Academia.eduAcademia.edu

Figure 3 – uploaded by International Journal of Electrical and Computer Engineering (IJECE)

See full PDFdownloadDownload figure
Table 3. Results of blood pressure for sensor and original device 3.3. Body heat analysis The patients’ body heat results as they are shown on the doctor’s application in the monitoring room device screen comprise the case study that is put into practice. Each patient’s body heat results are tabulated in Table 4. In the patient “Fahad Ahmed”, the test result is shown in red since the patient’s temperature is higher than usual. This table also displays the results comparison with a medical thermometer. The results as seen on the device screen for the doctor’s desktop program in the monitoring room, two patients in the case study are now in use. Each patient’s blood pressure readings are added together in the table. In order to compare the results, the patient’s blood pressure rate is also monitored using a typical blood pressure monitor, as shown in Table 3. As seen, between the sensor and the natural device, the output data values are so close, demonstrating the outperformance of the proposed system [21].

Table 3 Results of blood pressure for sensor and original device 3.3. Body heat analysis The patients’ body heat results as they are shown on the doctor’s application in the monitoring room device screen comprise the case study that is put into practice. Each patient’s body heat results are tabulated in Table 4. In the patient “Fahad Ahmed”, the test result is shown in red since the patient’s temperature is higher than usual. This table also displays the results comparison with a medical thermometer. The results as seen on the device screen for the doctor’s desktop program in the monitoring room, two patients in the case study are now in use. Each patient’s blood pressure readings are added together in the table. In order to compare the results, the patient’s blood pressure rate is also monitored using a typical blood pressure monitor, as shown in Table 3. As seen, between the sensor and the natural device, the output data values are so close, demonstrating the outperformance of the proposed system [21].

Related Figures (12)

