What happens if you have a 1D sensor capturing data, with a moving object? Like the Slit-scan cameras, which give a surreal feeling to portraits for example:

Well, you can also do this with ultrasounds =)

TLDR: In our project, communication and documentation are closely linked, and have a foremost place in our OSH. I'm detailing in the link how to document only in one place using markdown, and generate reports, docs, graphs, block diagrams and presentations on the fly.

I'm so lazy about documenting, that I want to do it clean, and efficient - so I have prepared a lazy documentation about it there: http://kelu124.github.io/echomods/DocumentationProcess.html

Hey community, a bit late, but ... KUDOS TO YOU! Thanks to you, we got accepted in the semifinalist list of hackadayprize!

To celebrate, let's share what we did with a probe found at 70$ on EBAY, which was supposed to be broken... well, that's what the vendor said! After opening, rewiring some cables, we found out we could get echoes:
Well, with the modules underway, and the existing Murgen board, we can now stopping worrying about the continuous rotation of the head and the noise it gives.

Sweet! And a great time opening and repairing:

Just to be curious: what would you use to convert the digitalized array into a geometrically correct shape?

How did we take these images?

Fantoms and the testing bowl	Distance to the fantom	Specs of the fantom

Comparing the image with the fantoms

Image	Specs of the fantom

Getting some images

Image	Time Stamp	Nb Lines	Nb Of Points Per Line	Spls. Freq.	Angle	Comments
	20160401 - 202119	120	991	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160401 - 202119
	20160328-142043	120	966	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160328-142043
	20160328-141354	120	988	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160328-141354
	20160328-142216	120	991	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160328-142216
	20160328-142100	120	991	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160328-142100
	20160401 - 202746	120	1008	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160401 - 202746
	20160328-141800	120	994	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160328-141800
	20160328-141312	120	998	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160328-141312
	20160401 - 201835	120	994	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160401 - 201835
	20160328-141521	120	998	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160328-141521
	20160401 - 202233	120	988	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160401 - 202233
	20160401 - 201601	120	987	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160401 - 201601
	20160401 - 202350	120	988	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160401 - 202350
	20160328-141136	120	997	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160328-141136
	20160328-142322	120	998	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160328-142322
	20160401 - 202638	120	994	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160401 - 202638
	20160401 - 201923	120	991	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160401 - 201923
	20160328-141815	120	994	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160328-141815
	20160401 - 203237	120	980	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160401 - 203237
	20160328-142200	120	987	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160328-142200
	20160401 - 201655	120	994	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160401 - 201655
	20160328-141152	120	988	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160328-141152
	20160328-142338	120	976	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160328-142338
	20160401 - 203127	120	990	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160401 - 203127
	20160328-142026	120	988	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160328-142026
	20160401 - 201413	120	1001	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160401 - 201413
	20160401 - 202033	120	988	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160401 - 202033
	20160328-141539	120	988	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160328-141539
	20160328-142230	120	990	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160328-142230
	20160401 - 202809	120	970	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160401 - 202809
	20160328-141840	120	997	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160328-141840
	20160328-141334	120	988	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160328-141334
	20160401 - 202429	120	991	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160401 - 202429
	20160328-141651	120	998	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160328-141651
	20160328-141207	120	988	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160328-141207
	20160328-141554	120	986	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160328-141554
	20160328-142144	120	998	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160328-142144
	20160328-141223	120	991	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160328-141223
	20160328-141119	120	994	5MHz	60	File coming from the examples.tar.bz2 -- ./old/20160328-141119

Original repo of images at : https://github.com/kelu124/murgen-dev-kit/blob/master/software/examples/Readme.md

We had sort of a debate regarding the speed of the ADCs we are using. Since we're using 3.5MHz transducers, we would have been aiming at getting 7Msps minimum ADCs. However, we're using non-perfect transducers, so pulses are often quite large (2 to 5 periods), and we're looking at the envelope of the echoes. So, the echoes envelope have a width that spans around 1ms - a 1Msps ADC could be able to capture this.
Moreover, we discussed about introducing a peak-hold-reset to play the role of the envelope detector.
Since we wanted to see the impact of the ADC's speed, we simulated a decimation from 5Msps (the original data) down to 2.5Msps and 1Msps. For this decimation, we had three choices: averaging the points, getting a random one, or getting the highest point (emulating the peak-hold).
No more to say, but this, on data before the scan conversion:

Unsurprisingly, the random is less smooth than averaged, and the Max has quite an interesting answer, as it highlights the features of the image. Rest of the description (including scan converted images) is here.

All code, images and the scripts are available at : https://github.com/kelu124/murgen-dev-kit/tree/master/worklog/Images/Session_8/source_files

Exciting as it has been, the murgen-dev-kit project (GitHub repo) will be stabilized in the coming week. There's still some doc to fine tune, but, apart from that, the objectives are all reached:

• Getting a medical-grade image? Check
• Publishing the electronics sources? Check
• Developing the software framework? Check
• Documenting everything? Check
• Having a OpenSourceHardware-'compliant' repository? Check

However, on my path to hack ultrasound, murgen will give birth to a set of modules: instead of being a shield, the electronics are going to be transformed into a set of (breadboard) plugable boards, or modules.

To reflect, this I'm starting a new Hackaday project, the echomods. Feel free to follow this new adventure!

What do you think of this comparative? A DIY ultrasound machine at 500$ vs a commercial one costing 100k$ ;)

The USB isolator is not working for the bitscope noise.. so instead I focused on existing data and cleaned those..
Improvement was done compared to the previous image taken, as you can see below: we're quite on par with the medical image.

Still need to have a "clean" scan conversion using a 4Tap approach instead of a Nearest neighbour, anyone for a python challenge?

Another couple of hours worth of playing - full lab log on github (Session 6) !

We finally get some images from the fantoms (that is, condoms filled with agar gel and tapioca balls).. and below is the setup, upper left the clinical image (coming from a 100k$ machine) and our image (lower). We clearly see the bigger tapioca balls (orange circles) and smaller (blue marks).

Mission accomplished =)

Same type of config, from a different angle: seeing low diffusion areas =)

Last but not least, the noise (circled in red) is due to the bitscope itself, I'm getting a USB isolator to get rid of this damn noise.

A full image