An Android App for Training New Doctors in Mechanical Ventilation

Andrea Bombarda

1 a

, Sara Noto Milleﬁori

, Michela Penzo

, Luca Novelli

2 b

and Angelo Gargantini

1 c

Department of Management, Information and Production Engineering, University of Bergamo, Bergamo, Italy

Pulmonary medicine Unit, ASST Papa Giovanni XXIII, Bergamo, Italy

{s.notomilleﬁori, m.penzo1}@studenti.unibg.it

Keywords:

Mobile Application, Mechanical Ventilation, Application Prototype, Training.

Abstract:

Mechanical ventilation is essential for critically ill patients, as recently demonstrated by the COVID-19 pan-

demic. The experience of physicians in correctly selecting ventilation parameters and values plays a crucial

role in ensuring the best possible outcome. In order to aid physicians in setting up a mechanical ventilator,

several brands have implemented in their products an adaptive ventilation mode called Adaptive Support Venti-

lation (ASV). This mode automatically selects pressure and respiratory rate to require the patient the minimum

breathing effort possible. However, physicians are generally skeptical about adopting this ventilation mode, as

they prefer to have all parameters under their control. Nevertheless, we believe that comprehending how ASV

works is paramount important, to understanding the patterns used, and possibly exploiting them while manu-

ally setting mechanical ventilators. For this reason, in this paper, we present Ventilation App, an Android

app for training new physicians in mechanical ventilation. It allows the simulation of a ventilation process

for a patient unable to breathe and gives feedback to the user by exploiting the same operating principles of

the ASV mode. Thanks to the feedback received by a collaborating physician, we believe that our app can be

useful for allowing physicians-in-training to acquire proﬁciency in mechanical ventilation.

1 INTRODUCTION

Mechanical ventilation plays a crucial role in the

treatment of critically ill patients, and the proﬁciency

of physicians in this area is vital to provide optimal

patient care. However, acquiring the necessary skills

and knowledge in mechanical ventilation can be a

challenging task, as it requires a deep understand-

ing of complex respiratory physiology and the abil-

ity to interpret and adjust ventilation parameters ef-

fectively. To solve this issue, automatic ventilation

strategies such as the Adaptive Support Ventilation

(ASV) modes have been proposed. In this way, the

mechanical ventilator automatically deﬁnes the ven-

tilation parameters and values based on the patient’s

respiratory mechanics, and by computing them by us-

ing the Otis’ curve (Fern

andez et al., 2013). However,

the complexity of ASV can give rise to concerns and

skepticism among healthcare professionals. Physi-

cians may question its efﬁcacy and reliability, espe-

https://orcid.org/0000-0003-4244-9319

https://orcid.org/0000-0002-2705-248X

https://orcid.org/0000-0002-4035-0131

cially when compared to traditional modes of venti-

lation that they are more familiar with and that allow

them to manually manage the ventilation parameters.

Additionally, most clinicians may be hesitant to fully

embrace ASV due to a lack of comprehensive train-

ing or limited exposure to its implementation in real-

world clinical settings.

To bridge this gap and facilitate the learning pro-

cess for aspiring physicians, in this paper, we in-

troduce an Android application aimed at enhancing

training in mechanical ventilation by taking advan-

tage of the mechanisms used by the ASV strategy,

used by ventilators when automatically deciding the

ventilation strategy. It is based on the principle of

learning by doing, i.e., it lets the user manually set

the ventilation parameters after having decided on the

physiological and mechanical characteristics of the

patient. Then, the app simulates the mechanical ven-

tilation and gives feedback to the user by using the

measured effort and its distance from the optimal, i.e.,

the minimum one computed using the Otis equation

as automatically done by mechanical ventilators em-

ploying the ASV mode. In this way, physicians-in-

362

Bombarda, A., Milleﬁori, S., Penzo, M., Novelli, L. and Gargantini, A.

An Android App for Training New Doctors in Mechanical Ventilation.

DOI: 10.5220/0012323900003657

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 17th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2024) - Volume 2, pages 362-369

ISBN: 978-989-758-688-0; ISSN: 2184-4305

training can understand the ventilation patterns used

by the ASV mode and, possibly, exploit them while

manually setting mechanical ventilators.

With this paper, our contribution is twofold. On

the one hand, we contribute our Android app allowing

physicians to train with mechanical ventilation. On

the other hand, we apply the paradigm of learning-

by-doing in the context of healthcare, especially de-

voted to increasing the understanding of physicians

in automatic procedures that are normally avoided for

lack of conﬁdence. This approach can be generalized

to other ﬁelds, beyond mechanical ventilation, where

simulators are available.

The remainder of the paper is structured as fol-

lows. Sect. 2 presents the background on mechanical

ventilation, including operating principles and modes.

Moreover, it includes a brief analysis of existing An-

droid apps for training new physicians in terms of me-

chanical ventilation. In Sect. 3, we report the require-

ments and the architecture of the application we pro-

pose in this paper. Sect. 4 presents the implemented

app and explains its functioning. Finally, Sect. 5 re-

ports related work on apps for training physicians in

different ﬁelds, while Sect. 6 reports future work di-

rections and concludes the paper.

2 BACKGROUND

In this section, we report some background on the ba-

sis of mechanical ventilation, together with an analy-

sis of already existing apps for training physicians in

using mechanical ventilators.

2.1 Mechanical Ventilation

Artiﬁcial mechanical ventilation replaces (or inte-

grates) the activity of the inspiratory muscles to en-

sure a sufﬁcient supply of gas to the lungs. It is a nec-

essary tool in every modern intensive care unit (ICU).

The type and intensity of ventilation support required

by a patient vary over the course of treatment. Modern

mechanical ventilators are versatile and adapt to pa-

tient needs. They support patients who cannot breathe

or those who can still trigger a mechanical cycle by a

spontaneous inspiratory effort. Present-day mechan-

ical ventilators are complex machines, consisting of

many specialized components and featuring several

ventilation modes, which have to be set depending

on the conditions of the patients. There are numer-

ous conditions in which it is necessary to resort to

mechanical ventilation, for example in all those con-

ditions in which spontaneous natural breathing is not

possible (e.g., in the case of acute respiratory distress

syndrome - ARDS - or chronic obstructive pulmonary

disease - COPD).

Ventilation Modes. Depending on the patient and

the type of ventilator available, several ventilation

modes (i.e., speciﬁc combinations of breathing pat-

terns, control types, and operational algorithms) are

available in the products on the market, and the most

common are those based on pressure or volume con-

trol. Most of them operate in an open-loop way.

For what concerns volume-control ventilation

modes, the two most common modes are Mandatory

Minute Ventilation (MMV) and Controlled Manda-

tory Ventilation (CMV). With the former, the ventila-

tor provides a predetermined minute ventilation (i.e.,

a pre-deﬁned volume of air) when the patient’s spon-

taneous breathing effort becomes inadequate (e.g., be-

cause of an apnea). When this occurs, the manda-

tory frequency is increased automatically to compen-

sate for the decrease in minute ventilation caused by

the apnea and to ensure the desired minute venti-

lation. With the latter, the ventilator controls both

the patient’s tidal volume (i.e., the volume of air in-

spired/expired in each breath) and respiratory rate. In

practice, the ventilator controls the patient’s minute

volume. This means that a patient cannot change the

ventilator frequency or breathe spontaneously.

Instead, the two main pressure-control ventilation

modes are Pressure Controlled Ventilation (PCV) and

Pressure Support Ventilation (PSV). In PCV mode,

the ventilator controls the timing of the breathing cy-

cle and regulates the pressure applied to the patient.

PCV mode is used in the acute phase of the disease

when patients are deeply sedated or paralyzed, and

cannot breathe themselves. On the other hand, PSV is

an assisted ventilatory mode that is patient-triggered,

pressure-limited, and ﬂow-cycled. The main use of

this mode is for the weaning of the patient from me-

chanical ventilation because it gradually unloads the

work of breathing.

Ventilation Parameters. Depending on the venti-

lation mode, different parameters have to be set by

the physicians. In the following, a brief explanation

of the main ventilation parameters we exploit in the

work presented in this paper is given.

The respiratory rate (RR) indicates the number of

breaths per minute the patient has to take. For each

breath, the ratio between the inspiratory and expira-

tory times is set through the parameter I:E. During

the inspiration phase, the ventilator reaches the maxi-

mum pressure P

max

, which has to be set accurately in

order not to damage the patient’s lung with too high

pressure. On the other hand, during the expiration

An Android App for Training New Doctors in Mechanical Ventilation

363

Figure 1: Example of the Otis curve for the minimal WOB.

The red dot indicates the optimal target ventilation point.

phase, the ventilator sets its pressure to the positive

end-expiratory pressure (PEEP), which is greater than

zero to prevent the collapse of the airway.

2.2 Adaptive Support Ventilation

Adaptive support ventilation (ASV) is a mechanical

ventilation mode that operates in a closed-loop sys-

tem, adjusting the ventilator’s output for each breath.

It utilizes the patient’s measured mechanical charac-

teristics (e.g., airway resistance and compliance) and

breathing effort to determine the appropriate level of

ventilatory support. The primary objective of ASV is

to deliver a predetermined level of alveolar ventila-

tion while minimizing the combined work of breath-

ing (WOB) for both the patient and the ventilator.

ASV can provide complete or partial ventilatory sup-

port and is applicable during mechanical ventilation’s

initiation, maintenance, or weaning stages. The venti-

lation pattern is automatically decided by the ventila-

tor for reducing the WOB based on the Otis equation

RR(t) =

1 +2 · a · RC

min

− RR(t − 1) ·V

− 1

a · RC

where RC

is the expiratory time constant (deﬁned

through the resistance R and compliance C of the pa-

tient’s airway), V

min

is the target minute ventilation,

represents the dead space volume (computed as

2.2 · IBW , where IBW is the ideal body weight), a is

a factor depending on the ﬂow waveform (e.g., it is

0.329 for sinusoidal ﬂows), and RR is the respiratory

rate, i.e., the number of breaths per minute. Based

on Otis’s observations, if the ventilation pattern (res-

piratory rate and tidal volume) satisﬁes the proposed

equation, then the effort requested to the patient for

breathing is reduced to the minimum.

The ASV mode uses the Otis equation and con-

siders the curve in Fig. 1 for automatically adapt-

ing the respiratory rate (RR) and the tidal volume

min

/RR), i.e., the volume inspired or expired dur-

ing a single respiratory cycle. Based on this curve,

every ventilation pattern that guarantees to stay on the

blue line is optimal and lowers the patient’s effort to

the minimum possible.

Note that the Otis curve identiﬁes some safety lim-

its (see the red rectangle in Fig. 1), which are used by

ventilators when working in the ASV mode in order

to raise alarms and avoid dangerous conditions. In

terms of respiratory rate (x-axis in Fig. 1), the lower

limit is 5 respiratory cycles per minute, while the up-

per bound is computed as

/(R ·C) and, in any case, the

respiratory rate cannot exceed 60 breaths per minute.

On the other hand, in terms of tidal volume V

(y-

axis in Fig. 1), the safety limits depend on the pa-

tient’s IBW. Indeed, the minimum tidal volume can be

4.4

/kg · IBW , while the maximum is 22

/kg · IBW

and, in any case, the pressure applied by the ventilator

in order to achieve the intended tidal volume has to be

lower than 45cmH

Despite being automatic, the ASV mode is nor-

mally avoided by physicians who prefer to manually

set ventilation parameters in order to keep them under

control. However, we believe that it is essential for

every doctor to understand how ASV works. In this

way, physicians can learn some of the patterns that are

normally used by ASV and integrate them into their

manually deﬁned settings. This is the rationale behind

the application we propose in this paper.

2.3 Existing Apps for Training in

Mechanical Ventilation

Before starting the design of our app, we analyzed

existing applications related to training in mechanical

ventilation. With this analysis, we found four apps. In

the following, we give a brief overview of their fea-

tures and the differences between them and the appli-

cation we propose in this paper.

TruVent App (TruCorp, 2023) is a training appli-

cation in which the trainer and the physician interact.

The former sets patient characteristics, while the lat-

ter monitors and performs actions based on patient

status, which the trainer can continuously modify by

simulating different clinical conditions. However, the

TruVent App is not based on ASV nor uses the Otis

curve to give users feedback, as we do in our app.

Similarly, Ventsim (Ventsim, 2023) simulates interac-

tions between a patient and a ventilator and generates

waveforms. Different modes of ventilation are imple-

mented, and clinical scenarios can be tested by chang-

ing the ventilator and patient settings. However, as

TruVent App, it does not offer a way for the physician

to understand whether the settings are those leading

to the minimum breathing effort for the patient.

VentilO (IUCPQ, 2023) aims at training physi-

HEALTHINF 2024 - 17th International Conference on Health Informatics

364

cians in reasoning on how different values for the ven-

tilation parameters may inﬂuence the patient’s condi-

tion. Nevertheless, VentilO only implements volume-

based ventilation modes and, thus, the pressure-

controlled ones we want to deal with are not sup-

ported, nor training for users is given in those cases.

Finally, OpenPediatrics (OpenPediatrics, 2023)

provides a ventilator simulator embedding the most

recent pediatric and adult ventilation guidelines, as

well as COVID-19 information and COVID-19 pa-

tient cases. As for the TruVent app, the ventilator

simulator provided by OpenPediatrics is not based on

ASV nor uses the Otis curve to give users feedback,

as we do in our proposed app.

We can conclude that, to the best of our knowl-

edge, no Android app with the same operating princi-

ples as those we embed in Ventilation App exists.

3 APP DESIGN

At the beginning of the development process, we held

several meetings with physicians and experts to have

a deep understanding of how mechanical ventilation

works, especially in the context of the ASV mode,

and of the most perceived difﬁculties by physicians-

in-training when approaching this ventilation strat-

egy. Then, together with physicians and experts, and

thanks to the experience gained in the past by groups

working on the development of mechanical ventila-

tors (e.g., for treating patients affected by COVID-19

in ICUs) (Abba et al., 2021; Bombarda et al., 2021;

Bombarda et al., 2022; Pearce, 2020), we deﬁned the

functional requirements of the Android app for train-

ing new doctors in mechanical ventilation. In this

way, we have been able to identify critical features

and isolate those that were not of interest for the sce-

nario we wanted to deal with, i.e., that of the ASV

mode. Then, the app architecture has been deﬁned

according to the identiﬁed requirements.

3.1 Functional Requirements

During the process of the development of our An-

droid app, we started by individuating the require-

ments, which were then formalized in the Software

Requirements Speciﬁcation document. For each re-

quirement, we have identiﬁed a description (some-

times with a rationale, when it was provided by the

experts we contacted) and an ID. In the Software Re-

quirements Speciﬁcation, we have identiﬁed 4 differ-

ent modes, and, for each of them, we have deﬁned

the set of needed requirements. In the following, we

describe each mode in detail.

Startup Mode: In startup mode, the Android app

shows the logo and offers the user the possibility to

start using the app (see Settings mode) or to view in-

formation on its operating principles (see Info mode).

Info Mode: In the info mode, the Android app

reports a message explaining its operating principle,

including the thresholds used for deﬁning when the

breathing effort is low, medium, or high.

Settings Mode: In the settings mode, the user is

requested to set all the required parameters for the

simulation, i.e., both those of the patient and those of

the ventilator. For what concerns the patient, the app

asks the user to insert the gender, age, weight, height

(used to calculate the IBW), airway resistance, and

compliance. Then, for what concerns the ventilator,

the user must insert the basic ventilation parameters,

i.e., the PEEP value, the P

max

, the I:E fraction, and the

respiratory rate. When in this mode, the user may save

the data previously set into an embedded database, in

order to load them in another simulation, or load the

data from those previously saved. When all the pa-

rameters are set, the user may proceed by starting the

Simulation mode or by closing the app.

Simulation Mode: The simulation mode is the

core mode of the Android application. By using all

the data provided by the user during the Patient pa-

rameters set mode, it simulates the patient’s breath-

ing when the mechanical ventilator works in PCV

mode, i.e., when the patient is not able to breathe

autonomously. During the simulation, two plots are

shown: one that reports the pressure over time, and

one showing the ﬂow over time. In order to allow

users to understand the effect and level of correctness

of the settings, during simulation mode, a semaphore

is shown: the green light is highlighted when the mea-

sured effort is equal to the minimum possible (see

the Otis’ equation in Sect. 2.2) with a tolerance of

±150, the yellow light is highlighted when the mea-

sured effort is distant from the optimal one from ±150

to ±300, otherwise, the red light is shown. These

threshold values have been discussed and approved

by physicians and have been derived from the exam-

ples given in the operator’s manual of the Hamilton

Galileo ventilator

, which embeds the ASV mode.

As previously explained in Sect. 2.2, safety lim-

its are identiﬁed by the ASV mode. The red light

is shown even if the chosen ventilation pattern does

not imply a high effort for the patient, but it is out-

side those safety limits. When in simulation mode,

the user may change some conﬁguration parameters

(see Settings mode), freeze the simulation (see Freeze

mode), or stop the simulation and close the app.

https://bit.ly/3Zh4PIy

An Android App for Training New Doctors in Mechanical Ventilation

365

Exit

Click

Start

Simulation

Settings

Click

Freeze

Simulation

after insp.

duration

Inspiration

after expiration duration

Expiration

Click Stop

Simulation

Click Stop Simulation

Click

Resume

Adjust parameters

Open

App

Exit

Click Start

Startup

View Info

Click Info

Figure 2: State machine.

«Component»

INSPIRE

«Database»

PatientDB

«Component»

GUI

«Component»

Ventilator

Figure 3: High-level software architecture.

Freeze Mode: The freeze mode can be activated

by the user during the simulation. When the freeze

mode is active, the simulation pauses and the user

can visualize the plot details in order to better under-

stand the patient’s condition. Users may exit from the

freeze mode by restarting the simulation (see Simula-

tion mode) or by stopping the simulation and setting

new patient parameters (see Settings mode).

A general overview of the transitions between the

application modes and the functionalities is given in

the state machine in Fig. 2, while a diagram showing

the interaction between actors and software artifacts

is available in our open-source repository at https://

github.com/foselab/VentilationApp.

3.2 App Architecture

The high-level architecture of the training app is re-

ported in Fig. 3. It is composed of 4 main compo-

nents. The INSPIRE component implements the pa-

tient simulation logic. It handles the electronic equiv-

alent of human lungs (van Diepen et al., 2021), com-

posed by the resistance R and compliance C set by

the user through the GUI, and when a deﬁned pres-

sure is applied by the ventilator, it computes the cur-

rent ﬂow and pressure in the patient’s airway. The

Ventilator component reads the ventilator’s param-

eters from the GUI and simulates a simple mechanical

ventilator working in PCV mode (see Sect. 2.1). It

sets the pressure (between the PEEP and P

max

val-

ues) and sends its value to the INSPIRE component,

through a ZeroMQ

communication channel, which

executes a single simulation step. The PatientDB

stores the patient and ventilator parameters saved by

the user in previous simulations. It is also used by the

GUI for loading previously saved data. Finally, the

GUI component shows the user all the Android app

screens, including the two plots and the semaphore

that signals the effort. Through the GUI, the user can

start/stop/freeze the simulation, set the conﬁguration

parameters for the ventilator and those of the patient,

load them from the PatientDB, or decide to store the

data of the current simulation run into the database.

Extension Points: The architecture has been cho-

sen to promote the future extensibility of the app.

New ventilatory strategies may be implemented by

simply modifying the Ventilator component. Sim-

ilarly, considering that the accuracy of the simulation

depends on the patient models, more complex and

realistic ones (including other parameters beyond R

and C) can be easily added in future releases, as the

INSPIRE component can simulate any kind of circuit

representing the respiratory system.

4 APP PROTOTYPE

In this section, we describe the prototype of our train-

ing Ventilation App, available and open source

at https://github.com/foselab/VentilationApp. It has

been developed for mobile devices, with a responsive

design, in order to make it suitable both for smart-

phones and tablets. Fig. 4 shows the Android applica-

tion screens, which we better explain in the following.

The app starts with the splash screen shown in

Fig. 4a, which contains two buttons, i.e., one for

accessing the info section and one for starting the

training. When the info button is clicked, the app

https://zeromq.org/

HEALTHINF 2024 - 17th International Conference on Health Informatics

366

(a) Splash screen. (b) Info screen. (c) Main window.

Figure 4: Startup and View Info mode in the First prototype in Android.

shows information about its purpose and functioning

(Fig. 4b). Instead, when the start button is clicked, the

app moves to the main window (shown in Fig. 4c).

The main window contains the core functionali-

ties of our Ventilation App. It contains real-time

charts reporting the simulation outcomes, namely the

airﬂow in the patient’s airways and the pressure mea-

sured at the patient’s mouth. This information is used

by physicians to understand potentially dangerous pa-

tient conditions, such as the auto-PEEP (i.e., when the

pressure in the patient’s airways increases breath after

breath, due to a too-short expiration time). In addi-

tion, two buttons allowing the setting of patient char-

acteristics and ventilation settings are present. For

what concerns the patient (Fig. 5a), the user is re-

quested to insert the gender, age, weight, height (used

to calculate the IBW), airway resistance, and com-

pliance. Then, in terms of ventilation parameters

(Fig. 5b), the user must set the value of the PEEP, the

max

, the I:E ratio, and the respiratory rate.

As introduced in Sect. 3, Ventilation App al-

lows the user to load both patient conﬁgurations and

ventilation parameters from an embedded database

(see Fig.3). In the case of patients it can be done by

clicking on the load button in Fig. 5a. In this way, the

list of patients (including age, gender, height, weight,

resistance, and compliance) is shown (Fig. 5c). The

user may then decide to load the desired patient by

clicking on the ”check” button or to delete one of the

patients by clicking on the trash bin icon. If a pa-

tient is chosen, all the ﬁelds in the patient conﬁgura-

tion (Fig. 5a) are automatically ﬁlled with the values

read from the DB. Similarly, as described for the pa-

tient parameters, the user can load the data related to

the ventilator conﬁguration with a speciﬁc load button

(Fig. 5d). Therefore, a dedicated list is shown with

values of PEEP, PMax, I:E, and RR. Also in this case,

it is possible to select or delete the data with two dif-

ferent buttons (”check” or trash bin icon). When both

(a) Patient’s settings. (b) Ventilation set-

tings.

from DB.

(d) Ventilation

params selection

from DB.

Figure 5: Settings mode in the ﬁrst prototype in Android.

the patient’s parameters and the ventilator settings are

ﬁlled, the user can start the simulation by clicking the

play button in the main window (see Fig. 4c).

Thus, Ventilation App starts simulating a pa-

tient with the desired conditions who is undergoing

a mechanical ventilation procedure. During the sim-

ulation, at every breath, our app computes the WOB

for the patient and, based on it, lights up one of the

three semaphore lights, as explained in Sect. 3. If the

measured effort is low and near the target one, the

green light is shown, as in Fig. 6a. Instead, if the mea-

sured effort is higher but still acceptable, the yellow

An Android App for Training New Doctors in Mechanical Ventilation

367

(a) Simulation with

optimal settings.

(b) Simulation with

non-optimal settings.

wrong settings.

(d) Stop screen.

Figure 6: Simulation mode and Stop screen in the ﬁrst prototype in Android.

light is turned on, as in Fig. 6b. Finally, if the chosen

ventilation strategy is beyond the safety limits or the

measured effort is considered too high, the red light is

shown (see Fig. 6c). Indeed, from Fig. 6a to Fig. 6c

(where the chosen respiratory rate is too high and the

pressure cannot reach the set maximum value) we can

see that the pressure curve tends to become more ﬂat,

hence implying higher effort for the patient.

Having received the feedback on the chosen set-

tings and analyzed the relevant patient’s data through

the pressure and ﬂow curves, the user may decide

to freeze the ventilation (e.g., to take a screenshot)

with the pause button available in the main window

(Fig. 4c). Finally, when the simulation is completed,

the user may stop it by clicking on the stop button in

the main window (Fig. 4c). In that case, the pop-up

in Fig. 6d is shown. It lets the user choose whether to

stop and reset the simulation or to save the current set-

tings (both those of the patient and of the ventilator)

in order to load them for following simulations.

5 RELATED WORK

As reported in Sect. 2.3 no other Android apps offer-

ing the same functionalities we offer with our train-

ing app exist, to the best of our knowledge. However,

the effort spent by developers in designing and imple-

menting simulation apps for physicians demonstrates

the importance of such tools (Kunkler, 2006), espe-

cially in the most critical scenarios such as emergency

medicine (Reznek, 2002). A survey conducted by

the authors of (Wallace et al., 2012) reports that over

85% of the interviewed participants, including medi-

cal students and physicians, reported using a mobile-

computing device for their training. This is the reason

why we also decided to opt for a mobile app. Indeed,

especially due to the increasing complexity and pre-

ciseness of patient models, simulators can be useful

tools in determining a physician’s understanding and

use of best practices, management of patient compli-

cations, appropriate use of instruments and tools, and

overall competence in performing procedures. Also

in other medical domains beyond mechanical ventila-

tion, simulators have been used, for example, to train

doctors in the management of temporary pacemak-

ers (Crowe et al., 2013), in bedside cardiac examina-

tions (Takashina et al., 1997), in surgeries (Ph

e et al.,

2016), in chronic kidney disease treatments (Markos-

sian et al., 2021) and also during radiology proce-

dures (L. E. Wood, 2018; Toderis et al., 2022). Using

case-based simulation training software (as the app

we propose in this paper) has been shown to improve

physicians’ skills even more than simple textbook re-

views (Schwid, 2001; Couto et al., 2015).

6 CONCLUSIONS AND FUTURE

WORK

In this paper, we have shown the methodology

adopted to develop the ﬁrst prototype of a mobile

application to make the training of new physicians

in mechanical ventilation easier and more effective.

It exploits the principles of the minimum WOB, de-

scribed by the Otis equation, and used by ventilators

implementing the ASV mode to give users feedback

not only on the correctness of the set ventilation pa-

rameters, but also on their optimality, to reduce the ef-

fort required to patients when under mechanical ven-

tilation. Starting from the discussion of the idea with

pneumologists and the analysis of the already exist-

ing apps, we have ﬁrst deﬁned the requirements and

the architecture, and then implemented the initial pro-

totype presented in this paper. This prototype is just

preliminary work that we have used to verify the fea-

sibility of the idea and to stimulate discussion with ex-

perts in mechanical ventilation. In the future, we may

conduct further analysis with physicians-in-training

in order to reﬁne Ventilation App, and to investi-

HEALTHINF 2024 - 17th International Conference on Health Informatics

368

gate potential ethical and legal implications of using

it for medical education.

Thanks to the feedback received by the experts we

collaborated with, we have identiﬁed some key future

work. The ﬁrst future work direction will be focused

on having pre-deﬁned scenarios in which the evolu-

tion of the patient status is described depending on

the actions taken by the physicians. For now, it can be

done only manually by saving into the database some

relevant patient data and, then, changing them on-the-

ﬂy. Nevertheless, we believe that integrating some

formalism automatically describing the patient’s evo-

lution depending on environmental conditions and

ventilation choices, such as Markov Decision Pro-

cesses (Lakkaraju and Rudin, 2016), or including an

AI component, is feasible and worthwhile. Another

future work direction is related to the expandability

of the lung simulator on which our Android app is

based. Indeed, it would be very appreciated by physi-

cians to have different patient models, with different

granularities, and modeling patients having different

diseases. We believe that it can be easily done by ex-

ploiting the INSPIRE framework, which is on the ba-

sis of Ventilation App and supports the simulation

of complex circuits. Finally, more ventilation modes

beyond PCV can be implemented, in order to let new

physicians train under disparate patient conditions.

ACKNOWLEDGEMENTS

This work has been partially funded by

PNRR - ANTHEM (AdvaNced Technologies

for Human-centrEd Medicine) - Grant PNC0000003

– CUP: B53C22006700001 - Spoke 1 - Pilot 1.4.

