A frugal connected wearable to support caretakers in tending to premature babies #SDGclass2017

Similar projects worth following
Around the world, 1 baby out of 10 is born preterm, with about two third are born in low-income countries like India. Because premature babies encounter specific health challenges, complications after a preterm birth are the first cause of neonatal death. In 2012, the World Health Organization estimated that 3/4 of those neonatal deaths could be avoided even without access to neonatal intensive care.

Our goal is to develop a device to support parents and local health workers taking care of preterm children in the vulnerable neonatal period. It will be composed of two parts: monitoring and a companion device. The wearable monitoring device will determine essential data from different sensors such as temperature, heartbeat and oximetry. It will then relay the data to the companion device which will provide guidance to the parents. For example, if the temperature of the preemie is too low then the companion device will let the parent know that he/she must do the kangaroo care.

Why is it important to make a frugal monitoring device for premature babies?

Some figures

Goal 3 of the United Nations sustainable development goals is to ensure healthy lives and promote well-being for all at all ages. One of the targets for this is to end preventable deaths of newborns and children under 5 years of age by 2030. In particular, one aim is for all countries to reduce neonatal mortality to at least as low as 12 per 1,000 live births. In 2015, the UN Inter-agency Group for Child Mortality Estimation estimated an average number of 21 per 1000 neonatal deaths in developing regions. In comparison, developed regions have on average 3 neonatal deaths per 1000 live deaths. Overall, neonatal deaths account for 45% of under-five mortality worldwide. [Child Mortality Report (2015) ]

In 2010, complications of preterm birth were identified as the single largest direct cause of neonatal deaths, responsible for 35% of them. They are also the second most common cause of under-five deaths, after pneumonia. As such, both the reduction of preterm birth and the mortality reduction among babies born preterm are essential issues in the pursuit of the United Nations sustainable development goals. [Born too soon (2012) ]

Preterm births and possible complications

In 2012, a collective of organisations has published Born Too Soon , a global action report on preterm birth. While the figures cited need updating, the trends observed show that the described phenomena and problematics still stand.

Preterm births are defined according to gestational age, the measure of the duration of a pregnancy as the time elapsed since the woman's last normal menstrual period. Babies born alive before 37 weeks of gestational age are preterm. Below 32 weeks, babies are labelled very preterm (about 10% of all preterm births) and below 28 weeks, extremely preterm (about 5% of all preterm births).

Of the 15 million children born preterm each year, only one third have access to full intensive care or neonatal care units. The remaining babies are either born and cared for at home (5.6 million) or are born in health facilities that face limited space, staff and equipment. Very and extremely preterm children, of course, require the most care. Therefore, they have the highest mortality rate in low- and middle- income countries. However, even babies born after 32 weeks of pregnancy face specific challenges.

Preterm babies have higher risks of hypothermia, hypoglycemia, hypoxia, feeding difficulties, severe infections, respiratory issues, jaundice, brain injuries, retinopathy and anaemia. Some of those issues can be addressed easily with the proper training and low-cost material. For example, Kangaroo Mother Care, a skin-to-skin technique of newborn care, helps keep the baby warm enough. Overall, it was estimated in 2012 that 3 out of 4 neonatal death linked to preterm complications could be avoided even without access to neonatal intensive care services.

Existing baby monitors: reassure the parents

There are already a number of monitors for newborns on the market. Most of them are not specifically fitted to preterm babies' needs. They are also designed to reduce parents' fears and give them a feeling of control rather than addressing an actual medical need. Most work with a mobile tracking app and possibilities for integration of other smart devices. They keep track of a range of markers from the most useful like temperature to the most accessory like room light levels. In addition to this, their price range -minimum 100$- and format -some require constant new purchases as the baby grows- make it clear that they target high-income populations.

There is, however, one device worthy of attention in the context of preterm death in low- and middle-income countries: Bempu is a bracelet that constantly measures the body temperature of the wearer. It is intended for preterm babies and other newborns with complications. It does not rely on a mobile...