An internet of things-based healthcare system performing on a prediction ... (Fahad Ahmed Shaban)
An internet of things-based healthcare system performing on a prediction ... (Fahad Ahmed Shaban)
2.2. Blood pressure measurement
2.2. Blood pressure measurement
Figure 2. Heart pulse rate sensor with Raspberry Pi connection diagram The specifications of the heartbeat are mentioned in Table 1. A significant DC signal that indicates leftover venous and arterial blood as well as bloodless tissues is detected by the photodetector. The arterial pulse (1% of the observed signal) is represented by the AC signal. Through the use of a communication protocol, the system gathers data from sensors and send it to the doctor’s monitoring system [11].
Figure 2. Heart pulse rate sensor with Raspberry Pi connection diagram The specifications of the heartbeat are mentioned in Table 1. A significant DC signal that indicates leftover venous and arterial blood as well as bloodless tissues is detected by the photodetector. The arterial pulse (1% of the observed signal) is represented by the AC signal. Through the use of a communication protocol, the system gathers data from sensors and send it to the doctor’s monitoring system [11].
Figure 3. Blood pressure measurement SAe SAREE MEO SR Rn SOP EE EO I BES ee nr ninecenrens Pir crea eRenE Tran Neen a SERS aN TES EO Human blood pressure can be measured non-invasively with the use of a blood pressure sensor. It uses the Oscillo metric method to calculate the mean arterial pressure, diastolic pressure, and systolic pressure [12]. Here, the blood pressure value is displayed online via a Raspberry Pi gateway. The findings demonstrated that the design can use a USB TTL serial cable directly connected to the Raspberry Pi to transfer data from the blood pressure detector to the monitoring system in the treating physician’s computer over the network [13]. Figure 3 displays the embedded system.
Figure 3. Blood pressure measurement SAe SAREE MEO SR Rn SOP EE EO I BES ee nr ninecenrens Pir crea eRenE Tran Neen a SERS aN TES EO Human blood pressure can be measured non-invasively with the use of a blood pressure sensor. It uses the Oscillo metric method to calculate the mean arterial pressure, diastolic pressure, and systolic pressure [12]. Here, the blood pressure value is displayed online via a Raspberry Pi gateway. The findings demonstrated that the design can use a USB TTL serial cable directly connected to the Raspberry Pi to transfer data from the blood pressure detector to the monitoring system in the treating physician’s computer over the network [13]. Figure 3 displays the embedded system.
Figure 4. Heat body sensor with connection method Advanced external temperature measurement has also seen the light of day with the development of sensing technologies, improved monitoring tools, and improved accuracy. Its information might be crucia for patients suffering from various conditions, such as infections and hypothermia. It is possible to determine the body’s surface temperature using skin or body temperature sensors [14]. Various temperature-measuring applications employ thermocouples, negative thermal coefficient (NTC), digital temperature sensors, and thermistor to accommodate wide thermistors circumstances, accuracy, and packaging. As seen in Figure 4, sensors are affixed to the patient’s body, connected to the Internet and cloud via a transmission media and powered by the voltage across the diode ends. The temperature rises in response to an increase in voltage. A voltage drop in the diode between the transistor’s base and emitter terminals occurs next [15].
Figure 4. Heat body sensor with connection method Advanced external temperature measurement has also seen the light of day with the development of sensing technologies, improved monitoring tools, and improved accuracy. Its information might be crucia for patients suffering from various conditions, such as infections and hypothermia. It is possible to determine the body’s surface temperature using skin or body temperature sensors [14]. Various temperature-measuring applications employ thermocouples, negative thermal coefficient (NTC), digital temperature sensors, and thermistor to accommodate wide thermistors circumstances, accuracy, and packaging. As seen in Figure 4, sensors are affixed to the patient’s body, connected to the Internet and cloud via a transmission media and powered by the voltage across the diode ends. The temperature rises in response to an increase in voltage. A voltage drop in the diode between the transistor’s base and emitter terminals occurs next [15].
Figure 5. Random forest prediction approach architecture [18] Random forest prediction uses a cluster learning approach in which multiple decision trees are generated and combined for prediction. Each tree in the forest is predicted independently; however, the final prediction is determined by the combination of predictions for all trees [16]. This procedure helps reduce overfitting and increases model accuracy and robustness. This approach is versatile and effective for classification and post-processing as shown in Figure 5 and is widely used in a variety of industries including healthcare, finance, and commerce [17]. Below is the mathematical representation of this method.
Figure 5. Random forest prediction approach architecture [18] Random forest prediction uses a cluster learning approach in which multiple decision trees are generated and combined for prediction. Each tree in the forest is predicted independently; however, the final prediction is determined by the combination of predictions for all trees [16]. This procedure helps reduce overfitting and increases model accuracy and robustness. This approach is versatile and effective for classification and post-processing as shown in Figure 5 and is widely used in a variety of industries including healthcare, finance, and commerce [17]. Below is the mathematical representation of this method.
Table 2. Result of heart-beat for sensor and original device Int J Elec & Comp Eng, Vol. 14, No. 5, October 2024: 5755-5764
Table 2. Result of heart-beat for sensor and original device Int J Elec & Comp Eng, Vol. 14, No. 5, October 2024: 5755-5764
