Description:DESCRIPTION:

The Data Logging System is employed to measure various physiological parameters, including waveform data from a Thermistor Sensor, Electrocardiogram (ECG), Pulse Plethysmography (PPG), and Blood Pressure (BP) readings (Systolic, Diastolic, and Mean Arterial Pressure). These measurements are recorded in an Excel sheet on a Personal Computer.

For ANN Training (supervised manner), the inputs comprise age, sex, pathology, weight, height, pulse transit time, systolic BP, diastolic BP, mean arterial pressure, heart rate, and thermistor data. The output port of the ANN is trained with systolic BP, diastolic BP, and mean arterial pressure.

During Testing, the same inputs as during training are used, but in an online manner. The ANN's output port is not provided with any data. The systolic, diastolic, and mean arterial pressures are evaluated based on the desired time interval.

The invention distinguishes itself from prior art by directly mapping systolic pressure, diastolic pressure, and mean arterial pressure from input to output without the need for prior blood pressure data, making only the present blood pressure information sufficient.

Data collection involves acquiring information from thermistor, ECG, PPG, heart rate, and BP apparatus. Clinical data such as weight, height, kidney-related pathology, sex, and age are also collected.

The aim of the system is to forecast blood pressure for specific time intervals (e.g., 10/15/20/25/30 minutes) during cuff inflation, which occurs every 5 minutes, over a 4-hour dialysis session.

Training involves providing input data as shown in Figure 3 (page 4), using a 10-minute interval for a 2-hour duration. Heart rate and thermistor data should be included in the input for training.

Testing, on the other hand, is performed with inputs as shown in Figure 4 (page 6), using a 10-minute interval for a 2-hour duration. The ANN's output port provides forecast values based on the set time intervals.

The unique feature of our system lies in its ability to map both hyper/hypo tension and provide forecasts for both conditions using the ANN. Additionally, training is done offline, and testing is conducted online, ensuring accuracy and efficiency.

CONDUCTED AND SURVEYED THE METHOD:
Using Data Logging System, Waveform from Thermistor Sensor, ECG(Electrocardiogram), PPG(Pulse Plethysmography), and BP (Blood Pressure) Systolic, Diastolic and Mean Arterial Pressure is measured and the data is written in Excel sheet of Personal Computer.
From ECG & PPG Pulse transit time is calculated.
ANN TRAINING (SUPERVISED MANNER):
The inputs to ANN TRAINING is:
1. Age
2. Sex
3. Pathology
4. Weight
5. Height
6. Pulse transit time
7. Systolic BP
8. Diastolic BP
9. Mean Arterial Pressure
10. Heart Rate
11. Thermistor data

From the output port of ANN:
1. Systolic BP
2. Diastolic BP
3. Mean Arterial Pressure
Is given as input and the ANN is trained in SUPERVISORY MODE. This training is done through
OFFLINE.

TESTING:
1. The input is the same as in input during training. But, here it is ONLINE.
2. In the output side, no data is given in the OUTPUT PORT of ANN.
3. The Systolic, Diastolic and Mean Arterial Pressure is evaluated according to the time interval we need.

BRIEF DESCRIPTION OF THE DRAWINGS:
In page 1 of the attachment gives Figure 1, THE DATA LOGGING SYSTEM. The inputs are Autonomous Nerve Stimulation, Electrocardiogram, Plethysmogram, and Systole, Diastole and Mean Arterial Pressure from Ambulatory Blood Pressure Apparatus.
In page 2 of the attachment gives Figure 2, which explains calculation of pulse transit time.
Page 4 of the attachment gives Figure 3, which explains Training of ANN. Please add Heart Rate (HR) and Thermistor data in the input side of ANN in the attachment of page 4 in Figure 3. The training is done by OFFLINE method. Page 6 of the attachment gives Figure 4, which explains Testing of ANN. Please add Heart Rate(HR) and Thermistor data in the input side of ANN in the attachment of page6 in Figure 4. The testing is done by ONLINE method.
Table 1: (Page 5 of the attachment)
Gives mapping details of the time from input to output. If we measure BP (Systolic, Diastolic and Mean Arterial Pressure) the input at 5 min, the output will give BP at 15 min (Systolic, Diastolic and Mean Arterial Pressure).

DETAILED DESCRIPTION OF THE INVENTION:
We differ from the prior art by mapping the Systolic Pressure, Diastolic Pressure and Mean Arterial Pressure from input port to output port such that there won’t be any need for prior Blood Pressure required. Only present Blood Pressure is enough.

DATA COLLECTION:
We are collecting the data from:
1. Thermistor : to acquire autonomous nerve stimulation by placing the thermistor in nose of the
subject.
2. ECG & PPG (Electrocardiogram and Pulse Plythesmogram) is acquired to calculate Pulse transit time. Pulse transit time is the time difference between R wave peak to end of the same Cardiac
cycle in the Pulse Plythesmogram.
3. HR (Heart Rate) Number of R peaks in one minute present in ECG waveform.
4. From BP apparatus Systolic, Diastolic and Mean arterial pressure has to be collected in
stipulated time intervals.

CLINICAL DATA:
1. Weight of the Subject
2. Height of the Subject
3. Pathology related to Kidney
4. Sex
5. Age

Our aim is to forecast Blood Pressure in 10/15/20/25/30 (CHOOSE ANY ONE TIME INTERVAL)minutes period interval during which the cuff is inflated for every 5 minutes. Usually, dialysis tenure for a single patient is 4 hours.

TRAINING:
During training the inputs for ANN (Artificial Neural Network) is given as shown in Figure 3 of page 4 in the attached document. Here, we have taken a 10 minute interval and for 2 hours duration.

Note: Please add Heart Rate (HR) and Thermistor Data in the input of the ANN during training.
Preferably we can go for BPN (Back Propogation Neural Network – Figure 3 in page 4 of the attachment).
The ANN is trained in SUPERVISED MANNER ie., the Systolic Pressure, Diastolic Pressure, and Mean Arterial Pressure is obtained and given as input in the output port of ANN.
Here, we consider inflation of cuff for every 5 minutes and forecasting for every ten minutes for 2 hours. This is an example only. We have to go for 5 minutes once inflation of the cuff and forecasting for every 30 minutes. The training of the ANN is done through the OFFLINE method.

TESTING:
DATA COLLECTION:
We are collecting the data from:
1 Thermistor : to acquire autonomous nerve stimulation by placing the thermistor in nose of the
subject.
2 ECG & PPG (Electrocardiogram and Pulse Plythesmogram) is acquired to calculate Pulse transit
time. Pulse transit time is the time difference between R wave peak to end of the same Cardiac
cycle in the Pulse Plythesmogram.
3 HR (Heart Rate) Number of R peaks in one minute present in ECG waveform.
4 From BP apparatus Systolic, Diastolic and Mean arterial pressure has to be collected in
stipulated time intervals.

CLINICAL DATA:
5 Weight of the Subject
6 Height of the Subject
7 Pathology related to Kidney
8 Sex
9 Age

Our aim is to forecast Blood Pressure in 10/15/20/25/30 minutes (CHOOSE ANY ONE TIME INTERVAL) period interval during which the cuff is inflated for every 5 minutes. Usually, dialysis tenure for a single patient is 4 hours.

TESTING:
During testing the inputs for ANN (Artificial Neural Network) is given as shown in Figure 4 of page 6 in the attached document. Here, we have taken a 10 minute interval (for forecasting) and for 2 hours duration. Here, we won't give any input (Systolic, Diastolic and Mean Arterial Pressure) to the output port of ANN . Now, when we give input to the input port of ANN the output port of ANN will give forecast value with respect to the time interval which we set.
Note: Please add Heart Rate (HR) and Thermistor Data in the input of the ANN during testing.
Preferably we can go for BPN (Back Propagation Neural Network – Figure 4 in page 6 of the attachment). Here, we consider inflation of cuff for every 5 minutes and forecasting for every ten minutes for 2 hours. This is an example only. We have to go for 5 minutes once, inflating the cuff and forecasting every 30 minutes for the period of 4 hours. The testing of the ANN is done through the ONLINE method. Our system will map both HYPER/HYPO TENSION and give forecasts for both the conditions. This is a unique feature.
, C , C , C , C , C , C , C , C , Claims:1) Our unique approach involves mapping the Systolic Pressure, Diastolic Pressure, and Mean Arterial Pressure directly from the input port to the output port, eliminating the necessity for prior Blood Pressure data. With this method, only the present Blood Pressure readings are sufficient for our system to function effectively.
2) In our procedure, we inflate the cuff every 5 minutes and make forecasts every 10 minutes for a duration of 2 hours. However, for the actual implementation, we will perform cuff inflation once every 5 minutes and conduct forecasting every 30 minutes, spanning a period of 4 hours.
3)The Artificial Neural Network (ANN) is tested using the ONLINE method. Our system possesses the distinctive capability to map and provide forecasts for both HYPER/HYPO TENSION, making it stand out with this unique feature.