This project started with the funding from the Institute of Tropical Medicine. Gabriel Carrasco, who was the principal investigator of this project, asked me to join, because I was working in some other projects related to public health. I was working with the Department of Engineering. He asked me if we could develop something to track a patient’s movement. He wanted to understand the relationship between population mobility and malaria transmission. They have been several studies using this strategy to understand and to see if there is a relationship between these factors and how to control malaria transmission. One of them was using GPS trackers to do that and to see how the behavior of these patients was as part of this study. The main challenge to do this study and to collect data from patients‘ mobilization was the device. In previous studies, researchers used a commercial GPS tracker, a small one. I don’t remember the brand of this tracker. It seems that the brand, that they mentioned in one of the studies, was the favorite for these kinds of studies. A proprietary system that everyone uses.

This device was adapted for this kind of study, because you could track and load the different data while you were moving. It was used more for other purposes and not necessarily to understand population mobility. They adapted and it seemed to work well. One problem was about the sign up. If you start to communicate with satellites from these devices, it’s different in urban areas than in rural areas. If you try to use it in the Amazon, for example, you have a problem with the weather and trees. You can lose data and you have to assume or predict the movement, for example, if the patient goes from one point to another one. There was a limitation for doing these kinds of research. The other problem is about the technical side. Because it was a commercial device and we had to import it to Peru. Further the budget, that we had at the time, wasn’t enough to do that. If we wanted to do other studies or maybe other people would like to reproduce it, it would be difficult because they need to get the same device to get the same quality of data.

The other limitations were about the technical features, the design and the weather. You don’t have the same behavior of these devices, when you are in rural or urban areas. You could have some variability in this data. Because of those limitations we started to search if there were other projects. In this case, we looked for open source projects. I joined the open source movement as I started building 3D printers by using all information from the RepRap project. That’s why I knew about this movement and all these different devices. I knew about the DIY movement and then how the open hardware was used for biology. That’s why Gabriel asked me if I knew something that would help us for this project. At the beginning, I didn’t know. We started to search some other projects, but at that time, we didn’t find anything for doing this kind of research. Most of the projects, that we found, were related to conservation, the study of animal behavior and animal mobility. There are some open source projects for tracking animals in general.

We found it challenging to replicate or to use the same hardware for our projects. When Gabriel contacted me, I had a RePhone, which are open source cell phones. They launched a crowdfunding campaign to back this project. After I knew about this project, I wanted to know if I could use it for some research. We decided to use the RePhone because you have a platform same as Arduino. In this case, you only have a small board. If you wanted to develop something related or any wearable device, that could be helpful. That’s why we decided to use this platform. You can get several kits with this platform. One of them is a GPS module that you could add to the RePhone. We did some tests at the University. It worked well. That’s why we decided it could be an option for doing these kinds of devices and could be suitable in this project to study a population or village.

Yes, I was a research associate at the Department of Engineering. They had the Health Innovation Lab, a leader in the startup at the time.

Partially, some of the fund was from the Institute of Tropical Medicine. Gabriel got a funding from that Institute. It wasn’t a small budget. He thought that money would be enough for doing this kind of research and to use it for a sample. When we wanted to publish the results, the reviewer told us that the sample wasn‘t enough for these kind of studies. We should prove why we use this sample for this study.

We found this hardware, the RePhone, and all these different models that allow to develop a wearable device for doing this tracking. We decided to use these devices because they are small and have the power like an Arduino. At that time, it wasn’t enough using an Arduino Nano. This device uses small, rechargeable batteries. We could develop like a small device that patients, in this case, could use daily. The idea of using this was to give it to the people as part of the project and they could use it in daily activities. We wondered how the mobility was in a specific period of time. That’s why it was important in this case. We wanted to understand and connect the data, that we got from these devices, to the malaria test, which we did to verify if they got this disease or not in this period of time. We needed this kind of data to see if there was any correlation.

Other students were involved in this project: Ober Castillo, Jose Saldaña, and Armando Caballero. They joined the team at the time. They were working with the RePhone previously for other applications. I led the hardware development, and the students contributed to programming or mechanical designs. In the paper related to the Gorgas Tracker, which was published in Frontier in the Public Health, they appear as co-authors.

After we decided which device or platform to use, we knew that we needed to get some user requirements in this case. We were working mostly with the researcher from the public health or the biological side. Most of the feedback and requirements came from them. We couldn’t ask patients because we were working in Lima and the patients were in the Amazon. We didn’t have enough money to travel to the Amazon because there were no other ways than taking an airplane. Maybe you can take a boat, but it’s not common to go there. All the feedback was from researchers of the research group studying malaria. We decided to use some design thinking workflow to get some iterations after one cycle to have something at the beginning, that could be helpful and then try to improve this device.

That’s why we shared the several designs of all the different versions of the device in the paper. At the beginning, we got a small device and we realized that we needed something bigger, because we needed a bigger battery. The patients needed to charge as less as possible. They didn’t have electricity all the time. In the Amazon, they have electricity during some hours. The electricity in general was a challenge for this project. We tried to improve the code to have a sleep mode of the device, while there weren‘t using it or traveling. In general, the design thinking method allowed us to get different versions of this device and something that could meet the user requirement, at least for researchers.Because we didn’t involve patients and we couldn’t travel to the Amazon to test the devices, we didn’t learn other things to include or improve. Again, the weather was challenging because of the humidity and temperature. We tried to add some other accessories to avoid the humidity. Further, we decided to include some cooling or ventilation.

We got at least five versions of this device. For example, we wanted to make a device where the case could be 3D printable. We found that could be easy to replicate and try to use at least some screws and nuts for this hardware for easy assembling.

Because of my network, in this case the open hardware community, it was easy to contact the Seeed Studio people. They helped me to get several kits of this RePhone. We got a discount for doing this research, because they were interested to see if this device or their platform would be useful for other applications.

They were interested if we could do something that is relevant for public health. That could be like a case use that they could share with the community about this platform. For the platform of the RePhone it wasn’t a big deal, maybe just to ensure good shipping. Customs are always a challenge because you have to prove that these devices are not weapons and met requirements of the Ministry of Telecommunication. That was difficult, because in Peru, the Ministry of Telecommunications and Transportation regulate devices. If you have these kind of mobile phones, they have to pass all these processes of bureaucracy. Doing that for a young researcher could be worse because you need contacts, money and sometimes a lawyer. We had to say to the customs that these devices were similar to Arduinos. We proved that was more for prototyping purposes than using as a mobile devices or mobile phones. That’s why we got these kits and we didn’t have any problems. A challenge was to get several rechargeable batteries, in this case the LiPo batteries. We didn’t know which supplier was good for this project. At the beginning, we tried to import the kit from China, it was not easy.

We did like 30 or more, I don‘t remember. We had to buy more than 30 batteries, because we didn’t know if they work or have problems in the field and because the researchers should have some batteries to change. Overall, the battery was a problem. It was difficult on one side to get the batteries and find suppliers and on the other side the functionality, some didn‘t work well.

Another bottleneck was the 3d printers that we had available. We had only two 3d printers, that were part of our department. Actually those 3d printers were mine. I built these 3d printers and we used them to fabricate all the pieces and the cases for the wearable device. To use 3d printers, in this case the FDM technology, is slow for some applications. In this case we tried to do several cases in a short time. The time was other other constraint. Within a short time we needed to get several 3d printed pieces to build all the pieces and to send them to the Peruvian Amazon.

RePhone is a GPS tracker with the main function to gather data. The device records the data and is connected to the satellite and gets the latitude and longitude. You can program through a user interface how this device records data and what it means. We wanted to know if the patients go outside of the community. We could define by using this interface and a radius of interest and relate this case to the community. After we got the position of the community, we could tell the device the radius. The device can detect when the patients goes outside the community and moves to another different part. Then we could program the time we want to record data. We have the active mode and the sleep mode. For example, we can record from 8am to 7pm. Afterwards, the device turns to the sleep mode to save energy and then the recording starts again at the next day at the same hour.

I read your paper related to maturity level. In terms of the general functionalities, I think we met the user requirements. However, in terms of documentation and general licenses, I think we still need more work. I don’t know which level of maturity it would be. At least, we got a functional prototype. We published this paper in Frontiers. There is another one with the same data, we got from this device, that was published in another paper this year. In terms of the quality of data we are accurate, at least for this kind of study. We compared it with some previous studies. I can say that we got a functional device, which could be used for other studies. However, there is still more work needed in terms of design. At least if you want to use it again for doing something in the Peruvian Amazon. If you want to do something in the Andes or near the Coast, I think you won’t have any problems. Because I think the weather was our main problem and other things related to the supply chain, because of the geography.

At the beginning we wanted to have the device, that could be useful for doing these kinds of research. In this research group, they were studying Malaria. To be open source was not relevant. After they knew about this project and the possibility to use open source principles, they understood that using these kinds of approaches could be useful in general for the field. The main outcome was to have a wearable device that could be a useful for studying human mobilization. The other thing is to publish the paper. As part of almost any research in academia, these results should be in a paper. We knew that we needed a paper. With that we wanted to prove to the research community that it’s possible to use these kinds of devices, develop your own device and to use it for these projects. It could be useful for other projects. Despite the paper was a requirement or maybe one of the outcome as part of this funding, we knew publishing a paper demonstrates that it’s possible to get good data with open hardware. This could be useful to prove other researchers and try to encourage other researchers to use these kinds of practices.

The idea was to see if this device with these functionalities could be used in other ways. One idea was to use this device, for example, in boats from these communities. We could record data about how traffic is throughout the rivers. We couldn’t do that because of money, but we started to focus on other topics. This was one idea and we hope to do that in the future. Maybe people could use these devices to record this data from boats, which is another way of using public transportation in the Amazon. Those were the main outcomes to publish. How was the relationship between human mobilizations and malaria transmission and to share an open source device that could be used for similar studies and other applications. Additional, we looked for other projects that could use the devices.

We published the paper in Frontiers in Public Health. We chose Frontiers, because it’s open access. As part of this paper, we share why we choose this platform, how we developed it, and what we try to translate from user needs to functionalities. As part of supplementary material, we shared some figures about the different versions and some picture of the user interface. We shared on the GitHub repository almost all the information related to that project. There we have the 3d printed files and a picture from the different versions. We got actually five versions of six versions of this device. We designed one that could be 3d printed without supports. There were other things, that I didn’t mention, about some other considerations within fabrication, that could be useful for anyone if they wanted to reproduce this device.

If successful means to get the outcome, to get data and to study what we wanted to study, I think it was successful, at least for the project. However, if we go to project management, it might not be successful, because it might take more time and money. In terms of resources, maybe it wasn’t successful. In terms of outcome, we got the data we wanted, at least for the number of patients that were part of the study. I think, the devices could have worked better if we could include other requirements related to the environment. There are other lessons learned regarding sustainability. I think that’s not successful. We realized, it’s difficult to recycle electronics in general in the Amazon. You could try to include sustainability as part of requirements, but you need some other a programs in the communities that can manage the thrash or components from these devices. After the project we took all the devices back to the city. We didn’t leave anything there. If you would give these devices to the people, that’s something you should think about. How is the sustainability of this project around these devices? I think it’s relevant for these communities. Otherwise, you would get a negative impact there. You could help and at the same time you leave something. These communities first need to implement a management of resources.

I wasn’t part of the research that was in the field. Gabriel knows more about that and the relationship between the devices and people in general. It’s not simple because you ask people to track their movements. It’s difficult in terms of privacy. However, when you contact these communities, you have to first talk with the head of the community. Some people that they consider as reference or someone that they can trust.

Afterwards, you explain the study. That’s how all these processes starts. You have several meetings with the participants. They have to sign a consent form for participating in this project. It’s not all about research and publishing a paper. The idea is to help with this data, for example, the Ministry of Health or another initiative that could help to control malaria in the Peruvian Amazon. They have vaccinations programs, where they go directly to the community and vaccinate people. But they don’t realize that the people move. If you vaccinate maybe a specific number and a specific time, maybe some of the members of this community are in other parts of the Amazon and could come with a malaria to their community again. Or if they decided to start this vaccination program in different parts of the Amazon, the Ministry of Health doesn’t know if controlling only vaccinated people in the communities is enough. They should try further strategies and maybe other sites to control malaria. In general, with this data you could develop these policies.

I’ve seen citations from the paper and I’ve seen one fork. But we don’t know if someone else has produced it. At least in the citations they mentioned our project because of the strategy of using open hardware or accessible devices for studying human mobilization. I haven’t seen another research group that has reproduced our device or did similar studies. We haven’t seen any publication where the others show the device for the same or other applications.

We decided to use GitHub mainly because I have used it before for other projects. It’s easier to share information and to try to create this brief documentation about the hardware. It’s one of the most known repositories in general.

We wanted people to see the STL files before they download the files and I don’t know if it was available at that time. I don’t remember to be honest. We were worried about visualizing STL files on GitHub. That’s something we wanted to do as part of documentation. That’s why we shared some pictures. If you see the repository, there are some pictures of the device and the different versions.

I was in charge of the whole process of 3d modeling and 3d printing. That was my main role. We used an end user platform, the RePhone, we only had to design the buttons that we use in this device because there didn’t exist any physical interface, but this was developed for others together by one of the students that was part of the project. My main role was to the design and development with other students all 3d printed pieces and advise the electronics. The students were the developers.

Yes, we used the geo kit for the device. The GPS tracker is based on the geo kit of RePhone.

We were four in hardware development and there were other four who were authors of the article that we published frontiers. There were two other people, Gabriel Carrasco and Edgar Manrique. They are biologists and work on malaria and in Public Health. We got feedback from them.

Gabriel was an associate researcher. He had a master’s degree and now he’s doing his PhD. Edgar had finished his undergrad studies. He was a technician and research assistant. There was another supervisor, Dr. Alejandro Llanos. He was a researcher in this group and he helped us to get some feedback.

The people from the Institute knew more about the environment, because they went to the Amazon for various projects. They knew about the different challenges the hardware could face. They recommended the strategy for designing the code with active and sleep mode. They knew more about this behavior than us. Their feedback was useful and relevant for developing the user interface. We had to develop this software and give them something that could be relevant and useful for doing the field research. We wanted to make it understandable for other researchers in this field, too. At least they could use these kinds of interfaces to get the feedback and hopefully their experience was useful.

We use some platforms to communicate between members. For documentation, we use the Google Suite to share any documents or to create any new document. We used emails and WhatsApp to communicate between us. In current projects, I mainly use slack.

To be honest, I can’t say an amount. We could pay part time for three students. With the funding we bought the components, filaments and other mechanical components. For manufacturing we didn’t pay anything because they were my 3d printers. I was involved more because I was interested to see at least one example of doing an open source device. I would say that my payment was this motivation, the results and the paper. That’s the way how I got a payment. The money was mainly for funding this part time position for the students and materials. I would say overall it was between $6,000 and $7,000. We estimated the costs for each device between 15 and 30 dollars.

The students applied the knowledge that they had gained from their undergraduate studies. They apply it to a real problem. Research in Peru is not expected to have an application of knowledge. They learned a lot from this process in terms of designing, how to apply and trying to translate a requirement to the functional prototype. They learned about malaria, and they understood its relevance to public health. They also benefited from this paper. If they want to continue their studies while applying for a master’s or a PhD program, it’s good to have a publication. For Gabriel, Edgar, the other authors, and I, it was good to know how these processes work for hardware development in Peru. Most of the projects import commercial hardware for research. In my case, it was my first time doing a hardware project for Public Health studies. My main benefit was learning about hardware development and using different approaches, like design thinking. Maybe design thinking has limitations, and you could use other approaches, or you have to iterate. It makes sense to use design thinking because it’s flexible enough to work with these applications.

Because of the project Health Innovation Lab was started. Gabriel and I decided to work and start this research group at the Institute, because we wanted to use open source principles or approaches for research. The other benefit was to demonstrate it’s possible to use open source principles in research and that we could do more things using these principles. Gabriel had the opportunity to start this research group at the institute. Now Health Innovation Lab uses some open access and open source principles, mostly in data science.

We hope to start some other hardware projects using the same approach that we used for the Gorgas Tracker project. Other benefits were the paper, the data for doing research and to know more about the hardware development.

The outcome of this project is that we saw a relationship between the human mobilization and the malaria transmission. We demonstrate with the data, that we got from this study, that people can get infected in a specific period of time and the infection probability and the level of exposure increases while they were moving from one point to another. Therefore, vaccination programs should try to develop different strategies to control malaria.