We would like to thank Eleonora Vitali for the

preliminary work done for this project.

REFERENCES

Abba, A., Accorsi, C., et al. (2021). The novel mechan-

ical ventilator milano for the COVID-19 pandemic.

Physics of Fluids, 33(3):037122.

Bombarda, A., Bonfanti, S., Galbiati, C., Gargantini, A.,

Pelliccione, P., Riccobene, E., and Wada, M. (2021).

Lessons learned from the development of a mechani-

cal ventilator for COVID-19. In 2021 IEEE 32nd In-

ternational Symposium on Software Reliability Engi-

neering (ISSRE). IEEE.

Bombarda, A., Bonfanti, S., Galbiati, C., Gargantini, A.,

Pelliccione, P., Riccobene, E., and Wada, M. (2022).

Guidelines for the development of a critical software

under emergency. Information and Software Technol-

ogy, 152:107061.

Couto, T. B., Farhat, S. C., et al. (2015). High-ﬁdelity

simulation versus case-based discussion for teaching

medical students in brazil about pediatric emergen-

cies. Clinics, 70(6):393–399.

Crowe, M. E., Hayes, C. T., and Hassan, Z.-U. (2013). Us-

ing software-based simulation for resident physician

training in the management of temporary pacemakers.

Simulation in Healthcare: The Journal of the Society

for Simulation in Healthcare, 8(2):109–113.

Fern

andez, J., Miguelena, D., Mulett, H., Godoy, J., and

Martin

on-Torres, F. (2013). Adaptive support ventila-

tion: State of the art review. Indian Journal of Critical

Care Medicine, 17(1):16–22.

IUCPQ (2023). VentilO Android App. https://play.google.

com/store/apps/details?id=ca.qc.iucpq.ventilo.

Kunkler, K. (2006). The role of medical simulation:

an overview. The International Journal of Medical

Robotics and Computer Assisted Surgery, 2(3).

L. E. Wood, M. M. Picard, M. K. (2018). App review: The

radiology assistant 2.0. Journal of Digital Imaging.

Lakkaraju, H. and Rudin, C. (2016). Learning cost-effective

treatment regimes using markov decision processes.

Markossian, T. W., Boyda, J., et al. (2021). A mobile

app to support self-management of chronic kidney

disease: Development study. JMIR Human Factors,

8(4):e29197.

OpenPediatrics (2023). Openpediatrics. https://www.

openpediatrics.org/simulators.

Pearce, J. M. (2020). A review of open source

ventilators for COVID-19 and future pandemics.

F1000Research, 9:218.

e, V., Cattarino, S., et al. (2016). Outcomes of a virtual-

reality simulator-training programme on basic surgi-

cal skills in robot-assisted laparoscopic surgery. The

International Journal of Medical Robotics and Com-

puter Assisted Surgery, 13(2):e1740.

Reznek, M. (2002). Virtual reality and simulation: Training

the future emergency physician. Academic Emergency

Medicine, 9(1):78–87.

Schwid, H. A. (2001). Components of an effective medical

simulation software solution. Simulation & Gaming,

32(2):240–249.

Takashina, T., Shimizu, M., and Katayama, H. (1997).

A new cardiology patient simulator. Cardiology,

88(5):408–413.

Toderis, L., Vo, A., et al. (2022). Development of a mobile

training app to assist radiographers’ diagnostic assess-

ments. Health Informatics Journal, 28(2).

TruCorp (2023). Truvent app. https://trucorp.com/product/

truvent-app/.

van Diepen, A., Bakkes, T. H. G. F., et al. (2021). A model-

based approach to generating annotated pressure sup-

port waveforms. Journal of Clinical Monitoring and

Computing.

Ventsim (2023). Ventsim. https://ventsim.cc/#/.

Wallace, S., Clark, M., and White, J. (2012). ‘it's on my

iPhone’: attitudes to the use of mobile computing de-

vices in medical education, a mixed-methods study.

BMJ Open, 2(4):e001099.

An Android App for Training New Doctors in Mechanical Ventilation

369