Figure 6. The proposed system hardware diagram In this section, the values of the sensors connected to the patient’s body are compared with the regular devices and the prediction values based on random forest algorithm, mean square error, as well as the percentage of the error are s crutinized. Figure 6 displays the hardware block diagram implemented for system operations. Here, three input signals are obtained for each patient (temperature, pulse rate, and blood pressure sensors) and then control ed and gathered by the system. To choose the patient, the Raspberry Pi microcontroller sends six output control signals to the database. Wi-Fi is used to transfer control signals (patient and lead selection) and acquired data (temperature, pulse heart rate, and blood pressure sensors) between the database modu e and the base station computer. Through a Wi-Fi network, the obtained data and control signals are sent from the base station (close to the patient) to the distant mobile device or from the mobile device to the caller data. Based on the reported data and verifications, the correct performance of the proposed approach is discussed and certified.
Figure 6. The proposed system hardware diagram In this section, the values of the sensors connected to the patient’s body are compared with the regular devices and the prediction values based on random forest algorithm, mean square error, as well as the percentage of the error are s crutinized. Figure 6 displays the hardware block diagram implemented for system operations. Here, three input signals are obtained for each patient (temperature, pulse rate, and blood pressure sensors) and then control ed and gathered by the system. To choose the patient, the Raspberry Pi microcontroller sends six output control signals to the database. Wi-Fi is used to transfer control signals (patient and lead selection) and acquired data (temperature, pulse heart rate, and blood pressure sensors) between the database modu e and the base station computer. Through a Wi-Fi network, the obtained data and control signals are sent from the base station (close to the patient) to the distant mobile device or from the mobile device to the caller data. Based on the reported data and verifications, the correct performance of the proposed approach is discussed and certified.
3.4. Error percentage Table 5 contains the test values for temperature, blood pressure, and heart rate. The values generated through the traditional device are compared with the samples generated by the sensors in the proposed method to reach the error rates and determine the accuracy of the results [22]. The error calculation of data is done through (3). From this table, it can be inferred that there are slight differences in pulse rate and temperature between the reading of the sensor and the device due to the generally low and non-variable error percentage rate. It is noted that there is a greater variation in the error rates for blood pressure, which indicates the presence of potential discrepancies in blood pressure measurements between the sensors and devices.
3.4. Error percentage Table 5 contains the test values for temperature, blood pressure, and heart rate. The values generated through the traditional device are compared with the samples generated by the sensors in the proposed method to reach the error rates and determine the accuracy of the results [22]. The error calculation of data is done through (3). From this table, it can be inferred that there are slight differences in pulse rate and temperature between the reading of the sensor and the device due to the generally low and non-variable error percentage rate. It is noted that there is a greater variation in the error rates for blood pressure, which indicates the presence of potential discrepancies in blood pressure measurements between the sensors and devices.
3.5. Random forest prediction = aS Ee RES RIERA Smee ses pee ee a= Se hae Table 6 uses random forest regression models. These models are trained based on supplied data to estimate heart-rate, body-heat, systolic blood pressure (SBP), and diastolic blood pressure (DBP) for each individual to obtain clearer measurements compared to the results of the sensor and traditional device. The projected values for SBP and DBP are displayed in the output. As can be seen, appropriate performance of the developed models could be verified certifying the outperformance of them [23].
3.5. Random forest prediction = aS Ee RES RIERA Smee ses pee ee a= Se hae Table 6 uses random forest regression models. These models are trained based on supplied data to estimate heart-rate, body-heat, systolic blood pressure (SBP), and diastolic blood pressure (DBP) for each individual to obtain clearer measurements compared to the results of the sensor and traditional device. The projected values for SBP and DBP are displayed in the output. As can be seen, appropriate performance of the developed models could be verified certifying the outperformance of them [23].
3.6. MSE analysis The MSE in Table 7 displays the average squared error difference between the sensor values and the predicted body temperature, blood pressure, and coronary heart rates. A lower MSE score indicates a better settlement between the actual and the expected values [24]. The computer MSE, which is based on the capabilities offered, shows that the random forest prediction approach performs rather well in terms of prediction when it comes to determining body heat and coronary heart rates.
3.6. MSE analysis The MSE in Table 7 displays the average squared error difference between the sensor values and the predicted body temperature, blood pressure, and coronary heart rates. A lower MSE score indicates a better settlement between the actual and the expected values [24]. The computer MSE, which is based on the capabilities offered, shows that the random forest prediction approach performs rather well in terms of prediction when it comes to determining body heat and coronary heart rates.
BIOGRAPHIES OF AUTHORS
BIOGRAPHIES OF AUTHORS
Related topics:
Random Forest Regressionphysiological indicators
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved