Clinical IoT in Practice: A Novel Design and Implementation of a Multi-functional Digital Stethoscope for Remote Health Monitoring

Abstract

p>This study introduces a cutting-edge digital stethoscope optimized for remote health monitoring. Integrating Micro-Electromechanical Systems (MEMS) microphones with Bluetooth Low Energy (BLE) 5, the device ensures high-fidelity, real-time audio transmission. It seamlessly connects to an Internet of Things (IoT) platform through an embedded system, highlighting its potential in remote healthcare scenarios, especially during global health emergencies. Advanced features encompass Active Noise Cancellation, extended battery life, and intricate internal sound filtration. Paired with a dedicated Android application, the stethoscope streamlines the capture, storage, and visualization of auscultation data. Crucially, its integration with Electronic Health Records (EHRs) and its capability to generate vast datasets can significantly advance diagnostic precision, leveraging Digital Signal Processing and Artificial Intelligence methodologies. </p

Full text

BARAEINEJAD et al.: CLINICAL IOT IN PRACTICE: A NOVEL DESIGN AND IMPLEMENTATION OF A MULTI-
FUNCTIONAL DIGITAL STETHOSCOPE FOR REMOTE HEALTH MONITORING
1
Abstract— This study introduces a cutting-edge digital
stethoscope optimized for remote health monitoring. Integrating
Micro-Electromechanical Systems (MEMS) microphones with
Bluetooth Low Energy (BLE) 5, the device ensures high-fidelity,
real-time audio transmission. It seamlessly connects to an
Internet of Things (IoT) platform through an embedded system,
highlighting its potential in remote healthcare scenarios,
especially during global health emergencies. Advanced features
encompass Active Noise Cancellation, extended battery life, and
intricate internal sound filtration. Paired with a dedicated
Android application, the stethoscope streamlines the capture,
storage, and visualization of auscultation data. Crucially, its
integration with Electronic Health Records (EHRs) and its
capability to generate vast datasets can significantly advance
diagnostic precision, leveraging Digital Signal Processing and
Artificial Intelligence methodologies.
Index Terms— Digital Stethoscope, Remote Health Monitoring,
Internet of Things (IoT), High-fidelity Audio Transmission,
Digital Signal Processing (DSP)
I. INTRODUCTION
ardiovascular diseases (CVDs) and chronic respiratory
diseases (CRDs) are formidable global health
challenges. Annually, CVDs are responsible for
approximately 18.6 million deaths [1], and the prevalence of
CRDs, including conditions such as asthma, is rising,
affecting hundreds of millions worldwide [2]. Early detection
and monitoring of these diseases are critical for effective
management and treatment. In this regard, remote patient
monitoring through telemetry devices has gained traction as
an innovative approach, demonstrating improvements in
patient outcomes [3].
Corresponding author: Daryoosh Vashaee
Bardia Baraeinejad, Morteza Shams, Maryam Forouzesh, Shayan
Fakhraeelotfabadi, Radmehr Karimian, Saba Babaei, Diba Rashidi, Danesh
Germchi, and Pouya Gorji are with BIOSEN Group, Tehran, Iran (email:
bardiabarai@gmail.com; Mortezashams291@gmail.com;
maryam.forouzesh8@gmail.com; Shayan.Fakhraee@gmail.com;
radmehr.karimian@gmail.com; ssabababaeis@gmail.com;
dibra.rashidi@gmail.com; danesh.germchi@gmail.com;
pgxiii@gmail.com).
Mohammad Saberi Hamedani is with the School of Medicine, Shahid
Beheshti University of Medical Sciences, Tehran, Iran (e-mail:
m.saberihamedani@gmail.com).
Amirhossein Nasiri-Valikboni is with Physiology Research Center, Iran
University of Medical Sciences, Tehran, Iran (e-mail:
nasiri.amirho@gmail.com).
Yasaman Torabi is with Electrical and Computer Engineering
Department, McMaster University, Hamilton, Ontario, Canada (e-mail:
torabiy@mcmaster.ca).
Daryoosh Vashaee is with the Electrical and Computer Engineering
Department, North Carolina State University, Raleigh, NC 27606 USA (e-
mail: dvashae@ncsu.edu).
One of the classic tools in the diagnosis and monitoring of
CVDs and CRDs is the stethoscope. Employed for over two
centuries, it remains indispensable in clinical practice.
However, conventional acoustic stethoscopes have
limitations, including low sensitivity to faint heart sounds [4]
and susceptibility to ambient noise [5]. Moreover, they rely
heavily on the clinician's experience for the interpretation of
sounds, and subtle changes can easily be missed.
Digital stethoscopes have emerged to address these
limitations [6]. Through digital signal processing, these
devices can amplify heart and lung sounds and reduce
ambient noise, thus offering a more precise and sensitive tool
for auscultation [7]. This technological leap is particularly
valuable for detecting faint heart sounds such as S3 and S4,
which are often indicators of severe cardiac conditions.
Moreover, in the realm of pulmonary assessment, the
stethoscope plays a crucial role. The analysis of respiratory
sounds provides valuable insights into the health of the lungs.
Adventitious sounds such as crackles or wheezes can indicate
various pathologies. Digital stethoscopes can facilitate a more
quantitative examination of these sounds, enabling clinicians
to make more informed diagnoses [8, 9].
As the fields of biomedical engineering and digital
technology continue to converge, the integration of these
domains into the design of advanced digital stethoscopes
represents an important development. In this article, we
describe the design and implementation of two versions of an
advanced digital stethoscope, specifically engineered for
remote monitoring of cardiovascular and pulmonary sounds.
These designs are characterized by extended battery life,
facilitated through the incorporation of high-efficiency
switching converters, energy-efficient components, and
streamlined firmware. Bluetooth Low Energy (BLE) is
employed as the communication protocol to optimize energy
usage during wireless data transmission [10]. The initial
version of the stethoscope features a single microphone,
while the second iteration employs dual microphones to
capture a broader range of acoustic information.
Additionally, this paper introduces an Android application
developed to complement the digital stethoscope. The
application offers capabilities for visualizing, recording, and
replaying auscultation sounds. Notably, both the device and
the application are endowed with independent filtering
functions tailored for cardiac and respiratory applications.
This work represents a significant contribution to the
development of advanced diagnostic tools in the context of
cardiovascular and respiratory diseases. The integration of
digital technology into the design of these stethoscopes has
the potential to revolutionize patient monitoring, particularly
in remote or resource-limited settings. Furthermore, it sets the
stage for the generation of large-scale datasets that can be
Bardia Baraeinejad, Morteza Shams, Mohammad Saberi Hamedani, Amirhossein Nasiri-Valikboni,
Maryam Forouzesh, Shayan Fakhraeelotfabadi, Radmehr Karimian, Saba Babaei, Diba Rashidi,
Danesh Germchi, Yasaman Torabi, Pouya Gorji, and Daryoosh Vashaee
Clinical IoT in Practice: A Novel Design and
Implementation of a Multi-functional Digital
Stethoscope for Remote Health Monitoring
C

BARAEINEJAD et al.: CLINICAL IOT IN PRACTICE: A NOVEL DESIGN AND IMPLEMENTATION OF A MULTI-
FUNCTIONAL DIGITAL STETHOSCOPE FOR REMOTE HEALTH MONITORING
2
invaluable for research and the development of machine
learning algorithms for automated analysis and diagnosis.
II. HARDWARE DESIGN
A. Overview of Hardware Design
This digital stethoscope represents a blend of low-power
consumption, ergonomic design, and high sensitivity. It
captures and encodes audio streams, which are then
transmitted to the Android host device via BLE for
processing, visualization, playback, and storage.
Simultaneously, it supports wired playback from the
stethoscope to the headphone, adding flexibility in usage.
The overarching aim of this design is to surpass existing
models in sound quality and battery life. To achieve this, the
stethoscope utilizes multiple interconnected Printed Circuit
Boards (PCBs). To strike a balance between performance and
cost, these PCBs are designed with a 2-layer metalized
configuration, which offers significant cost reduction without
compromising the effectiveness of the device [11].
B. Hardware Block Diagram
Figure 1 displays the digital stethoscope’s hardware block
diagram, centered around a Micro Controller Unit (MCU)
with an ARM Cortex M4F processor, alongside peripherals
such as BLE 5 [12]. The diagram also displays a USB Type-
C interface, adeptly integrated with a lithium polymer (LiPo)
battery charger [13]. LiPo batteries were chosen due to their
high energy density, rechargeability, extended lifecycle,
lightweight structure, and leak resistance [14]. The USB port
is multifunctional, also serving as a headphone connector. An
integrated level translator controls the red, green, and blue
(RGB) light-emitting diodes (LEDs) under the MCU's
supervision [15].
User interaction is facilitated through two tactile push-
buttons and a rotary encoder for adjusting amplification
levels. The push-buttons serve for power toggling/BLE
activation and switching filter modes. The device integrates a
digital input amplifier for converting, filtering, and
amplifying acoustic signals [16]. RGB LEDs offer visual
feedback on BLE connectivity, amplification levels, filter
modes, and battery status. Initially, the system integrated
seven LEDs [17], while ten LEDs are used in the updated
version [18]. A DC-DC step-down converter reliably
provides 1.8V (sourced from the LiPo battery voltage, which
varies between 3.2V and 4.2V) to the components, as
depicted in the block diagram [19].
A noteworthy distinction between the two implementations
lies in the microphone configurations. The initial version
incorporates a single microphone, whereas the updated
version employs a dual-microphone setup for active noise
cancellation (ANC). This enhancement significantly
improves auscultation sound clarity and fidelity, which is
crucial for precise and reliable diagnostics.
C. MCU
The digital stethoscope utilizes an nRF52832 MCU, which
features a 32-bit ARM Cortex M4F processor operating at 64
MHz. This MCU was chosen for its optimal balance between
low power consumption and high performance. According to
its datasheet [12], the power consumption ranges from 0.3 μA
in idle mode to 58 μA/MHz when flash memory operations
are involved.
The nRF52832 MCU boasts various peripherals including
Bluetooth Low Energy, Pulse Density Modulation, Serial
Peripheral Interface, Inter-IC Sound Bus, Analog to Digital
Converter, and Pulse Width Modulation. These peripherals,
in conjunction with the processor's capabilities, enable the
digital stethoscope to achieve the necessary sound quality
while adhering to specific sample rate and resolution criteria.
Audio captured from the microphone(s) is initially in Pulse
Density Modulation format. The data undergoes down-
sampling, reducing the sample rate from 15,625 to 6,250
samples per second, enhancing the reliability of audio
streaming via Bluetooth Low Energy. Each sample has a 16-
bit resolution.
Serial Peripheral Interface is critical for transmitting the
amplification level to the amplifier and setting filter
coefficients. Inter-IC Sound Bus facilitates audio data
exchange between the MCU and the Digital to Analog
Converter. The built-in Analog to Digital Converter is used
to measure battery charge levels. General-Purpose
Input/Output pins receive interrupts regarding battery
charging modes and activate MOSFETs controlling power
consumption of LEDs and the level translator. Pulse Width
Modulation is used for controlling LED colors through the
level translator.
Collectively, these features and functionalities enable the
digital stethoscope to achieve high precision and efficiency in
audio capture, processing, and transmission.
Fig. 1. Block diagram of the digital stethoscope. In the second
implementation, an additional microphone is integrated for ANC. Most
system components operate at 1.8 V, except for the Filtering, DAC,
Amplifier, Level Translator, MOSFETs, and Step-Down Converter, which
are directly powered by a 680 mAh LiPo battery. The remaining units draw
power from the Step-Down Converter.
D. Microphone(s)
Bioacoustic signal monitoring is increasingly adopting Micro-
Electromechanical Systems (MEMS) microphones over traditional
Electret Condenser Microphones due to MEMS's compact size and
enhanced accuracy [20]. The initial version of the digital
stethoscope incorporates a single IM69D130 microphone, chosen
for its low inherent noise, outstanding Signal-to-Noise Ratio, and
broad dynamic range [21]. Positioned at the center of the lower
Printed Circuit Board, the microphone’s flat frequency response,
ranging from 28 Hz to 20 kHz, is especially beneficial for faithfully
capturing low-frequency sounds.
Recognizing the importance of reducing ambient noise, the
second version includes an additional IM69D130 microphone for
Active Noise Cancellation. This microphone is situated on a
different Printed Circuit Board.
Cardiac acoustic signals predominantly range between 70 Hz and
120 Hz [22], while respiratory acoustic signals vary more widely,
from 50 Hz to 2,500 Hz [23]. Considering the limitations of the
MCU and microphone operating on Pulse Density Modulation, the

BARAEINEJAD et al.: CLINICAL IOT IN PRACTICE: A NOVEL DESIGN AND IMPLEMENTATION OF A MULTI-
FUNCTIONAL DIGITAL STETHOSCOPE FOR REMOTE HEALTH MONITORING
3
PDM clock is set at 1 MHz, resulting in a sample rate of 15,625
samples per second (with a decimation factor of 64). This rate
satisfies the Nyquist criterion [24], ensuring the accurate capture of
acoustic signals without aliasing.
E. LEDs
RGB LEDs are integrated into the design as indicators for
various operational states. Each RGB LED is capable of
displaying over 16 million colors through a combination of
256 intensity levels for red, green, and blue colors. Distinct
colors and blinking patterns are assigned to represent
different operational modes, including battery charging
status, Bluetooth connectivity, filter selection, and audio
amplification levels.
An examination of the LEDs' datasheets indicated that they
operate at a minimum voltage of 3.3V and have considerable
current leakage when turned off, around 0.6 mA per LED [17,
18]. To address this, the design includes two parallel
MOSFETs as drivers to minimize power consumption.
Additionally, a level translator (TXB0101) is incorporated to
reconcile the voltage level differences between the
microcontroller unit and the LEDs. This translator ensures
seamless communication between the microcontroller and the
LEDs without the risk of damage or inconsistent
performance.
F. Filtering, Digital to Analog Converter (DAC), and
Amplifier
For amplification, noise filtering, and direct playback
through headphones, the device employs the TAS2521, a
low-power digital audio amplifier with an integrated
miniDSP [16]. It is compatible with 16-bit digital I2S mono
audio playback and features a variety of programmable audio
processing blocks, which are instrumental in filtering out
unwanted noise and ensuring the clarity and fidelity of the
captured audio.
The TAS2521 allows for programmable digital volume
gain between 0 dB and 24 dB. Configuration for the audio
section is achieved via the Serial Peripheral Interface (SPI).
The amplifier is distinguished by its high audio integrity,
evidenced by its Total Harmonic Distortion (THD) plus noise
rating, which typically registers at -78.2 dB. This low THD
plus noise metric signifies that the amplifier generates
minimal distortion and noise, crucial for the accuracy of
auscultation and subsequent clinical evaluations.
In the domain of digital stethoscopes, where the clarity and
accuracy of audio signals are vital for diagnostic purposes,
the selection of an amplifier is critical. The incorporation of
TAS2521 with its embedded miniDSP offers a powerful
combination for amplification and filtering, assuring that
healthcare professionals have access to high-quality audio
data for informed clinical decision-making.
G. Power Management
The digital stethoscope draws power from a rechargeable
680 mAh Lithium Polymer (LiPo) battery. The selection of a
LiPo battery is strategic due to its high energy density and
lightweight properties, which contribute to the portability and
longevity of the device. Voltage regulation is handled by the
TPS62840 step-down DC-DC buck converter, which steps
down the battery voltage to 1.8 V to efficiently power the
relevant component [19]. The adoption of a switching
converter is intentional for its high efficiency, which
translates into an extended battery life.
TABLE I
POWER CONSUMPTION FOR EACH PART
A comparison between components power consumption.
The battery is designed to continuously supply power to
the entire circuitry for an impressive 72 hours while
connected to headphones. Charging the battery is facilitated
through a USB Type-C connector, utilizing the BQ24091
charging unit [13]. To foster an extended battery life cycle,
the design deliberately accommodates a 3-hour charging
duration for a full battery recharge [25].
Table I offers a comprehensive summary of the power
consumption associated with each device function and the
respective components tasked with executing these functions.
With the power consumption profile in perspective, the
subsequent equations break down the total power
consumption, the power supplied by the battery, and the
resulting battery lifespan:
• Total Power Consumption ≈ (Effective Processing Power
≈ 3 mW) + (Effective BLE Power ≈ 10 mW) + (Effective
Microphones Power ≈ 2 mW) + (Effective RGB LEDs Power
≈ 3 mW) + (Effective Playback Power ≈ 16 mW) ≈ 34 mW.
• Battery Power ≈ Average Battery Voltage x Battery
Capacity ≈ 3.7 V x 680 mAh ≈ 2516 mWh.
• Battery Life = Battery Power / Total Power Consumption
≈ 72 hours.
In summation, through adept power management and
judicious component selection, the digital stethoscope
accomplishes a considerable operational duration. This is
indispensable for extended usage without the need for
frequent recharging - a feature that is particularly
advantageous in clinical environments where reliability and
availability are paramount.
H. Printed Circuit Boards (PCB) Design
The digital stethoscope employs two double-sided Printed
Circuit Boards, both measuring 0.8 mm in thickness, in its
initial configuration. Each of these PCBs plays a critical role
in the function and user experience of the device.
On the upper side, the first PCB (referred to as PCB #2)
hosts a rotary encoder and seven RGB LEDs (Fig. 2) [17].
The role of the rotary encoder is to provide an intuitive and
tactile way for the user to interact with the device, offering
adjustments to various settings. The RGB LEDs, on the other
hand, serve as vital visual cues, signaling different
operational statuses, making it easy for the user to monitor
the device's functioning at a glance.
Part Number
Functionality
Quantity
Supply Current
Per One Unit
Total Power
Consumption
While Active
nRF52832
Processing
1
3.7 mA
(Maximum)
6.7 mW
(Maximum)
nRF52832
BLE
transceiver
1
10.7 mA
(Maximum)
14 mW
(Maximum)
IM69D130
Microphone
2
550 µA
2 mW
WS2812B-
2020
RGB LED
10
20 mA
(Maximum)
740 mW
(Maximum)
TAS2521
Mono sound
playback
1
4.5 mA
16 mW
(Maximum)

BARAEINEJAD et al.: CLINICAL IOT IN PRACTICE: A NOVEL DESIGN AND IMPLEMENTATION OF A MULTI-
FUNCTIONAL DIGITAL STETHOSCOPE FOR REMOTE HEALTH MONITORING
4
PCB #1, located on the lower side, contains the remaining
core components, including the MCU, filtering and
amplification systems, the power management system, and
the primary microphone. This configuration was selected
specifically to allow for a 680 mAh LiPo battery to be snugly
sandwiched between the two PCBs, optimizing space
utilization and contributing to the compact form factor of the
device.
Fig. 2. Images of PCBs #1 and #2 from the first stethoscope implementation.
Notable components and PCB dimensions are indicated.
Fig. 3. Images of PCBs #1, #2, and #3 from the second stethoscope
implementation. Important components and PCB dimensions are marked.
The PCBs comply with the Restriction of Hazardous
Substances (RoHS) directive, featuring an Electroless Nickel
Immersion Gold (ENIG) surface finish [26]. This choice not
only aligns with environmental standards but also provides a
surface that is ideally suited for mounting fine-pitch
components - a necessity for devices with sophisticated and
compact designs like a digital stethoscope.
In the second implementation (Fig. 3), modifications were
made to further enhance the device's capabilities. PCB #2
now hosts ten RGB LEDs [18], and a third PCB (PCB #3) has
been introduced. This new PCB exclusively houses the ANC
microphone, a significant upgrade to the device's ability to
mitigate environmental noise, which can potentially interfere
with the captured acoustic signals.
I. Firmware Functionality Flowchart
The firmware functionality plays a central role in
managing the digital stethoscope's operation, facilitating
seamless interaction among the hardware components,
providing a user-friendly interface, and enabling wireless
communication for remote monitoring and data analysis. The
flowchart, illustrated in Fig. 4, provides a comprehensive
view of how the digital stethoscope operates, the interactions
among its internal components, and its communication with
external devices, primarily a smartphone, via BLE. At the
very outset, the stethoscope functions as a BLE peripheral
device, establishing a connection with a smartphone, which
serves as the BLE central device. This connection facilitates
the transmission of the audio stream and battery status from
the stethoscope to the smartphone. Sound signals captured by
the microphone are sampled by the MCU. This sound data is
received in Pulse Density Modulation (PDM) format with a
16-bit resolution and a sample rate of 15625 samples per
second.
Fig. 4. Firmware flowchart illustrating data transmission and system
functionalities. Sound data is conveyed to the MCU via PDM. The I2S
interface transmits this data from the MCU to the Filtering, DAC, and
Amplifier unit. Filter and amplification configurations are transferred to the
same unit via an SPI connection. The remaining firmware operations pertain
to data transmission through BLE and user interactions facilitated by
input/output pins.
Following the audio sampling, the data undergoes signal
processing, which entails several critical steps. The sampled
data is dispatched to an amplifier, the TAS2521, via the Inter-
IC Sound Bus (I2S) protocol. Here, the amplifier carries out
numerous functions such as filtering the sound data,
converting the digital sound data to analog signals, and
amplifying these signals for clear audio playback. To control
the amplification level and filter configuration, the MCU
communicates these settings to the amplifier using the SPI
protocol. This ability to manipulate the amplification and
filters plays a crucial role in ensuring that the audio is
accurately representative of the biological sounds of interest.
The firmware also supports user interaction through two
push-buttons and a rotary encoder. These controls, linked to
the MCU via General-Purpose Input/Output (GPIO) pins,
serve various functions. The push-buttons can turn the device
on/off, initiate BLE advertising, and cycle through different

BARAEINEJAD et al.: CLINICAL IOT IN PRACTICE: A NOVEL DESIGN AND IMPLEMENTATION OF A MULTI-
FUNCTIONAL DIGITAL STETHOSCOPE FOR REMOTE HEALTH MONITORING
5
filter modes, while the rotary encoder lets the user adjust the
amplification level.
The USB Type-C connector, a multi-functional
component, facilitates headphone connection, battery
charging, and MCU programming through Serial Wire Debug
(SWD). RGB LEDs, with their capacity to display various
colors and patterns, serve as visual indicators, notifying users
about different operational statuses and modes, including
BLE connectivity, battery charging status, filter modes, and
amplification levels.
Lastly, in the second implementation, an additional
microphone is incorporated to provide ANC, which
significantly improves the clarity of the audio signals by
reducing ambient noise.
Fig. 5. Stethoscope Body Design: (a) 3D rendering of the body for the first
stethoscope implementation, (b) Exploded 3D view of the body for the
second stethoscope implementation, (c) Top-view 3D rendering of the body
for the second stethoscope implementation, (d) Images of the final
manufactured bodies for both the first and second implementations,
displayed from left to right respectively.
J. Body Design
The body design of the digital stethoscope (Fig. 5) reflects
careful consideration of ergonomics, aesthetics, and
performance, which have been refined across two
implementations. In the first implementation, the main
enclosure comprises 3D-printed parts fabricated via
Stereolithography (SLA) resin printing. This technique
confers a smooth surface finish. To enhance aesthetics, parts
of the stethoscope are made of stainless steel, fabricated
through laser cutting and Computerized Numerical Control
(CNC) bending. However, the use of stainless steel is
moderated to prevent interference with the antenna's
performance. The diaphragm, a critical part for capturing
sound, is composed of a rigid plastic material such as epoxy
fiberglass or Bakelite [27]. The device is ergonomically
designed, featuring a cylindrical base for easy handling, with
two conveniently placed push-buttons on the side facilitating
user interaction. The removable diaphragm is situated at the
bottom of the device. Notably, a curved air cavity placed
between the diaphragm and the microphone enhances sound
sensitivity [28].
In the second implementation, the main enclosure and
volume knob are fabricated using HP Multi Jet Fusion (MJF)
technology with HP-PA12 material [29]. This approach
improves the stethoscope's structural strength and aesthetic
appeal. The fastening part is made of 316L stainless steel,
produced through Selective Laser Melting (SLM)
technology, which adds strength and precision to the metal
components. The diaphragm material remains consistent with
the first implementation. The design in the second
implementation is streamlined, featuring fewer parts, thereby
enhancing handling and aesthetic appeal. The cap situated
over the diaphragm is crafted from a rigid resin, 3D printed
using SLA technology to maximize sound capture efficiency.
Both implementations of the digital stethoscope succeed in
delivering a sturdy, user-friendly, and efficient medical
device. The addition of a diaphragm and ergonomic design
elements optimize sound sensitivity, thereby elevating the
device's effectiveness in medical diagnostics.
III. SOFTWARE DESIGN
A. Overview of Software Design
The software platform developed for this digital
stethoscope offers comprehensive capabilities for viewing,
filtering, and analyzing the captured signal across two
interfaces: an Android application and a web application.
The Android application provides real-time functionality,
enabling users to view, filter, play, and save the received
signal directly on their devices [30]. It offers on-the-go
accessibility and flexibility, making the stethoscope highly
versatile and user-friendly.
In addition, a robust web application has been developed
for a more detailed diagnosis, effective data management, and
educational endeavors. The application's interface is intuitive
and user-friendly, supporting a wide range of functionalities
from basic signal processing to advanced data analytics.
For future research and improvements, cloud processing
can be leveraged to perform sound segmentation and
facilitate computer-aided differential diagnosis. This feature
could greatly enhance the digital stethoscope's utility,
potentially enabling it to identify a wide range of health
conditions based on the captured sound data.
B. Android Application
The Android application, developed utilizing the Android
Software Development Kit (SDK) [31], displays real-time
raw audio data received from the digital stethoscope and
enables users to save this data on their Android device. The
application permits users to fine-tune sound levels and filter
settings in accordance with their preferences and specific use
cases (Fig. 6).
Two distinct Finite Impulse Response (FIR) filters are
incorporated in the app, each designed for heart and lung
modes respectively. These FIR filters are favored due to their
linear phase attributes, stability, and simpler implementation
[32]. To cater to various user needs, these filters can be
toggled on or off. Designed using the Kaiser windowing
method, these 48th order filters provide superior lobe width
and ripple ratio characteristics [33].
To optimize audio capture for cardiac and respiratory
sounds, two band-pass filters are devised. The first filter,
targeting cardiac sounds, spans from 20 Hz to 500 Hz. The
second filter, intended for respiratory sounds, covers the
(a)
(b)
(c)
(d)

BARAEINEJAD et al.: CLINICAL IOT IN PRACTICE: A NOVEL DESIGN AND IMPLEMENTATION OF A MULTI-
FUNCTIONAL DIGITAL STETHOSCOPE FOR REMOTE HEALTH MONITORING
6
range of 50 Hz to 1000 Hz (Fig. 8) [34]. These filters help to
enhance the clarity of the relevant audio frequencies, further
improving the usability of the digital stethoscope.
Fig. 6. Screenshots from the Android application displaying signal
monitoring after applying heart (left one) and (right one) lung sound filters.
C. Web Application
The web application offers an additional platform for end-
users to manage their data. Accessible via standard web
browsers, it leverages Representational State Transfer
(REST) Application Programming Interface (API) to
facilitate various tasks such as user login, data uploading,
sound plotting, and playback (Fig. 7).
Similar to the Android application, the web application
incorporates the same set of filters (heart and lung modes) for
refining the audio data. These filters can be manipulated as
per the user's requirements, providing flexibility in data
analysis.
One of the critical features of this web application is its
data storage capability. Users have the option to upload their
data to a secure cloud storage
(https://dataset.biosengroup.com). This cloud storage
platform is currently under evaluation for ethical compliance.
Once approved, it could provide a valuable repository for
future studies and research in this field, fostering
advancements in digital stethoscope technology and its
applications.
Fig. 7. A user interface view from the web-based auscultation platform.
D. Data Encoding and Compression
Data transfer and processing efficiency is a significant
aspect of the digital stethoscope system, especially when
considering telemetry and data storage requirements. This is
addressed using the Free Lossless Audio Codec (FLAC)
encoding format, known for its efficiency in compressing
audio data. It leverages the high correlation between
consecutive samples of an audio signal, segmenting input
based on sample rate and spectral characteristics, and reports
residuals [35].
In parallel with FLAC, the Android application records
data using a more human-readable in-house encoding format.
This encoding starts with a line containing key parameters
(sample rate, resolution, patient gender, age, stethoscope
position, user comments) separated by commas. The next line
contains audio samples expressed as hexadecimal two's
complement characters (e.g., "FFFF" = -1, "0000" = 0, and
"012A" = 298 for a 16-bit audio sample). These data encoding
and compression techniques ensure efficient data
management, crucial for subsequent analysis.
Fig. 8. Frequency responses of designed FIR filters for cardiac (top) and
respiratory (bottom) sounds.
IV. COMPARATIVE ANALYSIS AND DIAGNOSTIC POTENTIAL
Table 2 provides a comparative analysis of our digital
stethoscope with four other models, examining key features
such as connectivity, battery life, sound amplification, noise
cancellation, and the use of MEMS microphones. Our model
stands out with its versatility, offering both wired headphones
and wireless transmission via BLE to a smartphone
application. It also boasts a long continuous runtime of 72
hours, representing a more energy-efficient design than
comparable models.

BARAEINEJAD et al.: CLINICAL IOT IN PRACTICE: A NOVEL DESIGN AND IMPLEMENTATION OF A MULTI-
FUNCTIONAL DIGITAL STETHOSCOPE FOR REMOTE HEALTH MONITORING
7
Moreover, our device amplifies sounds by nearly 160 times
(precisely 44 dB), enhancing the clarity of auscultation.
Inclusion of ANC minimizes environmental disruption,
aiding in accurate diagnosis. The MEMS microphones
employed provide a wider dynamic range, higher Signal-to-
Noise Ratio (SNR), and a flatter frequency response,
resulting in superior sound quality.
The pivotal comparison between digital and traditional
mechanical stethoscopes lies in their diagnostic capabilities
for diverse diseases. Our digital stethoscope introduces
innovative design elements, enabling noise removal,
amplification of faint sounds, and potential integration of
artificial intelligence in future designs. This suggests a
potentially increased diagnostic power for healthcare
professionals, subject to confirmation through statistical
studies and expert feedback as digital stethoscopes gain more
widespread use.
TABLE II
DIGITAL STETHOSCOPES COMPARISON
A comparison between this device and commercial devices.
Our stethoscope's extended battery life, high sound quality,
effective resolution, and sample rate render it an ideal tool for
telemonitoring, especially in cases requiring regular re-
examination. It also facilitates patient self-monitoring in
chronic diseases following proper instructions, enabling more
efficient treatment modifications. Additionally, it provides a
valuable resource of audio data for the development of
disease diagnostic algorithms and establishment of a cloud-
based analytical platform.
Lastly, our ongoing dataset collection opens up
possibilities for an AI warning system to aid in preliminary
disease screening, giving patients the autonomy to seek
medical help when needed, and enhancing the prognosis of
various conditions.
