@sunbiz@shbucher I spoke to some doctors and people involved in the medical profession. They said that urine output is also an important parameter. We track that as well? I specifically asked this as it might need manual entry by hospital staff and not from sensors.
I have been trying to make a KMC device based on an ESP board in the college lab and has gone well so far. The board and technology involved can be different but it can do similar things based on the description in the IEEE research papers I read. Can you please guide me about the data schema that is collected by the device?
@sunbiz@shbucher I have made a sample app that fetches the data related to battery levels, heart rate, and temperature from a BLE-enabled peripheral device but when we are talking about NeoRoo App there are Three important features that are to be integrated in-app: -
Sharing neonatal vital signs and related operations in real-time is straightforward when the app is visible but
the task is we need to track this data even when the NeoRoo app is not visible means we need to make the NeoRoo app always run in the background which is a high power consumption task.
As soon as the app goes in the foreground or dispose of activity lifecycle we need to run some set of services in the background which fetch the data from the KMC device and sync with the NeoRoo app and Cloud storage and show local notifications alerts based on the KMC device characteristics value.
This is an important task when we are working with this kind of use case.
So what I have achieved till now
Able to scan and establish a connection between the client and the peripheral device.
Able to read data from the peripheral device in realtime and show notification alerts with given conditions
Able to store peripheral device data periodically to the cloud storage and once the device is disconnected to the peripheral device app automatically syncs the latest data from the cloud.
I am currently figuring out a solution how we can get data from the peripheral device to the app even when our app is not visible in flutter
Is BLE the only protocol that the KMC device uses? I dont think it is extremely reliable to push the data after connecting to the phone. Is there another protocol? Can Gsoc contenders work on the KMC device side as well?
@Hrishikesh-Bhagwat as far as I understand this project I don’t think there is any work related to the KMC device but this is an open-source project if you have any idea related to the KMC device communities are always welcomed proposals for it.
There can be a project idea related to embedding flutter in KMC devices but that will be an altogether different project.
Yep. I know it is not about the KMC device. The device I am making must not have more or less features than original device which seems likely. If I add more features there will be a faulty plan of action.
Infact I already have made a device like that during a past internship. I cant access it as its proprietary now.
This question complements your question of what to do when device is not connected to KMC device @adarsh-technocrat .
Another communication protocol other than BLE will be needed when mobile is far away from the baby. Like the doctor is at home and the baby is in the hospital. In this case the KMC device should publish data to DHIS right?
I sincerely apologise for continuosly asking questions. I have been brainstorming for a while and am curious and excited.
Understood, it would be better @Hrishikesh-Bhagwat if mentors can share jitsi link here as well for communication it would be better if we can discuss this thing there, there will be more clarity
this is the beauty of opensource @Hrishikesh-Bhagwat people don’t offend when you ask more questions thus it has to be relevant and you are doing a great job, its good to ask questions
Some examples of goals that can be set and tracked (this list is not exhaustive – there can be other goals/objectives/metrics) include:
Number of hours of skin-to-skin care/kangaroo mother care (by day, week, month)
Metrics related to feeding – such as, number of times baby breastfeeds per day; number of times baby requires supplementation with cup feeding; volume of feeds delivered via NG tube
Number (and type) of neonatal care educational sessions attended by family members during hospitalization
Tracking of metrics related to the baby’s vital signs status. For example, proportion of time (over shift or day or week) baby’s body temperature remains within the normal zone; number of episodes of apnea)
Hello – For this phase of the project, assume that the device + app are being used within a health facility. That is, not in the community or home setting. That being said, there are often very large teaching and referral hospitals, and it is true that the device and mobile phones (especially of the providers) may not always be right next to one another, thereby, limiting the effectiveness of bluetooth. Thus, exploring and proposing other data stream options to incorporate into your proposals is highly encouraged.
For purposes of this use case, at this particular time, we are not collecting information about urine output. There is no way to do that via the integrated technology that is currently within the sensorized KMC device. The sensors embedded within the device can, at the current time, gather vital signs information regarding heart rate, breathing (rate and pattern), body temperature, and blood oxygen saturation. Those are the vital signs that are automated.
To collect information regarding urine output, or feeding volume/modality (see another comment in this thread), these data would have to be entered MANUALLY by the parents/healthcare providers. For urine output, probably the most PRAGMATIC indicator, for the current use case, would be to enter the number of times the baby had his/her nappy changed within each 24 hour period, and whether there was pee/poo/both for each diaper change. However, please note – for your proposal to GSoC, @sunbiz and I are not currently stressing the need for information related to urine output or feeding volume. These are “nice to have” features, not “must have” features, for this phase of the project.
Will the data for the goals be automatically collected? Most of the programmers do not possess domain knowledge of the medical field so it is difficult to think of goals independently. Will this information be provided.
@shbucher it was said earlier that the device is patented and developed outside the scope of the project. Since the app is the main project, unless we are able to modify the device itself in terms of hardware and software, a new communication protocol cannot be set up. I have seen projects using MQTT(pub-sub) for remote tracking. However any new protocol will require us to change the source code of the KMC device.
Well I have a solution to this in mind. We can create a middleman device thats constantly next to the KMC device. It will collect and push data to the client smartphones via websocket or MQTT. However, it will need some work on the embedded systems side.
I own the patents to the device – the purpose of this is because I wish for the device to remain low-cost, and affordable within the LMIC setting. This is one of the primary reasons we are developing open-source, rather than proprietary apps, to link with this device. It is another key reason why we wish for this app to be built within the architecture of the DHIS2 platform, and why we are incubating these efforts within the LibreHealth ecosystem. The overarching objective of the initiative is to save newborn lives, with mHealth solutions that are affordable, safe, effective, feasible, and acceptable within the low/middle-income country setting. Thus, we are interested in the most effective functional solutions for NeoRoo (apps). Develop the best possible proposal for the open-source apps, according to the specifications provided by @sunbiz and myself for the 2022 GSoC initiative, without worrying too much about the patents on the hardware (KMC device).
