A Model Oriented Approach for Managing Traceability of Biological

Samples and Tests of Patients in Assisted Reproduction Clinics

L. Morales-Trujillo

, V. Cid de la Paz Furest

, J. G. Enríquez

and José Navarro

Computer Languages and Systems Department, University of Seville, Avenida Reina Mercedes, s/n 41010, Sevilla, Spain

Assisted Reproduction Clinic - Inebir, Hospital Victoria Eugenia, Avenida la Cruz Roja, 1, 41009, Sevilla, Spain

Keywords: Traceability, Control Process, Assisted Reproduction.

Abstract: Assisted reproduction has become a service that more and more people access. Current problems such as the

delay in the age of motherhood, single-parent couples, etc. they have proliferated the options and the different

treatments that are put at the service of society. A fundamental part of these processes lies in the work of

laboratories, in the samples that are handled in clinical processes and then be implanted in future mothers.

The management of the samples is a critical aspect that requires all the opportune mechanisms that guarantee

the traceability of said samples, avoiding fatal errors. The correct identification, monitoring and control of

them is a fundamental aspect and of special relevance. However, the systems currently offered to clinics

present important problems. On the one hand, they offer little security, are very expensive or very independent

of a specific provider, so that the traceability system cannot be connected to the hospital central management

system, or they are very intrusive control systems in the daily work of the laboratory technicians. In this paper

a software solution based on the definition of automatic models and protocols is proposed. It includes the

appropriate devices, to manage the traceability of the samples in parallel to the work of the laboratory

technician.

1 INTRODUCTION

For some years now, assisted reproduction techniques

have been part of the reproductive history of many

couples. According to the most extensive

epidemiological studies, infertility affects 15% of the

population of reproductive age in Western countries,

that is, one in six couples, and experiences an

increasing evolution (País, 2017).

Spain is a leader in Europe in assisted

reproduction techniques, with a total of 127,809 'in

vitro' fertilization cycles and 38,903 artificial

inseminations in 2015. In 2015, a total of 36,318

children were born in Spain thanks to assisted

reproduction techniques, representing 8,6% of the

more than 420,000 births that occurred that year,

according to the latest data from the Ministry of

Health (Mundo, 2017). Figure 1 and 2 show some

results graphically.

Figure 1: Assisted reproduction techniques, Spain 2015.

Figure 2: Children born in Spain in 2015.

Morales-Trujillo, L., Furest, V., Enríquez, J. and Navarro, J.

A Model Oriented Approach for Managing Traceability of Biological Samples and Tests of Patients in Assisted Reproduction Clinics.

DOI: 10.5220/0006961003010307

In Proceedings of the 14th International Conference on Web Information Systems and Technologies (WEBIST 2018), pages 301-307

ISBN: 978-989-758-324-7

301

However, although this has allowed the centres to

reach technological and service levels "leading", has

also put Spain "in the crosshairs" of many European

scientific societies and groups of other countries,

since we still have a lot for improving in terms of

control mechanisms (BMD, 2016).

Currently there is no control and transparent

system for the professional to verify if the biological

samples and patient tests are right, which prevents

serious errors. This paper marks the challenge of

increasing trust on the part of the patients involved by

ensuring the control and monitoring of the samples

and tests of the patient in such a way that the risk of

error is reduced to minimum levels by proposing a

software solution based on the definition of automatic

models and protocols, which by incorporating the

appropriate devices, control the traceability of the

samples in parallel to the work of the laboratory

technician. This solution must be affordable and

compatible with the standards that govern hospital

management processes and laboratory samples in

assisted reproduction.

2 PROBLEM

Human reproduction laboratories are exposed to

multiple incidents, but one of the most serious is that

which causes the erroneous identification of

biological samples (ovules, sperm and embryos),

being the misidentification a common problem in all

areas of health.

An identification error occurs when a patient is

incorrectly paired with a test, treatment or procedure

and is usually caused by stress, work overload,

multiple interruptions, material errors, among many

other factors.

University Hospital Utrecht has made public the

possible fertilization of oocytes of 26 women with

sperm outside their partner, that is, a score of women

or couples have sired children with the sperm of the

man who was not the indicated. Finally, the center

detected that the pipette that had been used in some

oocyte fertilization procedures was contaminated

with the sperm of another patient (País, 2016).

Certain processes, such as the mixture of ovules

and sperm, and the transfer of embryos to the uterus,

are seen as critical, since they represent "the point of

no return." If an erroneous identification is produced

in assisted reproduction laboratories, it can be go

unnoticed practically in each of the steps of the

process involving gametes and embryos. The final

result will be catastrophic for both the patient, as for

the professional and the clinic, with legal implications

that can lead to sanctions and even, in extreme cases,

at the close of the clinic.

One of the most used solutions to avoid these

possible errors, is the manual double-test protocol

(MDT), defined as the obligation of "double control

performed in all clinical and laboratory procedures"

with the expectation that, if a " operator "makes a

mistake, it will be captured by the other" witness "in

time to be resolved. However, the evidence suggests

that it may not be as safe and effective. The effects of

mechanization of tasks can reduce the effectiveness

of double testimony because the levels of attention

decrease when the same action is carried out

repeatedly by the same person. Therefore, the risk of

error during the double-checking protocol may

increase due to numerous problems, such as process

redundancy, attentional blindness, ambiguous

responsibility, errors in the verification, additional

work overload and increased stress. In addition, the

process of double control produces additional

"paperwork" to an already overwhelming work

environment and entails the duplication of resources

in an already expensive process.

For these reasons, several alternatives have been

developed based on less manual identification

techniques in order to replace most of the steps

required by human witnesses in assisted reproduction

laboratories. The most widespread are:

 Systems based on barcode labels, which,

moreover, are often used in collaboration with

MDT protocols. It consists of identifying the

pipettes with a bar code so that as a process is

executed in the laboratory, the technician uses

a reader of said bar code to collect the life cycle

of the sample. In this regard, it is expected that

the "Single European Code" (SEC) will be

launched soon, which would increase

confidence in the realization of these

techniques in a quality framework (RHA

Professional). However, this code consists of

40 characters, too large to be placed in

cryopreservation devices: Semen, oocytes and

embryos are conserved in devices called straws

or vitrification supports. Even a simplification

is not feasible through the use of 20 characters

(RHA Professional).

 Systems based on silicon bar codes that are

injected directly into the ovules or embryos,

work in a similar way but the identification is

found in the sample itself. This option is a bit

aggressive in the eyes of the donor of the

sample, since a marker is being introduced to

his biological sample, to his possible son.

WEBIST 2018 - 14th International Conference on Web Information Systems and Technologies

302

 Systems based on radio frequency

identification (RFID) technology RFID

systems involve placing an adhesive RFID tag

on all fungibles that contain embryos and

gametes related to a single patient. These labels

are analysed in each station of the process to

identify the patient with whom you are working

at that moment. A digital record keeps track of

the location of each label at each stage of the

overall process and which staff member is

manipulating the samples. Visual and auditory

alarms indicate when an imbalance has

occurred so that it can be corrected promptly.

The RFID system also keeps a record of the

imbalances, so it is possible to review and

analyse the different stages, detecting the steps

with the highest risk in order to reduce the

imbalances accurately. There are already

companies that have developed solutions, with

a high cost, based on RFID, such as Cooper

Surgical Company and its RI Witness.

However, these solutions have little

adaptability to different laboratory concepts,

are not affordable and are closed to the specific

fungibles of the companies, making it difficult

to intercommunicate with the rest of the

hospital management system.

In short, there is currently no control system,

transparent for the professional, to verify if the

biological samples that are being worked on are those

indicated, which prevents serious errors from

occurring.

For all this, our research aims to create a

technological solution that allows to take control of

the work of technicians in assisted reproduction

laboratories when they execute processes that involve

work with samples. This solution should be little or

nothing invasive in the work of the technician,

compatible with the management system and

affordable and adaptable to the different realities of

assisted reproduction clinics. It must also guarantee

that any sample identification must be in accordance

with current regulations and follow the guidelines set

by the Ministry of Health.

3 GOALS

This project marks the challenge of designing a

strategy that reduces the risk of error minimum levels

making use of information technologies as the central

axis of the solution. The objectives and requirements

defined to achieve it are described below.

The main objective of the research is the

definition and implementation of an ICT solution,

which can be integrated in a simple way in the

processes that follow the biological samples to

provide a complete, univocal and safe traceability of

each sample during its life cycle.

In order to carry out this objective, we propose the

next research steps.

 Study the current situation in the identification

and monitoring of samples in assisted

reproduction.

 Define a protocol to incorporate the monitoring

of the life cycle of biological samples and tests

in a non-invasive and safe way in the work of

health personnel.

 Develop a technological solution that allows

executing the protocol in an affordable and

adaptable way to the reality of the different

clinics.

 Validate the solution in a real context.

4 HYPOTHESIS AND SOLUTION

PROPOSAL

The solution to be developed to achieve these

objectives must be composed of the following

aspects:

4.1 Theoretical Framework

For the management of traceability in clinics, which

provides alignment with traceability management

standards and allows the design, development and

application of a theoretical framework, which, among

other things, allows:

 A work methodology for the management of

traceability biological samples and patient tests

based on standards, good practices and

international standards on biological samples

and patient tests that contribute to structure and

order the activities, documents or controls that

are planned and implemented inside a clinic.

 The definition of the life cycle of the

traceability management process focused on

biological samples and patient tests that

includes planning, compliance, evaluation and

the correct adaptation of the standards that have

been contemplated in the proposal.

Stabilization of a standard does not only entail

A Model Oriented Approach for Managing Traceability of Biological Samples and Tests of Patients in Assisted Reproduction Clinics

303

compliance with it, it also implies the

mobilization of the clinic in a process that

ensures its future fulfilment.

 Following the paradigm MDE (Model-Driven

Engineering) will be defined:

 A meta-model of reference that describes

the processes and artefacts necessary to

carry out the entire life cycle of the

management of the traceability of

biological samples and tests of the

patient in the clinic. This metamodel will

serve as a basis to verify the control and

monitoring of all biological samples and

patient tests.

 A series of checking mechanisms

between the reference metamodel and

the metamodel that is instantiated or

executed in the information system.

4.2 Why the MDE Paradigm?

Model-Driven engineering (MDE) is one of the most

deeply rooted paradigms in the area of software

engineering. It focuses on creating and exploiting

domain models, which are conceptual models of all

the issues related to a specific problem. Therefore,

highlight and point out abstract representations of

knowledge and activities that require a particular

application domain, instead of computer concepts.

An important aspect when using MDE is to

guarantee traceability between the generated process

models. This is essential in the context of the proposal

that is made here and that allows maintaining the

identity of a process among all the modules that

guarantee traceability and the possibility of finding

errors in the early stages, thus avoiding irrecoverable

failures.

By ensuring that the traceability between the

different processes that have to be carried out in a

technique of assisted reproduction and avoiding

possible irrecoverable failures during this process, the

levels of error are reduced to minimum levels and

they provide greater security to patients who want to

undergo a process of assisted reproduction.

After explaining why we use MDE in our

proposal, it is important to know what it is.

MDE came up in order to tackle the complexity of

platforms and the inability of third generation

languages to relief this complexity and effectively

express the domain concepts of the problem. This

new paradigm, apart from raising the level of

abstraction, intends to increase automation during the

life cycle of software development.

This paradigm works, as the primary form of

expression, with definitions of models and

transformation rules among these models which

entail the production of other models. Every model

corresponds to a phase of the life cycle and is

generally specified by means of UML modelling

language.

Standardization was necessary in order to

implement this new paradigm in real projects. OMG

presented MDA, which stands for Model-Driven

Architecture (OMG, 2003), as a platform to support

the paradigm of Model-Driven Engineering.

The main ideas of MDA consist in dividing the

specification of the system functionality from its

implementation on a specific technology platform, as

well as control the evolution from abstract models to

implementations. Thus, the degree of automation

usually increases. MDA proposes to base the software

development on models which make transformations

be performed to generate code or another model with

characteristics of a particular technology (or lowest

level of abstraction). As transformations go on, it may

be noticed that the models become more concrete and

the abstract model changes into another one

compatible with a particular technology or platform.

MDA is based on four types of levels or models:

 The CIM level (Computation-Independent

Model) is considered the highest level of

business model and the most abstract level. It

focuses on requirements specification and

intends that anyone who knows the business

and its processes can understand a CIM model,

as this avoids any contact with the specific

system.

 The PIM level (Platform-Independent

Model) represents the business process model

and system structure, without any reference to

the platform on which the application will be

implemented. It is usually the entry point for all

the support tools for MDA.

 The PSM level (Platform-Specific Model)

specifically relates to the platform where the

system will be implemented, for example, with

operating systems, programming languages or

middleware platforms, among others.

 Finally, the Code level refers to the codification

and suitable implementation of the system.

Figure 3 (Koch, 2006) represents a diagram with

the adaptation of the MDA standard in Web

development.

WEBIST 2018 - 14th International Conference on Web Information Systems and Technologies

304

Figure 3: Model-Driven Web Engineering.

In this context, a set of metamodels in the CIM

level is given which are requirements models. These

models allow information requirements to be

captured.

Analytical models are obtained systematically by

means of the transformation CIM-to-PIM: content

model, navigation model, presentation model, and

some others. In addition, PIM level allows the

application of some transformations (PIM-to-PIM) in

order to get design models. Subsequently, models on

the PSM level are obtained by applying

Transformations PIM-to-PSM. Finally, the

application of Transformations PSM-to-Code

generates the system code.

4.3 Support Tool

The proposal of the solution will be supported by a

support tool that allows accessibility from any

position or workplace to the management process of

biological samples and tests of the patient of the clinic

with secure access through user profiles. In addition,

it will act as a document manager of all the

information generated in the process. On the other

hand, a module that allows to represent the clinical

process will be required. This system must be

compatible with the central hospital management

system, be in accordance with the standards and allow

the administrator, or person who defines the process,

to "mark" the points that have to be controlled.

5 VALIDATION

A module that allows to represent clinical processes

and artefacts is required. This system must be

compatible with the central hospital management

system, be in accordance with the standards and allow

the administrator, or person who defines the process,

to even "mark" the most critical points that have to be

controlled.

On the other hand, a second module will be

developed that, by means of a specific device,

automatically, and based on the process itself, make a

control of those control points and refer them to the

central hospital management system, launching an

exception in the case the technician makes an error.

Finally, there will be a module to control and

monitor biological samples and patient tests that will

allow accessibility from any position or workplace to

them and, in addition, act as a document manager of

all the information generated in the process.

For its validation, a case study will be made taking

the Inebir clinic as a reference (Schmidt 2016). Inebir

is an assisted reproduction clinic located in Seville

(Spain), more specifically in the Victoria Eugenia

Hospital. This clinic has a laboratory designed and

built following strict guidelines that result in a work

space with the latest technological advances in the

field of fertilization, where a team of highly qualified

embryologists follows an exhaustive method of work

(García-García, 2017).

Next, 4 activity diagrams are presented where 4 of

the processes that are carried out in the laboratory of

the Inebir clinic are illustrated (Figure 4 to 7). These

Figure 4: Preparation and extraction of ovules.

A Model Oriented Approach for Managing Traceability of Biological Samples and Tests of Patients in Assisted Reproduction Clinics

305

diagrams can be visualized, apart from the sequence

of steps followed to carry out the corresponding

process, the actors that are involved in this process

and the places where they are carried out:

The first process that is illustrated is that which

goes from the extraction of oocytes from a patient, to

the conservation of the ovules extracted from said

oocytes.

Afterwards, you can visualize the process that

goes from the collection of semen samples to your

treatment so that an ovule can be fertilized later.

Figure 5: Preparation of semen samples.

The final step to which all the processes described

here are directed is the one that can be visualized in

this diagram, where the fertilization of the ovules and

the subsequent transfer to the uterus of a patient is

described.

Figure 6: Fertilization and transfer.

Finally, the last process shown here is optional. It

is the process that results in the freezing of samples.

Figure 7: Freezing of samples.

For all this, it is a perfect scenario to validate the

proposal presented here.

6 EXPECTED RESULT

The following are the results that await the solution

presented here in a process of assisted reproduction:

 Greater control and monitoring of biological

samples and tests of patients reducing the risk

of error to minimum levels, an aspect that is

fundamental and of special relevance to

guarantee safety.

 Increase the confidence of the patient since

reducing the risk of error brings greater peace

of mind.

WEBIST 2018 - 14th International Conference on Web Information Systems and Technologies

306

 Improve the common understanding and

facilitate the continuous improvement of the

clinic in terms of the implementation of

standards, standards and good practices carried

out by third parties.

7 CONCLUSIONS

The paper presents a global view of a model-driven

approach to work with traceability in laboratories.

This approach was obtained for a real necessity in the

industry

The paper introduces the current situation and

analysis the problem that it exists in the field of

human reproduction.

We have, however a lot work to do. Fortunately,

we count with a real clinic support and one of our

main advantage is the real connection with users.

The use of the Model-driven paradigm also results

a suitable and promised idea. In fact, we have used in

other important areas with successful results (García-

García, 2012; Escalona, 2013).

ACKNOWLEDGEMENTS

This research has been supported by Pololas project

(TIN2016-76956-C3-2-R) of the Spanish Ministry of

Economy and Competitiveness.

REFERENCES

País, (periódico el País), «España lidera la reproducción

asistida en Europa,» 2017. [En línea]. Avalaible:

https://elpais.com/elpais/2017/07/04/mamas_papas/14

99166213_758427.html.

EL MUNDO, «El 8,6% de los niños nacen en España

gracias a técnicas de reproducción asistida,» 2017.

Available: http://www.elmundo.es/ciencia-ysalud/

salud/2017/10/11/59ddebb0ca474103188b45a1.html.

BMD (Banco Mundial de Datos) «Tasa de fertilidad, total

(nacimientos por cada mujer), Banco Mundial de

datos,» 2016. Avalaible: https://datos.banco

mundial.org/indicador/SP.DYN.TFRT.IN?view=chart.

EL PAÍS, «Holanda investiga la posible fecundación de 26

mujeres con esperma equivocado,» 2016. Avalaible:

https://elpais.com/internacional/2016/12/29/actualidad

/1483021366_741815.html.

RHA Profesional, «El Código Único Europeo. Toda la

información necesaria para entender su

funcionamiento,» Avalaible: http://www.rhapro

fesional.com/codigo-unico-europeo-ya-puedeconsulta

do-traves-la-plataforma-creada-la-comision-europea/.

Schmidt, D. C. Model-Driven Engineering. IEEE

Computer, Computer Society, vol. 39, no. 2, pp. 25-31,

2006.

Garcia-Garcia, J. A., Enriquez, J. G., Garcia-Borgonon, L.,

Arevalo, C., Morillo, E.: A MDE-based framework to

improve the process management: The EMPOWER

project. In: Proceedings - 2017 IEEE 15th International

Conference on Industrial Informatics, INDIN 2017

(2017).

García-García, J. A., Ortega, M. A., García-Borgoñon, L.,

& Escalona, M. J. (2012, July). NDT-Suite: a model-

based suite for the application of NDT. In International

Conference on Web Engineering (pp. 469-472).

Springer, Berlin, Heidelberg.

Escalona, M. J., García-García, J. A., Mas, F., Oliva, M., &

Del Valle, C. (2013). Applying Model-Driven

Paradigm: CALIPSOneo Experience. In CAiSE

Industrial Track (pp. 25-32).

OMG, 2003. MDA Guide of OMG. Version 1.0.1,

http://www.omg.org/docs/omg/03-06-01.pdf

Koch, N., Zhang, G., Escalona, M. J., 2006. Model

Transformations from Requirements to Web System

Design.

A Model Oriented Approach for Managing Traceability of Biological Samples and Tests of Patients in Assisted Reproduction Clinics

307