Read more »


General Graph of the process. File to be opened with

generalgraph - 4.68 kB - 08/09/2017 at 12:58



Decision tree for data interpretation [WORK IN PROGRESS]

Portable Network Graphics (PNG) - 58.83 kB - 08/09/2017 at 09:52



Show-case at the Cité des Sciences et de l'Industrie

Adobe Portable Document Format - 130.61 kB - 08/07/2017 at 10:16



Show-case at the Cité des Sciences et de l'Industrie

Adobe Portable Document Format - 101.88 kB - 08/07/2017 at 10:16



Show-case at the Cité des Sciences et de l'Industrie

Adobe Portable Document Format - 93.57 kB - 08/07/2017 at 10:16


View all 6 files

  • 1 × MAX30102 High-Sensitivity Pulse Oximeter and Heart-Rate Sensor for Wearable Health
  • 1 × MLX90614 Temperature Sensor
  • 2 × ESP8266 Used for communication between the monitoring and companion devices
  • 1 × Lithium Battery very small size

  • CRI and ZMQ

    Emma Barme01/19/2018 at 16:04 0 comments

    Just a short log to share excellent news: we have been discussing a collaboration with ZMQ, a "Technology for Development" social enterprise in India, for a few months now and we are officially kick-starting our working together on this project!

    I am personally very excited to be working with Subhi, Ahmad and others from their IoT team as they have both a wonderful framework for developing system-changing approaches to address critical problems of poor, rural, under-privileged and unreached communities, and the technical skills to carry this project out.

    If all goes well, we could have a full prototype by summer and start organizing tests with the future end users!

  • Simplified companion device

    Emma Barme01/19/2018 at 15:55 0 comments

    As stated in the previous log, we met with a design professor while in Shenzhen iSDG assembly and we left we new ideas for the design of the companion object.

    The main driving concern about our previous design, a cube-shaped speaker with symbols on the side that could light up, was that it was not obvious how it should be used. Part of the problem is that symbols, while more understandable across cultures that written words, are far from universal. For example, how can we be sure that the standard occidental heart shape will be recognized as indicative of a cardiac problem by parents in rural areas in India?

    We are now thinking of a button with a very simple shape, similar to that of a mushroom or a buzzer -see example below.

    The buzzer would hold a speaker to sound an audio warning and give oral advice, as well as LEDs of different colors so the device can light up with a color corresponding to the level of emergency of the situation.

    As the voice advice will be translated into local languages and adapted for each community, it would also be reasonable to adjust the emergency color coding to the standards of the community.

    We believe that this design has a better chance to be easily interacted with by people from very different backgrounds than the previous one, though it of course needs to be submitted to the feedback of users.

    All the electronic necessary to communicate with the sensors will also be hosted under the button.

    For now we have identified only one big technical challenge with the new design: for the sound to carry out of the device, there should be holes in the button. However, it needs to be resistant to humidity and rain, so we need to find a way to protect the electronics properly.

  • Shenzhen iSDG assembly and new insights

    Emma Barme08/31/2017 at 10:33 0 comments

    At the end of the CRI Labs Summer School, we were offered the opportunity to take part in the iSDG Shenzhen Assembly. The goal of this assembly was to gather students, researchers and makers involved in projects around the UN Sustainable Development Goals and "co-design innovative solutions for global sustainability".

    With four other projects from the CRI Labs Summer School, we presented PreemieAlert during the assembly. We received a lot of positive feedback as well as interesting suggestions and questions that are going to feed our continued development. We also had the chance to interact with several mentors from very diverse fields.

    In particular, a discussion with an experienced design professor led us to consider an entirely new design for our companion device. We will dedicated a specific log to this new idea, but it will roughly be a mushroom-shaped button with visual and audio feedback.

    We also participated in a small workshop on how to establish a business model and will work on ours with the canvas from

  • Layered companion device

    Emma Barme08/09/2017 at 14:37 0 comments

    Short update on the design of our companion device. It's composed of one layer of white plexiglass engraved with symbols, one layer of black paper with windows under the symbols, and two layers of wood to hold the LEDs, the speaker, and the wiring. Those layers are held together by a 3 printed plastic case (the one we have currently is slightly too small so we need to change the 3D model). The rest of the electronic (microchip, wifi, music shield, battery) fits in the case under the layers.

