Deep Prenatal Care Unit

Project Title

Deep Prenatal Care Unit

Project Area of Specialization Biomedical EngineeringProject Summary

Every 11 seconds a pregnant woman or her baby dies due to some sort of disorder that could have been prevented if addressed sooner. Awaiting a newborn baby is a blissful yet a worrying period for a mother and she’d do anything to ensure the safety of her to be born baby. During gestation period, a woman’s body goes through various physiological changes which if overlooked might lead to complications. Pregnancy is a delicate situation and therefore, needs special attention from medical experts. But certain changes in a woman’s body remain unnoticed by both medical practitioners and the pregnant women until these changes aggravate to levels where an emergency situation arises. When an emergency situation is created the health of pregnant woman gets complicated and difficult to normalize and, in such a critical situation, both life of mother and the child are at stake. According to our survey woman is recommended to visit her doctor at least 20 times during her 9 months period for regular checkup, apart from unexpected visits which becomes problematic for woman who live in remote areas and can’t afford travelling to cities every other week. Furthermore, pandemic like current has raised the health risk to the mother and her baby caused by exposure to virus by visiting hospitals or leaving house in case of regular checkup or false contractions. Therefore, to make this process feasible for both mother and the doctor, we propose a low-cost non-invasive AI enabled tele-monitoring device which would continuously monitor pregnant woman remotely and alert doctor only in case of emergency. Our device would not only continuously remote-monitor the physiological parameters of the woman but, with the help of machine learning, would simultaneously use that information to make real-time predictions about abnormality in other basic physiological parameters like Sugar level and Blood pressurue and any other severe complications. The results will then be provided to the doctor on a user-friendly application. This feature will enable the doctor to make required arrangements and take measures beforehand to prevent complications. Our device consists of three major subsystems which are DPCU, cloud computing, and smartphone application. DPCU will acquire the required data i.e. ECG and then it is transmitted to a cloud where it'll fed to deep learning models to make predictions which will be displayed on user's smartphone application.

Project Objectives

Develop a remote health-monitoring device for pregnant woman in their third trimester to ensure their safety by making future predictions about the possible complications through data analysis such that it reduces the unnecessary prenatal visits and the rising mortality rate.

Project Implementation Method

Project implementation can be divided into 2 parts as:

1) Hardware implementation: In hardware we have two separate modules one is responsible to acquire ECG signal of mother and the other acquires abdominal ECG. It is very challenging to acquire ECG signals because they are in milivolts and are suppressed by other noises so we had tried many instrumentation ICs and amplifiers and different combinations of filters but we received satisfying results with AD8232 IC. One the ECG was acquired then we had to amplify it so that it meets the minimum threshold of Arduino's reading port. Once the signal is amplified then it goes through highpass, lowpass, notch and butterworth filter which results in a smooth and noise-free ECG signal.

2) Software implementation: The ECG signal is then transmitted through Wifi to cloud where all the computation happens. Firstly, the abdominal signal is decomposed into its constituents and Fetal ECG is identified and further filtered through digital signal processing techniques. Then heart rate is computed of the filtered fetal and maternal ECG. Moreover, these ECGs will be used to classify as normal or abnormal ECGs and predictions about other fundamental physiological parameters like sugar level and blood pressure will be made using Computer Vision techniques.

Benefits of the Project

This will mainly benefit pregnant woman in remote areas who can't travel to cities for their checkup and can't afford staying in any big city. This will also benefit working class woman who live in cities like Karachi where there is a huge problem of traffic and therefore, this device can help them be more independent and confident as they and their baby are being monitored by experts. The ability of the device to make early predictions will help them manage their time accordingly. This same technology could later also be used by anyone who has heart problems and would also want their other physiological parameters being monitored without having to wear huge machinaries and equipments they can easily carry our portable device.

Technical Details of Final Deliverable

Our final product will look like a strappable/patch device that can easily be attached and dettached by the subject and she can carry it around easily. At the moment, we intend to have 2 separate circuits one to monitor mother's well being and the other to monitor her to-be-born baby's. It'll be connected to a cloud server where all the computation will take place and AI models will be deployed which will make predictions in real-time. All the informations will then be displayed on a user-friendly mobile application.

Final Deliverable of the Project HW/SW integrated systemCore Industry HealthOther Industries Medical , Telecommunication Core Technology Artificial Intelligence(AI)Other Technologies Internet of Things (IoT), Cloud Infrastructure, Wearables and Implantables, Big DataSustainable Development Goals Good Health and Well-Being for People, Decent Work and Economic Growth, Reduced InequalityRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	50000
ECG electrode cable	Equipment	1	7500	7500
Arduino Due	Equipment	1	2500	2500
GSM modules	Equipment	2	1500	3000
PTA registration	Equipment	1	4000	4000
Veroboard	Equipment	4	250	1000
Arduino UNO	Equipment	4	500	2000
Resistor sheet	Equipment	2	250	500
Capacitor Sheet	Equipment	2	250	500
PCB printing	Equipment	5	2000	10000
AD8232 module	Equipment	1	1500	1500
reusable Electrodes packet	Equipment	1	500	500
Wifi Module	Equipment	2	500	1000
ICs	Equipment	5	1000	5000
db15 pcb mount	Equipment	4	250	1000
soldering iron	Miscellaneous	2	2500	5000
overhead cost	Miscellaneous	1	5000	5000