Page 81 - AIH-1-4

Artificial Intelligence in Health Robotics and Vivaldi AI for ALS assessment

A B

Figure 1. (A and B) Schematic representation and key features of Sanbot Elf. Image created by author.

and activities. Control over Sanbot is facilitated by seven With these features, the Sanbot Elf is a versatile service
modules or managers accessible through API libraries, robot designed to fulfill multiple roles, particularly in
encompassing functionalities ranging from voice and healthcare settings. At hospitals, it functions adeptly as a
hardware control to multimedia and motion control. receptionist, warmly welcoming and assisting visitors.
In developing the Sanbot application, client-server It excels in providing comprehensive information about
architecture was adopted. The client component, written the facility, guiding visitors to various departments, and
17
in Java for Android, interacts with users and collects data ensuring smooth navigation throughout the premises.
to personalize user-robot interactions. The server, also Within hospital departments, the Sanbot Elf engages with
coded in Java and hosted on a dedicated machine, acts patients in a compassionate and supportive manner. It
as the central processing unit or “brain” of the robot in enhances patient experience by offering entertainment
the cloud. The architecture comprises various services options such as music and videos and providing
including emotion and face recognition, user modeling, companionship during periods of loneliness or anxiety. In
and speech-to-text functionality. addition, the robot enhances security measures by actively
monitoring its surroundings and promptly notifying staff
The server orchestrates communication between of any unusual situations or emergencies that may arise.
different software components, each offering distinct Moreover, the Sanbot Elf facilitates telemedicine sessions by
services such as emotion recognition using Affectiva seamlessly connecting patients with healthcare providers
libraries, face recognition with the “FaceRecognition.py” remotely, thereby enhancing access to healthcare services.
library, and speech-to-text conversion through the Wit
18
service. Given the diverse languages and libraries involved, 3. Related works
13
a central Java server was devised to manage requests and Assistive robotics are gaining significant attention for
data storage in a central database, facilitating seamless their potential to enhance the quality of life for various
integration and operation. Communication occurs between patients, especially those with neurological disorders
the client and the Java server, as well as between the Java such as Alzheimer’s disease, cognitive impairments,
server and internal and external components. neuromuscular diseases, spinal cord injuries, to name a
The Sanbot application commences with a natural few. These conditions often lead to declines in cognitive
language dialogue aimed at user profiling, encompassing and motor functions, severely impacting daily living
factors such as facial recognition, name, gender, age, activities. By improving the management of these robotic
interests, and mood. This information is utilized by systems, it is possible to alleviate the burden on physicians,
the user modeling component to tailor subsequent caregivers, and family members. These robots offer a range
interactions, enabling Sanbot to adapt its behavior, offer of functions, including connecting patients with distant
recommendations, and encourage positive behaviors family members, providing companionship, promoting
through persuasive techniques. health, and assisting with daily tasks. Furthermore, Gao

Volume 1 Issue 4 (2024) 75 doi: 10.36922/aih.3732