V. CONCLUSION
In the ever-evolving landscape of healthcare and
technology, the integration of traditional medical instruments
with modern technological advancements is crucial. This
paper has successfully introduced a state-of-the-art digital
stethoscope, leveraging the potential of the Internet of Things
(IoT) for remote health monitoring. Through its high-quality
sound capture, transmission capabilities, and advanced
features such as Active Noise Cancellation, this device holds
the potential to become a valuable asset in clinical
auscultation. Moreover, its compatibility with an Android
application further facilitates the capture, storage, and
visualization of audio data, bridging the gap between patients
and healthcare providers, especially in remote or pandemic-
stricken areas. The integration with Electronic Health
Records (EHRs) and the potential for large-scale data
collection offers promising avenues for further research,
especially in the realm of Digital Signal Processing and
Artificial Intelligence. As telemedicine and remote healthcare
become increasingly relevant, tools like the digital
stethoscope described herein hold the potential to play a
significant role in advancing patient care and refining
diagnostic accuracy.
REFERENCES
[1] G. A. Roth et al., "Global burden of cardiovascular
diseases and risk factors, 1990–2019: update from
the GBD 2019 study," Journal of the American
College of Cardiology, vol. 76, no. 25, pp. 2982-
3021, 2020.
[2] P. Song et al., "Global, regional, and national
prevalence of asthma in 2019: a systematic analysis
and modelling study," Journal of Global Health,
vol. 12, 2022.
[3] S. Wongvibulsin, S. S. Martin, S. R. Steinhubl, and
E. D. Muse, "Connected health technology for
cardiovascular disease prevention and
management," Current treatment options in
cardiovascular medicine, vol. 21, no. 6, pp. 1-15,
2019.
[4] A. Tassin et al., "Relationship between amplitude
and timing of heart sounds and endocardial
acceleration," Pacing and Clinical
Electrophysiology, vol. 32, pp. S101-S104, 2009.
[5] V. V. V. Vis-à-Vis, V. Varactors, J. Williams, and
D. Beebe, "Low Noise Varactor Biasing with
Switching Regulators," 2000.
[6] M. Choudry, T. S. Stead, R. K. Mangal, and L.
Ganti, "The History and Evolution of the
Stethoscope," Cureus, vol. 14, no. 8, 2022.
[7] S. Swarup and A. N. Makaryus, "Digital
stethoscope: Technology update," Medical devices
(Auckland, NZ), vol. 11, p. 29, 2018.
[8] F. A. Loscalzo J, & Kasper D, & Hauser S, & Longo
D, & Jameson J(Eds.),Eds. Joseph Loscalzo, et al.,
Harrison's® Principles of Internal Medicine, 21st
edition. McGraw Hill LLC, 2022.
[9] M. Peter Libby, PhD, Robert O. Bonow, MD, MS,
Douglas L. Mann, MD, Gordon F. Tomaselli, MD,
Deepak Bhatt, MD, MPH, FACC, FAHA, FSCAI,
FESC, Scott D. Solomon, MD and Eugene
Braunwald, MD, MD(Hon), ScD(Hon), FRCP,
Braunwald's Heart Disease, Single Volume, 12th
Edition. Elsevier, 2022.
[10] J. Tosi, F. Taffoni, M. Santacatterina, R. Sannino,
and D. Formica, "Performance evaluation of
bluetooth low energy: A systematic review,"
Sensors, vol. 17, no. 12, p. 2898, 2017.
[11] M. K. Armstrong, "PCB design techniques for
lowest-cost EMC compliance. Part 1," Electronics
& communication engineering journal, vol. 11, no.
4, pp. 185-194, 1999.
[12] "Nordic Semiconductor: nRF52832 Product
Specification v1.4," ed, 10th October 2017, p.
Product
Name
Amplification
Active Noise
Cancellation
Sensor Type
Battery
Life
Data
Transmission
Thinklabs
One [36]
100x
No
Patented
Capacitive
Transducer
4 Hours
Bluetooth
3M
Littmann
CORE
[37]
40x
Yes
Piezoelectric
Sensor
8 Hours
Bluetooth
eKuore
Pro [38]
20x
No
N/A
7 Hours
External Wi-Fi/
Bluetooth
Module
Connected to
the Headphone
Jack
E-Scope
[39]
30x
No
N/A
8 Hours
Internal Device
Memory
This
Work
160x
Yes
MEMS
Microphone
72 Hours
BLE

BARAEINEJAD et al.: CLINICAL IOT IN PRACTICE: A NOVEL DESIGN AND IMPLEMENTATION OF A MULTI-
FUNCTIONAL DIGITAL STETHOSCOPE FOR REMOTE HEALTH MONITORING
8
https://infocenter.nordicsemi.com/pdf/nRF52832_P
S_v1.4.pdf.
[13] "Texas Instruments: BQ2409x 1-A, Single-Input,
Single-Cell Li-Ion and Li-Pol Battery Chargers," ed,
January 2010 – REVISED April 2021, p.
https://www.ti.com/lit/ds/symlink/bq24092.pdf?ts=
1669883934204&ref_url=https%253A%252F%25
2Fwww.google.com%252F.
[14] H. Venkatasetty and Y. Jeong, "Recent advances in
lithium-ion and lithium-polymer batteries," in
Seventeenth Annual Battery Conference on
Applications and Advances. Proceedings of
Conference (Cat. No. 02TH8576), 2002: IEEE, pp.
173-178.
[15] "Texas Instruments: TXB0101 1-Bit Bidirectional
Level-Shifting and Voltage Translator
With Auto Direction-Sensing and ±15-kV ESD Protection,"
ed, January 2007–REVISED March 2017, p.
https://www.ti.com/lit/ds/symlink/txb0101.pdf?ts=
1669798829862&ref_url=https%253A%252F%25
2Fwww.ti.com%252Fproduct%252FTXB0101.
