Get best of the week

Telemedicine. Reverse Engineering Electronic Stethoscope


I’ve been engaged in the development of electronics for more than 10 years, but it somehow happened that my first project in the field of medicine only appeared last year, and after that I didn’t get off this topic . As usual in such cases, the development, it was decided to start with the study of prototypes. In this article I want to share the side results of the reverse engineering of the fetus of a telemedicine startup with Chinese roots - an electronic stethoscope. If you are interested in what the insides of a typical representative of startups in the field of telemetry, launched in the early 2010s, are hidden, and why such a circuit design solution was chosen, you are welcome to cat.



Click if you don’t know, but what kind of beast is this - a stethoscope?
C , . , , , , . : ( ), ( ) ( ). «» . «» , , .

Despite the fact that the guys started with their project in Australia, they managed to raise the first round of investments only in China. Given that the backbone of the team was from the Eternal Empire, and China has a serious government program to help startups, this is not at all surprising. However, we were distracted, the topic of this article is the internal structure of the device, and not the history of the startup's development. About this, perhaps next time.

Let's look inside the package




Looking inside, an extremely high-quality cardboard packaging, we find a pleasant-to-touch case with two devices. The one that is larger and extended like a crocodile is a non-contact thermometer. Round is our today's patient, a stethoscope.

The guys missed a lot with the thermometer - not only did the stability of the measurements turn out to be extremely low, but the reliability of the work also let us down. For reverse engineering, we bought a used device and the non-contact thermometer in it turned out to be faulty. So let's focus on the stethoscope. First, let's take a closer look at the connecting cable.



You can see that on the one hand it ends with an audio jack, and on the other hand with a microUSB connector. Why is it easier to understand if you mentally travel back to 2012 - the year the startup was born. In those days, on Kickstarter, you could find a huge number of a wide variety of devices connected to the gadget in such an exotic way.

All of them were designed primarily to work with the then generation of the iPhone. He did not have a USB connector, and the organization of signal transmission over wireless channels with him was extremely difficult. Therefore, the gadgets were connected via an audio cable, through a headset jack, and the exchange of commands was carried out using audio signals encoded by the frequency and duration of the packets. It is this principle of communication that is also used in our experimental subject. It’s time, however, to look under his hood!

It's time to move on to the details


The cast metal case consists of two halves connected by a thread. Inside the case are two scarves. One of them is not worthy of our attention - only the holder of a three-volt battery is not located in it, but the second is worth considering in more detail and even on both sides.



On the left side of the boards we see a microUSB connector. The choice fell on him because it is much smaller in size than a standard audio jack with a diameter of 3.5 mm. The upper part of the board is shown on the right. It houses numerous power-blocking capacities and protective elements in three-lead SOT23 cases resembling transistors. Recall that we are dealing with a medical device, and even the simplest level of certification requires them.

The heart, and part-time and ear, of this device is certainly a microphone. A cheap electret, which, similarly to a headset microphone, can be powered from a phone, cannot be used here. Useful information of sound signals arriving at the headset of a stethoscope / phonendoscope lies in the range from 20 Hz to Hz 600. The lower range is limited by the frequency properties of the ear, the upper maximum frequency in which the noise of the heart and internal organs are located. Finding a compact microphone with such a frequency response is not easy. This device used an analog microphone manufactured by MEMS technology. The signal from it, without any additional amplification, entered the audio section of the smartphone. The microphone was carefully covered with a rubber pad, which I took off for a better view.

For convenience, I assigned numbers to the key components, and depicted below a greatly simplified functional diagram of the device.



It would seem that it is enough to put a microphone with a battery and not complicate everything with additional chips. However, even in the absence of an input signal, the current consumption of the microphone is enough to drain the battery in a few days. Of course, it would be possible to install a small-sized mechanical switch on the instrument case, but this is not the way of the Jedi, primarily because it is easy to forget to turn it off or, on the contrary, accidentally turn it on. As a result, at a time when you urgently need to listen to the patient’s lungs, the device will be inoperative. For medical applications, this is not at all buzzing, you can even say more - is unacceptable.

Our Chinese, reluctantly, had to go on to complicate the scheme. First of all, they added a microcontroller from the STM32L series with ultra-low consumption 1. (For low- power microcontrollers, see one of my previous articles ) Then they installed a voltage regulator in the power supply circuit of microphone 2 and finally went bankrupt on an analog switch chip.

The following functions are assigned to the control microcontroller:

  • Microphone Power Supply
  • Determination of the voltage level of the battery and, if necessary, the formation of a signal about its imminent discharge
  • Detection of a control sinusoidal audio signal from a smartphone
  • Generation of an analog sinusoidal signal for the process of data exchange with a smartphone
  • ,


After the user applies a stethoscope to the body of a person, he presses the button located on the screen of the smartphone. The smartphone sends a control audio sequence in which the listening time is encoded. She goes to the comparator microcontroller. Having sensed a voltage drop at its input, the microcontroller interrupts the state of deep sleep, decrypts the input signal and sends a response sound sequence containing information about the battery level to the microphone input of the phone. After that, the microcontroller supplies power to the microphone for a specified time and switches the audio switch to the mode of transmitting the signal from the microphone to the microphone input of the phone. At the end of a predetermined period of time, everything returns to its original state and the microcontroller plunges into deep sleep.

The smartphone uses the firmware to record the audio signal as a file in its own memory. In the future, you can listen to it, view it on the smartphone screen and even transfer it to the clouds for analysis and detailed processing.

Device disadvantages




Like many medical first-wave startups, to which I refer projects launched in the early 2010s, this one ultimately crashed. However, there is an idea to devote my next article to a detailed analysis of the reasons for the ups and downs of medical startups of that generation.

In this, I note only the obvious miscalculations associated with technical performance.
As a small charge for the mind, I suggest finding obvious blunders by ourselves, and then open the spoiler and compare with those

which I noticed
. , . , .

, , , . , , , , .

— . . , . , . , . , .

In conclusion, a big request to take a few seconds on a small but important poll for me. Thank.

More articles:


All Articles