
72 |
96 |
36.6 |
18 |
0% |
|
77 |
96 |
36.3 |
27 |
10% |
|
72 |
96 |
36.2 |
19 |
0% |
|
72 |
97 |
36.3 |
21 |
10% |
|
73 |
96 |
36.2 |
21 |
10% |
|
78 |
96 |
36.2 |
26 |
10% |
|
77 |
97 |
36.2 |
25 |
10% |
|
74 |
97 |
36.2 |
20 |
0% |
|
72 |
96 |
36.2 |
20 |
0% |
|
73 |
97 |
36.1 |
23 |
10% |
|
70 |
97 |
36.3 |
24 |
10% |
|
70 |
97 |
36.2 |
19 |
0% |
|
82 |
96 |
36.3 |
26 |
10% |
|
71 |
96 |
36.2 |
19 |
0% |
From the obtained diagnostic risk percentage values, there is one similar pattern. The similarity is that all the values obtained for heart rate, oxygen saturation and body temperature are still within normal limits and the risk scale is zero (R = 0). As a result, the data obtained along with the diagnostic risk percentage value is less able to describe the sensitivity of the health monitoring device to at least those who have abnormalities or indeed one of whose vital signs is slightly outside the normal range other than healthy people. Apart from that, for dozens of respondents, the respiratory frequency reading values obtained on the health monitoring device also tended to be inaccurate. In other words, obtaining normal values for other vital signs is not followed by normal respiratory frequency readings. This is because it is influenced by the range of vital sign values which is not that large. So, more data collection is needed, with a greater range of values for vital signs than before - including for body temperature and skin temperature, the largest range of which can only reach 0.6 °C. So that later when training is carried out, up to the validation and testing stages, predictions of respiratory frequency readings can also be included when vital sign values are obtained which are currently not within the range, so that the sensitivity of the equipment readings can be better described.
The most crucial aspect that influences the diagnostic risk percentage value obtained in this study is the sensitivity of the health monitoring device made by researchers - this is most visible in the prediction of respiratory frequency vital sign readings. This means that there is still a need for data collection with a more diverse range of acquisition and range of vital sign values. So that it can further increase the accuracy and precision of vital sign readings from health monitoring devices - especially respiratory frequency vital signs. So that later the diagnostic risk percentage results obtained can be accounted for and reflect the mapping of health diagnoses based on vital conditions in the entire human body. Regarding the questions and objectives of this research, the following conclusions can be outlined; (1) the diagnostic risk percentage parameter covers the need for parameters that do not yet exist in smartwatches - as well as portable monitoring equipment. In diagnostic risk analysis, the biggest role of this parameter lies in the selection of vital signs which are considered to play an essential role in the medical world, (2) the development of a diagnostic risk percentage framework in the form of a combination of a regression approach for predicting respiratory frequency readings and classification using a decision tree algorithm in a diagnostic risk scoring and scaling mechanism is able to provide an actual picture of a person's physical health condition, (3) the existence of a diagnostic risk percentage parameter can fill the practical gap in health monitoring in the form of a comprehensive summary of health diagnosis mapping which is not yet available on patient monitoring equipment in hospitals, as well as portable monitoring equipment and smartwatches, and (4) in this research, the interpretation results have not been accommodated in the implementation of algorithm programming with the Arduino IDE.
BIBLIOGRAPHY
Chennam,
K. K., Uma Maheshwari, V., & Aluvalu, R. (2021). Maintaining IoT healthcare
records using cloud storage. In IoT and IoE Driven Smart Cities (pp.
215–233). Springer.
Castiglione, V., Aimo, A., Vergaro,
G., Saccaro, L., Passino, C., & Emdin, M. (2022). Biomarkers for the
diagnosis and management of heart failure. Heart failure reviews, 1-19.
Da Costa, C. A., Pasluosta, C. F.,
Eskofier, B., Da Silva, D. B., & da Rosa Righi, R. (2018). Internet of
health things: toward intelligent vital signs monitoring in hospital wards.
Artificial intelligence in medicine, 89, 61-69.
Huecker, M., Genova, G., & Shreffler, J. (2023). A
Scoping Review of Physical Activity Interventions for Physician Wellness. American Journal of Lifestyle Medicine, 15598276231178485.
Kang, S., Kim, S., & Kim, J.
(2020). Forensic analysis for IoT fitness trackers and its application.
Peer-to-Peer Networking and Applications, 13, 564-573.
Leite, H., Hodgkinson, I. R., & Gruber, T. (2020).
New development:‘Healing at a distance’—telemedicine and COVID-19. Public money & management, 40(6), 483-485.
Kebe, M., Gadhafi,
R., Mohammad, B., Sanduleanu, M., Saleh, H., & Al-Qutayri, M. (2020). Human
vital signs detection methods and potential using radars: A review. Sensors,
20(5), 1454.
Kumar, V. S., & Krishnamoorthi, C. (2021). Development of
electrical transduction based wearable tactile sensors for human vital signs monitor:
Fundamentals, methodologies and applications. Sensors
and Actuators A: Physical, 321, 112582.
Levental, S., Picard, E., Mimouni, F., Joseph, L., Samuel, T.
Y., Bromiker, R., ... & Goldberg, S. (2018). Sex‐linked difference in blood
oxygen saturation. The
Clinical Respiratory Journal, 12(5), 1900-1904.
Lourenço, D. E. M. (2021). EldyIoT:
IoT Assistive System for Elderly (Master's thesis,
ISCTE-Instituto Universitario de Lisboa (Portugal)).
Michaud, K., Pedro, S., Wipfler, K., Agarwal, E.,
& Katz, P. (2021). Changes in Disease‐Modifying Antirheumatic Drug
Treatment for Patients With Rheumatoid Arthritis in the US During the COVID‐19
Pandemic: A Three‐Month Observational Study. Arthritis care & research, 73(9), 1322-1331.
Masoumian,
H. M., Masoumian H. S. T., Qayumi, K., Hosseinzadeh, S., & Sajadi Tabar, S.
S. (2023). Smartwatches in healthcare medicine: assistance and monitoring; a
scoping review. BMC Medical Informatics and Decision Making, 23(1),
248.
Sahu, M. L., Atulkar, M., Ahirwal, M. K., & Ahamad, A.
(2022). Vital sign monitoring system for healthcare through IoT based personal
service application. Wireless
Personal Communications, 122(1), 129-156.
Swaroop, K. N.,
Chandu, K., Gorrepotu, R., & Deb, S. (2019). A health monitoring system for
vital signs using IoT. Internet of Things, 5, 116-129.
Weenk, M., van Goor,
H., van Acht, M., Engelen, L. J., van de Belt, T. H., & Bredie, S. J.
(2018). A smart all-in-one device to measure vital signs in admitted patients.
PloS one, 13(2), e0190138.
Youssef A. A. A., Wouters, F., Vranken,
J., de Korte-de Boer, D., Smit-Fun, V., Duflot, P., ... & Vanrumste, B.
(2020). Vital signs prediction and early warning score calculation based on
continuous monitoring of hospitalised patients using wearable technology. Sensors,
20(22), 6593.
|
Copyright holder: |
|
First publication right: Syntax Literate: Jurnal Ilmiah
Indonesia |
|
This article is licensed under:
|
_3836%20-%203847.fld/image004.jpg){kind=small}