[16] "Texas Instruments: Digital Input Class-D Speaker
Amplifier with Embedded miniDSP and mono
headphone
amplifier," ed, February 2013–REVISED February 2013, p.
https://www.ti.com/lit/ds/symlink/tas2521.pdf?ts=1
669884620366.
[17] "Worldsemi: WS2812B Ver. No.: V5 Intelligent
control LED integrated light source," ed, pp.
http://www.world-semi.com/DownLoadFile/108.
[18] "Worldsemi: WS2812B-2020 Intelligent control
LED integrated light source," ed, pp.
https://www.mouser.com/pdfDocs/WS2812B-
2020_V10_EN_181106150240761.pdf.
[19] "Texas Instruments: TPS62840 1.8-V to 6.5-V, 750-
mA, 60-nA IQ Step-Down Converter," ed, June
2019–REVISED March 2020, p.
https://www.ti.com/lit/ds/symlink/tps62840.pdf?ts=
1669818194652&ref_url=https%253A%252F%25
2Fwww.ti.com%252Fproduct%252FTPS62840.
[20] M. Fueldner, "Microphones," in Handbook of
Silicon Based MEMS Materials and Technologies:
Elsevier, 2020, pp. 937-948.
[21] "Infineon: IM69D130 High performance digital
XENSIVTM MEMS microphone," ed, 19th
December 2017, pp.
https://www.infineon.com/dgdl/Infineon-
IM69D130-DataSheet-v01_00-
EN.pdf?fileId=5546d462602a9dc801607a0e46511
a2e.
[22] P. Oktivasari, F. Haryanto, R. Riandini, and S.
Suprijadi, "A Real-Time Heart Rate Signal
Detection using an Electronic Stethoscope with
Labview," Journal of Biomedical Physics &
Engineering, vol. 10, no. 3, p. 375, 2020.
[23] M. E. Silverman, "Essentials of cardiac physical
diagnosis By Jonathan Abrams Lea & Febiger,
Philadelphia (1987) 300 pages, 170 illustrations,
$34.50 ISBN: 0‐8121‐1038‐2," ed: Wiley Online
Library, 1987.
[24] H. Nyquist, "American Institute of Electrical
Engineers," Transactions of The, vol. 47, pp. 617-
644, 1928.
[25] A. Tomaszewska et al., "Lithium-ion battery fast
charging: A review. eTransportation 2019, 1,
100011," Publisher Full Text, 2019.
[26] P. Label, P. Label, and R. oHS Compliance
Engineer, "Directive 2002/95/EC of the European
Parliament and of the Council of 27 January 2003 on
the restriction of the use of certain hazardous
substances in electrical and electronic equipment,"
Artesyn Technologies: Boca Raton, FL, USA, 2005.
[27] D. K. SITZ, Joseph P. JOHNSON, Mary Jo,
"Stethoscope Diaphragm," United States of
America, 28th July 2016. [Online]. Available:
https://patentimages.storage.googleapis.com/d4/9a/
61/5c5913a7ecdda0/WO2016118356A1.pdf
[28] M. Nussbaumer and A. Agarwal, "Stethoscope
acoustics," Journal of Sound and Vibration, vol.
539, p. 117194, 2022.
[29] "HP 3D HR PA12
Summary of Regulatory Compliance and Environmental
Attributes."
https://h20195.www2.hp.com/v2/GetDocument.asp
x?docname=4AA7-6989ENW (accessed.
[30] B. Baraeinejad et al., "Design and Implementation
of an Ultralow-Power ECG Patch and Smart Cloud-
Based Platform," IEEE Transactions on
Instrumentation and Measurement, vol. 71, pp. 1-
11, 2022.
[31] "SDK Tools release notes."
https://developer.android.com/studio/releases/sdk-
tools (accessed 4th December 2022.
[32] A. V. Oppenheim, Discrete-time signal processing.
Pearson Education India, 1999.
[33] T. Kaya and M. C. Ince, "The obtaining of window
function having useful spectral parameters by
helping of genetic algorithm," Procedia-Social and
Behavioral Sciences, vol. 83, pp. 563-568, 2013.
[34] A. A. Alanazi, S. R. Atcherson, C. A. Franklin, and
M. F. Bryan, "Frequency responses of conventional
and amplified stethoscopes for measuring heart
sounds," Saudi journal of medicine & medical
sciences, vol. 8, no. 2, p. 112, 2020.
[35] C. J. FLAC. "Free Lossless Audio Codec."
https://xiph.org/flac/index.html (accessed 1st
December 2022.
[36] Thinklabs. "Thinklabs ONE."
https://www.thinklabs.com/ (accessed 1st
December 2022.
[37] Littmann®. "The 3M™ Littmann® CORE Digital
Stethoscope."
https://www.littmann.com/3M/en_US/littmann-
stethoscopes/advantages/core-digital-stethoscope/
(accessed 1st December 2022).
[38] eKuore. "eKuore Pro Electronic Stethoscope."
https://ekuore.com/product/ekuore-pro-electronic-
stethoscope/ (accessed 1st December 2022.
[39] Cardionics. " e-scope II Operator’s Manual."
https://www.midwestern.edu/Documents/Simulatio
n%20Clinic/e-Scope.pdf (accessed.