    All the blueprints for laser cutting the layers and 3D models for printing the case have been uploaded on our Github.

  • General Representation

    ccccccaaaaaaa08/09/2017 at 13:05 0 comments

    Schematic of the general process

    Schema of the process

    Here is a global view of how the device works. 

  • MAX30102-- how does it work?

    ccccccaaaaaaa08/08/2017 at 09:48 0 comments

    The MAX30102 uses a method called photoplethysmography to measure the heart rate of someone. This method shines light on the skin and the perfusion of the blood is measured. One of the practical aspects of this approach is that it is possible to differentiate between the light reflected by the blood of an artery (produces an AC output) and other components of the body such as bones and tissues (produces a DC output).  The photo-diode in the sensor then converts the light to current that we can use as comprehensible data. 

    To counter difficulties such as skin tone differences LEDs with different wavelengths are used. In the MAX30102 there is an extra green LED for this purpose. 

    Photoplethysmographic Sensors

    Another log will be posted later on to provide precision on how we calculate the SP02 and heart-rate from the AC output of the sensor. 

  • Exhibition in science museum

    Emma Barme08/07/2017 at 10:56 0 comments

    We had a wonderful time on Friday at the Cité des Sciences et de l'Industrie, one of Paris's science museums. We presented our project among the others of the CRI Labs Summer School to invited scientists and curious families.

    We received heart-warming compliments and encouragements, both on the project itself and on our presentation.

    Our posters from the exhibition have been uploaded here in the files.

    We were able to show a working infrared temperature sensor. This allowed us to detect a problem: when the heat source stays on the sensor for too long the measurement becomes very imprecise.
    We also showed the various steps of our prototyping and the electronic components we used.

  • Heart Beat Sensor Update

    ccccccaaaaaaa08/07/2017 at 09:14 0 comments

    This is a follow-up on the previous post on the heart beat sensor MAX30102. The problem encountered previously was the death of the sensor with no explanations. We tested out a new sensor on the code and determined two possible causes. 

    Firstly, we based ourselves on code online which uses an Arduino with a 3.6V input. After reading the specifications we observed that the sensor can be submitted to a voltage of min 1.7V and max 2V with a suggested 1.8V input. 

    Secondly, the sensor should be tailored to the different skin and the location where we put the sensor. The intensity of the light emitted should be changed in order to find the best possible results. This needs to be experimented on to find the best solution. 

  • Working temperature sensor!

    Emma Barme07/31/2017 at 09:10 0 comments

    We have received and wired our MLX90614 Temp Sensor.

    We used the Adafruit library and you can find our code on our GitHub.

  • Rough prototype of foot bracelet with sensors

    Emma Barme07/31/2017 at 08:50 0 comments

    Taking into account the size of the various electronic components we have to include in the wearable, and the constraints associated with the sensors, we have decided to focus on a foot wearable. The sensors will be placed on the top and bottom of the foot thanks to a textile sock/bracelet part. The rest of the components (battery, wifi and micro chip) will be held in an ankle or thigh bracelet.

    Because we have not yet settled on the exact battery, wifi module and micro chip we want to use, we focussed first on the sensors part of our wearable.

    Below are photos of our first, very rough prototype. For each sensor, we are using a made-to-fit 3D-printed plastic case. Right now it is hard plastic but ideally, it would be made of medical-grade silicon.

    A window of the size of the case is cut into a thick slightly elastic fabric A. The case is placed in this window. Both the thick textile A and the case are sewn onto a thinner and more elastic textile B. [See diagram below, top is the view from the inside and bottom is a longitudinal cut]

    We placed the temperature sensor above the foot because it was slightly smaller and did not require contact, and the heart-rate and SPO2 sensor below it on the flatter sole part. We added a short elastic strap around the ankle to maintain the whole thing in place.

View all 14 project logs

Enjoy this project?



Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates