Energy expenditure estimation based on artificial intelligence and microservice architecture

Energy expenditure estimation based on artificial

intelligence and microservice architecture

Hieu Trung Huynh1,2

1

Industrial University of Ho Chi Minh City

2

Vietnamese-German University

[email protected]

Ho Dac Quan

Industrial University of Ho Chi Minh City

[email protected]

ABSTRACT

Nutritional status plays an important role in not only pregnancy

outcomes but also neonatal health. One of efficient techniques to

control the nutritional status is to estimate the energy expenditure.

There are some approaches for estimating energy expenditure.

However, they have limitations including high cost, relative

complexity, trained personnel requirements, or locality. This study

investigates in a system for data collection and analysis (IoH￾Internet of Health) developing based on microservice architecture,

and its application for energy expenditure estimation. The

proposed system has a good ability to scale and integrate with

other systems; the energy expenditure estimation is performed by

using artificial intelligence. The experimental results have shown

the promising results of the proposed system.

CCS Concepts

• Computing methodologies•Applied computing

Keywords

data collection; visualization; expenditure energy estimation; IoH

system; healthcare.

1. INTRODUCTION

Developing the systems for data collection and analysis are

attracting researchers on the world [1-10]. Several systems have

been proposed [3,5,7,10-13]. However, the demand for scalability

and integration still has challenges. The scalability allows a

growing system of services available to patients while they are

being evolved. One of approaches to deal with scalability is that

the system architecture must be modular and flexible.

Microservice is one of efficient architectures for this issue.

Microservice architecture emphasizes modular, lightweight

services with a high degree of cohesion. Unlike monolithic

applications where tightly integrated components implement the

application’s functionality and changing requirements affect the

system, microservices can be developed, deployed and scaled

independently of other services that make up the system. The

microservice architecture allows to split a single application into a

set of small services, each running its own process.

Communicating among microservices may be through well￾defined interfaces and standard lightweight protocols such as

gRPC or HTTP and they do not need to use the same platform or

development languages. These properties increase the modular

and continuously growing abilities to support the needs of patients.

They allow to add new functionalities addressing new

requirements as new microservices. Some approaches based on

the microservice platform have been proposed [14-16]. Richard

Hill et al. [14] introduced the microservices in Internet of Things

(IoT) and using an example in the community health care domain.

A microservice based platform for the monitoring of health￾related data via activity trackers has been proposed by Obrien et al.

[15]. Recently, Surya Roca et al. has proposed an approach using

microservice platform for chatbot architecture for chronic patient

support [16]. Although there are implementations, the

microservice model is still attracting researchers, especially in

healthcare systems.

Besides scalability, the ability of standard data sharing in e-health

environment has also been addressing. It allows to integrate the

gathered data relating to the patient such as Electronic or Personal

Health Record (EHR/PHR) into other bigger structures. This

problem has been addressed by Fast Healthcare Interoperability

Resources (FHIR) developed by HL7[17]. In the FHIR standard,

the list of data models representing a wide range of healthcare

related features, which include both clinical and administrative, is

defined. Instances of these data models are named resources,

which can be used to store or exchange the data using different

serialization formats including XML or JSON. Another interesting

point of resources is ability of integration into working systems

which make them suitable for using in a wide variety of contexts

[18].

In this paper, we introduce a system for data collection and

analysis, named IoH (Internet of Health). This system is

developed based on microservice platform, which allows to scale

and integrate with other systems easily. The IoH system is used to

estimate energy expenditure corresponding to each woman with

gestational diabetes mellitus.

Regarding energy expenditure estimation, the nutritional status

may have substantial impacts on not only pregnancy outcomes but

also neonatal health. Determining nutritional supply is mainly

affected by energy expenditure considering the physical activity

and specific health conditions. Several methods for energy

expenditure measurement were proposed which include indirect

calorimetry (IC), direct calorimetry (DC), doubly labeled water

(DLW), bioelectrical impedance analysis (BIA), and predictive

equations [19,20]. These methods can provide good results;

however, they have several limitations. Their results may be

affected by several factors including hydration status of subject,

diet, physical activities, diuretics use, age, menstrual period,

ethnic group, body shape, or health and nutrition condition [21,22].

Permission to make digital or hard copies of all or part of this work for

personal or classroom use is granted without fee provided that copies are

not made or distributed for profit or commercial advantage and that copies

bear this notice and the full citation on the first page. Copyrights for

components of this work owned by others than ACM must be honored.

Abstracting with credit is permitted. To copy otherwise, or republish, to

post on servers or to redistribute to lists, requires prior specific permission

and/or a fee. Request permissions from [email protected].

ICMLSC 2020, January 17–19, 2020, Haiphong City, Viet Nam

© 2020 Association for Computing Machinery.

ACM ISBN 978-1-4503-7631-0/20/01…$15.00

https://doi.org/10.1145/ 3380688.3380715

159