What this blog is about :-)

Posted: 19. August 2014 in Allgemein

myrijam-01Hello everyone!!

At first I’ll tell you who I am and why do I guide this page…

My name is¬†Myrijam Stoetzer, I’m 16 years old and I’m in the 10th class… ūüėČ

DSC06110I want to show you all my technical projects, I did since the 3rd class. So I won’t forget what I did as yet…

Here is my e-mail adress to contact me: myrijam.stoetzer@gmail.com


My friend Finja and me developed an optical procedure which analyses the video of a sleeping baby in order to recognize if the baby’s breathing is stable. If that is not the case for a few seconds, there will be an alarm that allows the parents to get help or to apply first aid. At first we tried our algorithm with a video of a real baby sleeping, which worked fine – the motion of the lung and stomach while breathing were detected. But of course we do not have a video or a real-life situation in which we could test out our program to detect a breathing arrest – and we didn’t want to edit the video to simulate that either.

Instead, we used a Raspberry Pi and a PiCam to simulate the situation with a puppet. A ballon and a syringe pump allowed small breathing movements as a breathing arrest as well.

The video-frames go through a differential-analysis in which two pictures get subtracted from each other so that the change (motion of the baby) gets visible. It also is filtered by a dilate-filter that makes the pixels bigger so that contours can be detected much easier. Then it is just a matter of checking for a contour with corresponding size to check for stable breathing.

This is another Jugend forsch project and we won the first prize with it on regional and the second prize on national level!!!

Hackaday Prize!!!

Posted: 13. September 2017 in Allgemein

We submitted to the hackaday prize 2017 in the assistive technology section Рand guess what happened yesterday: Our project was peer-reviewed and now we are finalists!

This is so amazing – we did not expect that at all. It gives a huge flush of enthusiasm to continue, even if school is stressing us now ūüėČ

If you have any ideas how we can improve our work, make it easier for people to access or more powerful in terms of diagnostic features, we are very happy about any comments or suggestions!

4th prize!!!!

Posted: 1. June 2017 in Allgemein

The german final was great I met a lot of people I know from the last years and a lot of new people! It was so much fun and there were a lot of interesting projects! And we won a 4th prize in the category “Arbeitswelt”!! It was a great experience and I hope to see all of the people next year!

German Final!!

Posted: 12. May 2017 in Allgemein

The regional¬†competition of “Jugend forscht” (German STEM competition) is now one month ago and we won the first prize with our vein detection project and Finja and I have won the third prize with MAG3D!!! We are so excited to participate in the Jugend forscht final again!

Till then we for sure want to improve our project! We want to make the project more practical and simple to use.
We are about to make it possible to see the videostream on the smartphone. We use a simple mjpeg streaming solution and try to make a step-by-step guide for people to reproduce. And, of course, it will be published under CC license!


Additionally we compared our project to a professional vein detection system at the coagulation center in Duisburg. This system costs about 4.000‚ā¨ while ours would cost about 100‚ā¨ ūüôā And the quality is very comparable – so in the end our system could be used to encourage patients. Having high end priced systems,no one has thought about this before – giving a product like ours to people who need to inject medicine to their veins, simply because of the tremendous costs of professional devices…
In the picture you can see the¬†professional one, it projects a red laser image¬†and where the veins are located, it simpy does not display anything (no light = dark – there are your veins!) But with our project the veins are¬†at least as good to see – and with the possibility to improve the project yourself, you could even add a feature to stream the process of vein puncture to your¬†physician, if you would like to have immediate feedback ūüėČ

Here is a link to a guidance how to rebuild our assistive vein detector: https://github.com/Myrijam/Venenfinder

MAG3D – Update!

Posted: 5. March 2017 in Allgemein

Finja and I developed our scanner for vectorfield analysis of magnetic flux further:¬†We have built a new design with a linear 3-axis delta approach – now we can truly scan a 3d-space mapping the magnetic flux as a 6dimensional vectorfield! And we have added some fine mathematics to cheat – eh, improve performance and/or resolution ūüėČ


Design of the Linear Delta for moving Sensorplattform

I am still not satisfied with this design because the ratio of the arms and the radius is not optimal and therefore you can only measure a very small area of 3x3cm because the sensor plate is raised at the edges of the measuring surface. I know this could be simulated e.g. with Geogebra, but I have no clue at the moment how to do simulations in this software…

New scans – examples

Here is a vectorfield plot of a round magnet from above:


And, of course, we have taken the traditional electric-current-in-a-wire magnetfic field to the next level: How do strange or funny shaped-wires produce magnetic fields? Have a look:

heart_vectorfield heart_surface

It is the shape of a heard with about 10amps going through it – but apart from the “girlish” attitude of showing a heart, there is real-world application for it. Think about the complex magnetic field necessary in fusion reactors… they need odd-shaped coils to produce them ūüėČ

Constructing the linear delta

Coming back to the new mechanical design we have chosen: Delta printers have a parallel kinematics, since the positions of the motors are not interdependent, so one motor does not move the other. On the other hand, the positioning becomes more complex, because, in the case of a positional change in one plane, all three motors must always run at different speeds but start and end at the same time. The mathematics behind the delta design, for the precise control of the stepping motors, is derived from the theorem of Pythagoras itself Рignoring acceleration and deceleration. We 3d-printed the carriages and motor attachments and first used an Arduino Mega, 2 motorshields Bluetooth-Dongle and a smartphone to control the scanner. In the meantime we have started on porting this to the Raspberry Pi using Python and Matplotlib to visualize the scans.

How to cheat nicely with math ūüėČ

In the case of different measurements it is noticeable that the visualization of the vector field can show “jumps”. This is why we asked ourselves which mathematical algorithms we can use to improve the measurement results, since we cannot change the sensor itself to reduce noise.

One way to increase the measuring accuracy or to reduce the “noise” is to carry out several measurements and calculate¬†the average. The sensor itself allows an average value formation and transmits to the micro-controller the average formed by 2-fold, 4-fold or 8-fold measurements. We have also implemented this accordingly and have the sensor now always perform an 8-fold measurement at each individual measuring point. This way only the mean variation at a given point is reduced.

Nevertheless, a clear noise remains in the further pictures. We have therefore researched how noise can be reduced in a vector field and found a contribution in a scientific¬†journal describing a corresponding filter for geophysical investigations: a so-called vector-median filter. Median is exactly in the middle of an ascending sorting series. MATLAB has the possibility to calculate a vector median filter. The specific characteristics of the field are preserved. We’ll combine that with another mathematical improvement:

A further mathematical optimization in measurement is the interpolation of data. Some of the measurements take quite a long time when a large measuring field is set Рbecause a doubling of the resolution or the measuring field on two axes leads to a fourfold higher data volume and scan duration (quadratic relationship). If this is transferred to the z-axis, the time for an overall measurement is doubled (cubic correlation). Since the individual measuring points lie at a very small distance from each other (a few mm), it is unlikely that there is an extreme deviation in the magnetic field between them not affecting the neighboring points. So Рcan we use an algorithm to do educated guesses for sup-pixels of our analysis? Or in other words: Can we cheat nicely with math?

In the simplest case, the intermediate values could be found by means of a linear function, that is to say the values would be connected to a straight line. More complex ways would be to find a curve that fits the data to estimate the “in-between” values not measured.

In MATLAB there is a function that mathematically calculates approximate values between individual measurement results: The spline function. Starting from the neighboring points (for example, in a 3×3 or 5×5 grid), it reconstructs a graphical course on which these measured values lie by different methods – linearly or by curve trajectory. Have a look:


This is a zoom into a surface plot – on the left is the original data, middle shows linear interpolation and right image shows the spline function. The noise is preserved, but we already have a solution for it…

This function can be used to measure fields faster or with higher resolution. If, for example, a measurement is carried out only once every 5 steps in a 100×100 field, instead of measuring every single step, the measurement process can be 25 times faster. From the upper graph, however, it can also be seen that the measured values still show noise¬†point despite the mean value formation. Just interpolating the data carries on¬†these inaccuracies so that it can be useful to perform a smoothing of the data:


This comparison shows that the same magnetic field. In the upper left you see a zoom into a surface plot of a 20×20 field. The downer left shows the exact same magnetic field scanned with a 25times higher physical resolution of 100×100 samples.

The images on the right side show the mathematical approximations: The upper right image is interpolated from 20×20 to 100×100, increasing the physical resolution to a virtual one by factor 25 as well as an applied vector median filter. It still shows some of the noise, but it is very smooth and accurate now. The image on the down right side is a median filter on the pure 100×100 physical resolution – no noise, no artifacts.

As our aim was to keep the project as open as possible, so we looked for alternatives to visualize the data. We still work with MATLAB, but have an alternative version with the Raspberry Pi Рthere is Wolfram Alfa (which is for free on the Raspberry) as well as powerful tools like Matplotlib..

We submitted this project this year to Jugend Forscht in Berlin and won first prize on Berlin’s regional level. Keep your fingers crossed for the next competition level, please ūüėČ


Medication that cant be ingested through the gastrointenstinal tract has to be injected intravenous Рusually by a doctor but in chronic diseases it can also be
carried out by yourself ¬†(“home treatment”, for example, venous puncture with
medication 3 times per week).

A chronic disease which requires frequent venous puncture is hemophilia.
In the blood clotting chain, one or more essential enzymes are missing or build in a non-working way due to DNA corruption. This can cause severe bleeding into joints, muscles or inner organs and is potentially live threatening if the rest activity of the clotting factors is below a few percent. Today, most of the clotting factors can be artificially synthesized through biomedical engineering (cell cultures with a changed DNA  produce them), but they still have to be 
administered externally Рinjected into the blood stream by venous puncture. Children and families often face great difficulties here.
Venous blood vessels used for medication are not always easy to recognize and if the needle is not secure in the vein, it must be punctured again at another point.

Here our project comes into play
: Using NIR (near infrared) Illumination and real-time image processing, we can make the veins more visible!


Screenshot: upper left original camera, upper right automatic image adjustment, manual adjustment, colormap

We (my brother Elias,¬†Lucie [a friend of mine] and me) submitted this project to Jugend Forscht and won first prize on Berlin’s regional level. Please keep your fingers crossed for the next level on the 21st and 22nd of March…

Our aim was to develop an assistive technology for venous puncture for diagnostic purposes (blood sampling) or for medical drug administration.
Easy reproduction and low costs are important criteria for development.

In addition to the research and review of specialist literature on methods of venous localization and optical characteristics of human skin, we carry out our own experiments: 

Illumination of skin with different wavelengths, cooling the arm and using thermography to watch the surface veins warming the skin again as well as a spektrometrical analysis of arterious and venous blood samples were among them.

With the results and findings, we developed two¬†prototypes for computer-assisted venous localization!¬†They¬†differ regarding¬†to the camera system used – one works¬†with the PiCAM, the other with a modified webcam. Both have their own pros and cons…


Frame for Webcam

Usually, cameras have an infrared blocking filter, which only allows the visible range of the electromagnetic spectrum to pass, because otherwise the image representation deviates strongly from our viewing habits. For the Raspberry Pi, there is a camera, where such a filter was not build in. It can be used without further modifications. However, this camera offers only a fixed focus and cannot focus the image scene automatically.


Modding Webcam – remove IR-Blocking Filter

Another possibility is the modification¬†of a webcam. The housing has to¬†be opened, the blocking filter removed and as an additional lighting, 1mm small infrared LEDs need to be¬†soldered in; along with some pieces of old analog developed film (blocking visible light, passing NIR). We have been able to use such a camera from my previous research project (“eye control wheelchair”).

We use the single board computer Raspberry Pi (will work from from version 2 on), which is available at low cost worldwide and has proven itself in many projects as sufficient for image analysis and image processing. Using¬†3D printing, all required housings for the Raspberry Pi, the cameras the infrared illumination and the display could be designed and changed easily ūüôā And you can build your own version by just downloading the code and the cad-files from my github…


Discussing the Project


Dr. Klamroth testing our Prototype

We presented our project to Dr. Klamroth, head of department of internal medicine at Vivantes Berlin Friedrichshain, who was impressed and gave us some tipps to improve the usability.

A possible future development would be the use of smartphones with¬†an external camera that can record infrared light – but until now I haven’t had success in installing Android Studio and working with openCV yet…

You can download the full report here (in German):Veindetection_German



In the autumn holidays Finja and I were invited to the Fraunhofer Institut IIS in Erlangen. They have developed the HallinSight, a sensor similar to our magnet field scanner, but they use a sensor array instead, which means they don‚Äôt have to move the sensor. And the most amazing thing about their sensor is that they can receive the measurements in real-time (up to 200 Hz) ‚Äď so it is a true magnetic field camera! It uses a sensor array of 16×16 sensors that measure the magnetic flux as a 3d vector field ‚Äď but instantly. And with some very sophisticated math they can interpolate the readings to sub-pixel precision! Unfortunately, this sensor is not used in schools so far ‚Äď just imagine visualizing the magnetic flux of a motor on action!

Our Visit started with an overview of the institute and the applications they develop sensors for. It was very interesting to hear about the different working places. They have a lot of amazing projects starting from facial expressions detections to an intelligent toothbrush! This is so cool!

Then we presented our project in front of the entire team that developed the HallinSight and did a few measurements ‚Äď they were very interested in our development and the decisions in design and software architecture we did during the development. And we got a lot of tips and ideas as well!


After that we had a very delicious lunch in their canteen, followed by a tour through the Institute. At first we got to know about the audio technology, because the Fraunhofer Institute is known for inventing the mp3 compression.

We were at their home cinema and it was in 3d! That was so cool to watch short films. Unfortunately the cinema is only for research ūüėČ

At the end of the tour they told us that they have a surprise for us and we were super nervous and excited!

It was like Christmas ‚Äď or even better: Dr. Hohe und Dr. Peters gave us two of their magnetic field cameras as a loan (respectively our schools in Duisburg and Berlin, since we are not attending the same school anymore because my family and I moved to Berlin). We were shocked, because we never would have expected that and it is such an honor to us!

I will post some measurements from our MAG3D and the HallinSight here in the future so you can compare both.