Content uploaded by Börge Kordts
Author content
All content in this area was uploaded by Börge Kordts on Jul 10, 2018
Content may be subject to copyright.
Requirements for a System Supporting Patient
Communication in Intensive Care in Germany
Börge Kordts, Jan Patrick Kopetz, Katrin Balzer, Nicole Jochems
University of Lübeck
UzL
Lübeck, Germany
kordts@itm.uni-luebeck.de, kopetz@imis.uni-luebeck.de, katrin.balzer@uksh.de, jochems@imis.uni-luebeck.de
Abstract—Weaning from the mechanical ventilation poses
substantial physical and psychical stress to the patients which is
intensified by the obstruction of verbal communication. Hence,
during the weaning phase, ICU patients often cannot impart their
basic needs adequately. A prolonged healing process, delirium,
and complications are possible consequences. The research project
ACTIVATE aims to develop an interactive system to support
communication and re-orientation in weaning patients and to
allow them early autonomous control of ambient devices. The
system will include an innovative ball-shaped interactive
rehabilitation device (BIRDY), designed for weaning patients
bound to the bed to control the proposed system.
As a result of the development process including two studies,
several workshops and a comprehensive user and context analysis,
non-functional and functional requirements for the overall system,
consisting of BIRDY, the system’s architecture and human-
machine interfaces were determined.
The target group requires a particular focus on usability
aspects addressing patients’ cognitive and physical impairments.
To save nursing staff time resources, the system should function as
automatically as possible. Besides, safety and security by design,
meeting infection control regulations, multilingual system
dialogues and a multimodal presentation of information are
crucial aspects.
Keywords — Intensive Care Unit, Mechanical Ventilation,
Weaning, Human-Computer Interaction, Ambient Computing,
Human Centered Design, Augmentative and Alternative
Communication
I. INTRODUCTION
To treat the most critically ill patients, an intensive care unit
(ICU) is staffed with experienced personnel and characterized
by a high nurse-patient ratio. A common intervention for
patients with life-threatening conditions is mechanical
ventilation and hence, those patients represent a large and
highly vulnerable patient group in intensive care. In 2016 about
425,000 of the 2.1 million reported cases of intensive care
treatment in Germany were mechanically ventilated [1].
When the medical staff decides to remove the mechanical
ventilation to have a patient breathe independently again, the
first step performed is a reduction of the sedation. This
initializes the weaning process where the human body has to re-
adapt to breathing independently from mechanical support. To
facilitate this re-adaption, the ventilation is gradually decreased.
During weaning from mechanical ventilation patients
perceive substantial physical and psychical stress. The patients’
serious conditions, the influence of sedating medication and the
endotracheal intubation renders the patients temporarily
voiceless. This obstruction of verbal communication potentially
intensifies the stress. Adequate communication of even basic
needs can be a major challenge to respective ICU patients.
Possible consequences are a prolonged healing process,
delirium, and complications. Patients lacking communication
ability are facing a higher risk of poorer treatment [2].
Insufficiently treated pain [3], physical symptom burden, fear
as well as feelings of unfamiliarity and identity loss [4] [5] are
often reported strains among mechanically ventilated patients.
On the other hand, effective communication with ventilated
patients has been linked to positive nursing care outcomes [6].
Regular and successful communication between patients and
ICU staff is therefore of paramount importance to foster
patients’ recovery from the critical health conditions and must
be established as early as possible. However, although this need
for early and continuous communication is well-acknowledged
by health professionals, effective methods are lacking to
support this communication and in many reported cases, the
interaction with mechanically ventilated patients is perceived as
onerous [4]. Thus, novel and easy-to-use tools are required to
facilitate communication between health professionals and non-
vocal ICU patients from the very beginning of the weaning
phase.
Information and communication technology (ICT)
underwent a rapid evolvement in recent decades. Especially, the
health technology sector has seen immense innovation in the
past few years. Based on these trends, we aim to develop an
interactive system to support communication and re-orientation
in weaning patients and to allow them early autonomous control
of ambient devices. The system’s design and features are
inspired by methods of augmentative and alternative
communication (AAC) and the concept of successive
information. Our approach involves providing the correct
amount of information at the correct time to prevent
overburdening the patient. The ACTIVATE system will make
use of an innovative ball-shaped interactive rehabilitation
device (BIRDY), designed for weaning patients bound to the
bed to control the proposed system. The system’s concept is
inspired by the paradigm of ambient computing. We combine
several smart devices to create a device ensemble that suits the
users and serves specific use cases. A proposal for the usage of
the ACTIVATE system can be seen in Figure 1.
In this article, we present key requirements for the
ACTIVATE system. A human-centered design (HCD)
approach including two studies, several workshops and a
comprehensive user and context analysis was applied to gather
these requirements.
II. RELATED WORK
Different approaches for AAC strategies to support patient
communication of the voiceless were already summarized [7].
Findings show various low- and high-tech approaches to
overcome the communication barrier.
There is previous work focusing on approaches that don't
require ICT, like pen and paper [8] or illustrated
communication material [9] for instance. Furthermore, one
approach relies on nurse training with low-tech AAC, electronic
AAC, and low-tech tools [10]. Some authors describe the
application of AAC software on standard PCs with eye trackers
and touch screens [11], the use of computers controlled by eye
blinking and/or hand or finger movement [12], the use of tablets
with AAC specialized content and speech synthesis [13–15] as
well as the operation of tablet computers with programmable
speech [16] [17].
The more advanced work is focusing on tablets or tablet-
like devices that are inspired by basic talkers known in AAC
and enhance this basic functionality using modern user
interfaces like gaze control or touch. However, no systems
specifically designed for the ICU context are presented in these
publications, particularly no approaches based on an ambient
system to fulfill the task.
Several requirements for systems supporting patient
communication in intensive care have already been reported
[18]. Recently, based on these insights, an interaction device
and a communication system controlled by the device was
developed [19]. This device can be adapted to the patients’
physical deficits and impairments and uses a vibration motor to
provide feedback. Domain-specific requirements such as
infection control, simple design, suitable content, and limited
motor skills in ICU patients were taken into account by the
design of this device. We adapted requirements for the ambient
system for patient information, communication, and control
targeted by the ACTIVATE project from these findings.
Nevertheless, they do not cover all aspects relevant to this
system.
III. METHOD – HUMAN-CENTERED DESIGN
The ACTIVATE system is planned to be deployed in
clinical practice. Consequently, it is going to be evaluated under
realistic conditions in a clinical study to prove its usability and
to examine its impact on the target groups. To achieve the
acceptance by potential users, namely patients, relatives and
nursing staff, all of their needs must already be considered in
the development process. Additionally, usability factors should
be in focus during the development. As a consequence, we
closely adhere to the HCD process as specified in DIN EN ISO
9241-210 [20] (Figure 2) at each stage of the development
process.
Understanding and specifying the context of use and the
users’ needs and requirements plays a major role in the
development process of interactive systems and particularly
within the HCD process. These requirements are used to
develop design solutions that are formatively evaluated within
the process. Development is finished as soon as the solution
meets the requirements in a summative evaluation. At this
point, the ACTIVATE system can be tested in a clinical study.
Figure 2. The Human-Centered Design process according to DIN EN ISO
9241-210.
Figure 1. A possible setup where a patient interacts with the
ACTIVATE system using BIRDY.
This emphasizes the critical role of requirements in the whole
development process, the part presented within this article.
The requirements for the total system were gathered in three
parallel sub-processes, namely a user and context analysis
(including qualitative interview studies as well as the creation
of personas and problem scenarios), user preference studies,
and finally, workshops with stakeholders to discuss insights and
derive the detailed requirements. These sub-processes are
described in the following.
A. User and Context Analysis
For a better understanding of the user groups, their needs
and the general context of use, we conducted a user and the
context analysis. First, a comprehensive literature search was
performed to identify similar work in the field and gather the
corresponding key insights. The search was focused on socio-
technical systems to support patient communication in intensive
care. Results confirm the need for AAC and emphasize the
demand for novel solutions for patient communication during
the weaning phase. Besides, the search revealed that there is
only limited work on this topic (refer to Related Work).
Another systematic literature search was carried out to
identify patients’ perceptions and experiences recalled by
themselves from the weaning process. The searches yielded one
meta-synthesis [21] underscoring the weaning patients’ largely
unsatisfied communication needs, particularly with regard to
possibilities to express their feelings and symptoms and to
receive information about their situation. Further literature [22]
was included to get additional information about the context
which were complemented and reflected by discussions with
domain experts.
Furthermore, a qualitative study comprising individual
interviews with 16 patients, 16 relatives and 6 medical doctors
as well as three focus group interviews with 26 nurses and other
health professionals was conducted. This study allowed an in-
depth analysis of the patients’ needs from their own and nursing
staffs’ perspective as well as exploration of the staff’s and
relatives’ own needs in the care for weaning patients. Besides,
facilitating factors and potential barriers for the use of the
planned system were identified.
Results of these various information sources were used to
create data-driven personas that represent our target user
groups. They were carefully designed according to chosen key
characteristics, namely (un-)planned hospitalization, degree of
physical, cognitive and behavioral impairments and
disturbances, medical discipline and native speaker-status
(German or other language). Additionally, we modeled a
typical weaning process and used it for the creation of persona-
based problem and solution scenarios.
At the final stage, our procedure resulted in an elaborated
user analysis, a detailed organization analysis, and a task
analysis. The user analysis includes descriptions of user groups
along with their characteristics and personas of different types,
namely primary, negative, served and customer personas. The
organization and the task analysis provided valuable insights
into the daily clinical routine.
B. User Preference Study
To identify key characteristics (namely shape, size, weight,
surface properties and deformability) of BIRDY, the interaction
device designed to control the ACTIVATE system, we
conducted a user study to identify future users’ preferences
regarding different device properties. First, 30 commercially
available objects having design characteristic attributes being
potentially suitable for BIRDY were evaluated by 12
participants in a preliminary study resulting in eight preferred
objects included in our main study.
For the main study, a setting that resembles realistic
conditions in a hospital was created. The participants wore
special gloves simulating swollen hands and reducing hand
mobility. Additionally, they laid in a hospital bed, with the
upper body being elevated by 30° [23]. 40 participants of two
different age groups (20 each), the first group ranging from 18
to 40 years (M=23.45, SD=3.03, 11 females) and the second
group ranging from 58 to 84 years (M=67.25, S=6.6, 12
females), chose their object preferences and told how they
would interact with their favorite object. After a pair-wise
comparison, they ranked their favorite objects regarding
predefined characteristics, namely size, weight, shape surface
properties and deformability, and their overall favorite object
regardless of a fixed characteristic. We created several rankings
based on the pair-wise comparison, preferences regarding fixed
characteristics, and the overall favorite object.
An analysis of the choice and the reason for the decision
[24] as well as an analysis of the first impulse in spontaneous
interaction [25], both with the favored object, were already
published.
C. Workshops
All previous results were discussed and refined in joint
workshops of our project members. The team consists of
experts of various disciplines, namely nursing research, ICU
nursing practice, hospital IT, hardware engineering, software
engineering, psychology, usability, and AAC. The workshops
were conducted to determine concrete requirements and pave
the way for further development.
First, the created personas and problem scenarios were
enhanced and used for potential use cases as well as the
discussion of possible approaches that resulted in solution-
based scenarios describing situations where the system can be
used and promises positive impact on the user groups. Both of
these tools were used for further considerations finally resulting
in requirements.
We conducted several workshops with stakeholders to
define requirements addressing (non-)functional aspects,
technical details as well as design options. Results of previous
work were shared; unresolved aspects were clarified, and the
feasibility of different design options was verified. In total, 20
experts were involved in the workshops. We had one workshop
focusing on the HCD process, eight workshops on technical
details, scenarios and personas, three workshops to consider
preliminary requirements, two workshops to realize safety and
security by design, and one workshop to analyze state of the art
devices for AAC. Additionally, we had seven telephone
conferences to finalize technical details and requirements with
our hardware engineers CogniMed GmbH, who will realize the
interaction device BIRDY. Finally, we found a consensus
among all stakeholders and had our design choices confirmed
by the team.
IV. RESULTS
We identified several barriers and enablers for the use of
systems supporting patient communication in intensive care.
On the barrier side, required time and expertise for the
installation and use as well as a slow system performance were
determined. Other risks are seen in a high effort resulting in
possibly only little benefit and a potentially negative stance
towards digital communication. On the enabler side, an
intuitive and natural operation, stability, and simple usage were
named. A fast and easy assembly and installation, an
uncomplicated preparation for a change of patients and a
possibly high acceptance based on good experiences are seen as
chances.
Our process resulted in non-functional and functional
requirements for the overall system, consisting of BIRDY, the
system’s architecture, and human-machine interfaces, were
determined. Firstly, we describe the non-functional
requirements (NFR).
The fact that weaning patients cannot be expected to learn
complex interactions reinforces the need for an intuitive design,
especially since we are planning an application at an early stage.
Short awareness phases of the target patient group, as well as
potentially lacking experience in controlling smart devices,
demand that interaction and its effect must be immediately clear
(NFR-1). Hence, typical interaction patterns of the target
groups should be taken into account, particularly for the
development of BIRDY. This applies also to other interaction
possibilities, which should be provided to maintain
controllability despite various impairments of possible users
due to critical health conditions, age or intervention. Since the
system is intended to not cause additional nursing resource use
and the staff is not always present, patients should be instructed
by the system itself most of the time (NFR-2). The usage of the
system must not cause any injuries, posing special restrictions
to possible interaction devices, BIRDY included (NFR-3).
The target group requires a special focus on usability
aspects to address cognitive and physical impairments. Hence,
in our development process we follow the seven general
ergonomic principles that are described within DIN ISO 9241
part 110. Firstly, to maintain suitability for the task, cognitive
and physical impairments should be regarded (NFR-4). Users
should not be overstrained and interaction alternatives (for
instance AAC) should be provided to guarantee controllability
despite possible impairments. Displays and audio messages
should be appropriate for patients in such conditions, also
requiring text with large characters and alternative modalities,
particularly for vision- and hearing-impaired patients.
Additionally, reduced manual force should be taken into
account when designing or selecting interaction devices. Next,
a wake-up mode and tutorials should be implemented to ensure
the suitability for learning (NFR-5). Suitability for
individualization (NFR-6) is corresponded by allowing nursing
and other healthcare staff to configure the system’s interaction
interfaces according to the patient’s and their own needs.
Considered configuration options include specific messages,
individual music therapy modes and BIRDY’s vibration or light
intensity. To address user expectations, changes in interaction
possibilities should be avoided or kept minimal, for instance
when an interaction option becomes unusable, perhaps due to a
deterioration of the medical condition. On this matter, self-
descriptiveness should be provided, not only for the human-
computer dialogue but for all system components and ambient
devices (NFR-7). Finally, the system must be error tolerant
(NFR-8), particularly due to the target group and the context of
use.
Referring to the previously described barriers, the system
should function as automatically as possible to save nursing
staff’s time (NFR-9). Besides, safety and security by design
(NFR-10), meeting sanitary regulations (NFR-11), providing
multilingual system dialogues (NFR-12) and a multimodal
presentation of information (NFR-13) are central factors.
Particularly, unnecessary acoustical and optical noise should be
avoided (NFR-14) and it should be possible to deactivate the
system in situations where is it not appropriate (NFR-15). To
meet infection control regulations, objects that are not directly
in contact with the patient (e.g. more than 1.5 m away) must be
designed in such a way that they can be disinfected by wiping
and objects closer must be hermetically sealed (NFR-16).
Following this regulation, all components that are in contact
with the patient, namely BIRDY and potential other interaction
devices, must be sealed. It should be noted that small objects
are, although not required, often immersed into a disinfectant
and should therefore be submergible. To allow for a continuous
operation, replaceability of the components (NFR-17) should
be ensured.
The ACTIVATE system is planned to be applied in German
hospitals and thus, must conform European Union’s General
Data Protection Regulation (EU-GDPR). Hence, all data
communication shall be encrypted (NFR-18) and the system
shall be secured to be resistant against attacks (NFR-19).
Besides, to maintain patient’s privacy, personal information
should be protected from prying eyes where possible (NFR-20).
Yet, it should be noted, that insights by third persons cannot
always be prevented, due to shared rooms and visits of relatives.
Functional requirements (FR) are described in the
following. The system consists of a user interface (UI) for the
patients and an UI for the staff, as well as a system architecture
to integrate all components.
Patients should be able to send messages to the nursing staff
(FR-1) notwithstanding their condition (smart nursing calling
system). Furthermore, the system should provide application
masks for AAC on different topics, like pain or
breathing/mechanical ventilation for instance, also to support
the nursing staff’s assessment and information gathering of
patient’s problems (FR-2). To foster re-orientation, basic
information about date, time, place, weather conditions, and if
available the relevant nursing staff member(s), as well as
possible further information about interventions (for instance
mechanical ventilation) shall be given to the patient (FR-3). The
basic information should be provided visually and during the
wake-up phase acoustically (FR-4). Further information about
interventions should be included in the audio message (FR-5).
Besides, schematic information about future steps in treatment
and process should be given (FR-6).
The wake-up mode should introduce patients to their current
situation and the ACTIVATE system (FR-7). Hence, the wake-
up mode should provide primarily information for the patient.
To avoid overstraining the patient, the functionality should be
reduced during this phase and it should be possible to unlock
further functions according to the patient’s condition.
Particularly during this phase, but also at later stages,
unintended input should be avoided (FR-8), e.g. by using a lock
screen and appropriate tutorials for the wake-up phase and for
later phases.
Due to the situation, feedback plays an important role in the
ICU setting. Hence, acoustic and visual feedback should be
used (FR-9a). Besides, BIRDY should provide tactile feedback
(FR-9b), a function that should be used by the ACTIVATE
system to address the patients.
The staff should be able to configure (FR-10) and control
(FR-11) the patient system (the information given, intervals for
audio messages, the wake-up mode and individual
customizations of the application masks according to the
patient’s needs) and receive information about any interaction
attempts of the patient with BIRDY and the system (FR-12).
Furthermore, the staff should be informed about messages from
patients (FR-13) by the application (smart nursing calling
system, cf. FR-1).
The system architecture should allow for the integration of
BIRDY, potential other interaction devices, smart room
components controlled by the patient and output devices, like
screens and audio devices (FR-14).
To enable patients to manage a feelgood atmosphere, they
should be able to control room components (devices of the
internet of things, like smart lights). This requires
corresponding control possibilities in the patient UI (FR-15) as
well as an integration of the components in the system (FR-16).
Furthermore, the playback of media provided by relatives, like
music files, photos, videos, or audio messages, should be
possible and controlled by patients as well as the staff (FR-17).
V. DISCUSSION
Gathered requirements provide valuable information for the
development of the ACTIVATE system. They can be used to
design and realize further systems for the support of
mechanically ventilated ICU patients during the weaning phase.
During our qualitative study we faced limited capabilities of
the participants to imagine possible sociotechnical solutions for
the communication barrier. To handle this shortcoming, we
implemented additional workshops with domain and
engineering experts to find and discuss potential solutions in
this team.
The conducted user preferences study was limited to
acquirable objects and hence, decisions were made based on the
physical characteristics of these objects and not based on a
combination of single favored attributes. Nevertheless, results
of our study addressing size, weight and first interaction pattern
are comparable to those of other studies [26].
In the future, we plan to evaluate the ACTIVATE system in
a clinical study in the care setting of interest (ICU hospital care)
to examine the system’s feasibility and potential impact on the
target groups. Consequently, the next step is to realize the
system with its components, including BIRDY, based on the
specified requirements.
ACKNOWLEDGMENTS
This work is sponsored by the German Federal Ministry of
Education and Research (BMBF) funded project ACTIVATE
(Code - 16SV7689).
LITERATURE
[1] Statistisches Bundesamt, Gesundheit: Grunddaten der Krankenhäuser
2016. Statistisches Bun desamt (Destatis) , 2017.
[2] C. Handberg and A. K. Voss, “Implementing augmentative and
alternative communication in critical care settings: Perspectives of
healthcare professionals,” J Clin Nurs, vol. 27, no. 1–2, pp. 102–114, Jan.
2018.
[3] T. Bohrer et al., “Wie erleben allgemeinchirurgische Patienten die
Intensivstation? Ergebnisse einer prospektiven Beobachtungsstudie,”
Der Chirurg, vol. 73, no. 5, pp. 443–450, 2002.
[4] A. Abuatiq, “Patients’ and health care providers’ perception of stressors
in the intensive care units,” Dimensions of Critical Care Nursing, vol. 34,
no. 4, pp. 205–214, 2015.
[5] L. Rose, K. N. Dainty, J. Jordan, and B. Blackwood, “Weaning from
mechanical ventilation: a sc opin g review of q ualitative studies,”
American Journal of Critical Care, vol. 23, no. 5, pp. e54–e70, 2014.
[6] M. L . Ni lsen et al., “Nurse and patient interaction behaviors’ effects on
nursing care quality for mechanically ventilated older adults in the ICU,”
Research in gerontological nursing, vol. 7, no. 3, pp. 113–125, 2014.
[7] H. Carruthers, F. Astin, and W. Munro, “Which alternative
communication methods are effective for voiceless patients in Intensive
Care Units? A systematic review,” Intensive and Critical Care Nursing,
vol. 42, pp. 88–96, Oct. 2017.
[8] A. H. El-Soussi, M. M. Elsha fey, S . Y. Othman, and F. A. Abd-Elkader,
“Augmented alternative communication methods in intubated COPD
patients: Does it make difference,” Egyptian Journal of Chest Diseases
and Tuberculosis, vol. 64, no. 1, pp. 21–28, Jan. 2015.
[9] M. Otuzoğlu and A. Karahan, “Determining the effectiveness of
illustrated communication material for communication with intubated
patients at an intensive care unit,” International Journal of Nursing
Practice, vol. 20, no. 5, pp. 490–498, Oct. 2014.
[10] M. B. Happ et al., “Effect of a multi-level intervention on nurse–patient
communication in the intensive care unit: Results of the SPEACS trial,”
Heart & Lung: The Journal of Acute and Critical Care, vol. 43, no. 2, pp.
89–98, Mar. 2014 .
[11] F. Maringelli, N. Brienza , F. Scorrano, F. Grasso, and C. Gregoretti,
“Gaze-controlled, computer-assisted communication in Intensive Care
Unit: ‘speaking through the eyes’.,” Minerva Anestesiol, vol. 79, no. 2,
pp. 165–175, Feb. 2013.
[12] M. A. Miglietta, G. Bochicchio, and T. M. Scalea, “Computer-Assisted
Communication for Critically Ill Patients: A Pilot Study,” Journal of
Trauma and Acute Care Surgery, vol. 57, no. 3, p. 488, Sep. 2004.
[13] M. B. Happ, T. K. Roesch, and K. Garrett, “Electronic voice-output
communication aids for temporarily nonspeaking patients in a medical
intensive care unit: a feasibility study,” Heart & Lung: The Journal of
Acute and Critical Care, v ol. 33, no. 2, pp. 92–101, Mar. 2004.
[14] M. B. Happ, T. K. Roesch, and S. H. Kagan , “Patient communication
following head and neck cancer surgery: a pilot study using electronic
speech-generating devices,” Oncol Nurs Forum, vol. 32, no. 6, pp. 1179–
1187, Nov. 2005.
[15] C. Rodriguez and M. Rowe, “Use of a Speech-Generating Device for
Hospitalized Postoperative Patients With Head and Neck Cancer
Experiencing Speechlessness,” Oncology Nursing Forum, vol. 37, no. 2,
pp. 199–205, Feb. 2010.
[16] C. S. Rodriguez, M. Rowe, B. Koeppel, L. Thomas, M. S. Troche, and G.
Paguio, “Development of a communication in tervention to assist
hospitalized suddenly speechless patients,” Technology and Health Care,
vol. 20, no. 6, pp. 519–530, Jan. 2012.
[17] C. S. Rodriguez, M. Rowe, L. Thomas, J. Shuster, B. Koeppel, and P.
Cairns, “Enhancing the Communication of Suddenly Speechless Critical
Care Patients,” Am J Cr it Care, v ol. 25, no. 3, pp. e40–e47, May 2016.
[18] M. A. Goldberg, L. R. H ochberg, D. Carpenter, J. L. Isenberger, S. O.
Heard, and J. M. Walz, “Principles of Augmentative and Alternative
Communication System Design in the ICU Setting,” 2017.
[19] M. A. Goldberg, L. R. H ochberg, D. Carpenter, J. L. Isenberger, S. O.
Heard, and J. M. Walz, “Testing a Novel Manual Communication System
for Mechanically Ventilated ICU Patients,” 2017.
[20] ISO, “9241-210: 2010. Ergonomics of human system interaction-Part
210: Human-centred design for interactive systems,” International
Standardization Organization (ISO). Switzerland, 2009.
[21] S.-F. Tsay, P.-F. Mu, S. Lin, K.-W. K. Wang, and Y.-C. Chen, “The
experiences of adult ventilator-dependent patients: A meta-synthesis
review,” Nursing & health sciences, vol. 15, no. 4, pp. 525–533, 2013.
[22] G. Marx, E. Muhl, K. Zacharowski, and S. Zeuzem, “Die
Intensivmedizin,” Springer-Verlag, 2014.
[23] L. Wang et al., “Semi-recumbent position versus supine position for the
prevention of ventilator-associated pneumonia in adults requiring
mechanical ve ntila tion,” The Co chrane Database of Systematic Reviews,
no. 1, p. CD009946, Jan. 2016.
[24] S. Burgsmüller, A.-K. Vandereike, J. P. Kopetz, M. Sengpiel, and N.
Jochems, “Study of Desirable Characteristics of a Communication
Device for Intensive Care Patients,” in Student Conference Proceedings
2018, Lübeck, Germany, 2018.
[25] A.-K. Vandereike, S. Burgs müller, J. P. Kopetz, M. Sengpiel, and N.
Jochems, “Interaction Paradigms of a Ball-Shaped Input Device for
Intensive Care Patients,” in Student Conference Proceedings 2018,
Lübeck, Germany, 2018.
[26] G. Perelman, M. Serrano, M. Raynal, C. Picard, M. Derras, and E.
Dubois, “The Roly-Poly Mouse: Designing a Rolling Input Device
Unifying 2D and 3D Interaction,” in Proceed ings of the 33rd Annual
ACM Conference on Human Factors in Computing Systems, New York,
NY, USA, 2015, pp. 327–336.
