8
IEEE TRANSACTIONS ON NEURAL NETWORKS, VOL. 16, NO. 6, NOVEMBER 2005 1393 NDRAM: Nonlinear Dynamic Recurrent Associative Memory for Learning Bipolar and Nonbipolar Correlated Patterns Sylvain Chartier, Member, IEEE and Robert Proulx, Senior Member, IEEE Abstract—This paper presents a new unsupervised attractor neural network, which, contrary to optimal linear associative memory models, is able to develop nonbipolar attractors as well as bipolar attractors. Moreover, the model is able to develop less spurious attractors and has a better recall performance under random noise than any other Hopfield type neural network. Those performances are obtained by a simple Hebbian/anti-Hebbian online learning rule that directly incorporates feedback from a specific nonlinear transmission rule. Several computer simulations show the model’s distinguishing properties. Index Terms—Associative memory, dynamic model, neural net- work, unsupervised learning. I. INTRODUCTION A TTRACTOR neural networks (e.g., [1] and [2]) define a class of formal models that are usually used as autoasso- ciative memory. The key mechanism common to all attractor neural networks is the presence of a feedback loop. Feedback enables a given network to shift progressively from an initial pattern toward an invariable state, namely, an attractor. If the model is properly trained, then the attractors should correspond to the learned patterns. Thus, the fundamental question is: How well can a given network develop good attractors through learning of proper patterns? Learning in attractor neural networks is usually carried out by a Hebbian type algorithm. At first, simple Hebbian algo- rithms were proposed [1], [2]. However, such networks had the problem of poor storage capacity, unconstrained weight matrix growth, and difficulties in learning correlated prototypes [3]. To overcome those difficulties, different learning algorithms have been proposed. The most popular solution uses a weight matrix that converges to an optimal linear associative memory (OLAM) base on the pseudoinverse. The pseudoinverse of a ma- trix [4] was proposed as a neural network learning algorithm by Kohonen [5] and was applied to a Hopfield network by Per- sonnaz and Guyon [6] and Kanter and Sompolinsky [3]. How- ever, the pseudoinverse algorithm is neither local nor iterative. Many authors proposed locally implemented learning rules that converge toward optimal linear associative memory [7]–[9] or Manuscript received January 10, 2004; revised July 5, 2004. S. Chartier was with Université du Québec à Montréal, Montréal QC H3C3P8, Canada. He is now with the Department of Psychology, Université du Québec en Outaouais, Gatienau QC J8X3X7, Canada (e-mail: chartier.sylvain@ courrier.uqam.ca). R. Proulx is with the Faculté des Sciences Humaines, Université du Québec à Montréal, Montréal QC H3C3P8, Canada (e-mail: [email protected]). Digital Object Identifier 10.1109/TNN.2005.852861 at scaling parameter [10], [11] (see [12] for a comparative anal- ysis of an optimal projection model in a Hopfield type network). Although those various models have a better performance than the simple Hebbian algorithm, they nevertheless have the problem of spurious attractors and lack the capacity to develop nonbipolar attractors. In all previous models, the output is bound in a hypercube that limits the unit’s values to 1 or 1. More- over, Vidyasagar [13] demonstrates that Hopfield type networks using a step function can only develop stable attractors at hyper- cube corners. Consequently, those models typically develop ex- treme behavior that restricts its cognitive explanation. A more powerful model would be able to develop attractors anywhere within a hypercube quadrant instead of only at its extremities. To accomplish this, researchers have used a different type of transmission rule using multiple limit output function [14], [15]. Although this solution yields good results, it does so with an in- crease in the learning rule complexity. Moreover, the proposed model is more sensitive to noise than its binary counterpart. In this paper, we introduce a new attractor neural network model that greatly reduces the number of spurious attractors and therefore has a better performance compared to the models men- tioned previously. In addition, this new model is able to learn and recall nonbipolar attractors without any special coding or an increase in the learning rule complexity. Moreover, the model is able to develop real value attractors without using a mul- tiple threshold transmission rule. We will start by presenting the model and its properties, followed with simulation results. II. PROPERTIES OF NDRAM As with any artificial neural network model, nonlinear dy- namic recurrent associative memory (NDRAM) is entirely de- scribed by its architecture, its transmission rule, and its learning rule. The network architecture is illustrated in Fig. 1. As we can see, the model is autoassociative and recurrent like general Hop- field models. Learning in this model is simply based on Hebbian learning with an added correction term also named anti-Hebbian [16]. The following equation describes the learning rule: (1) where represents the initial bipolar input vector, the weight matrix, the value of the state vector after iterations, and the general learning parameter. The learning rule is a sim- plification of previous iterative optimal projection learning rules in which a combination of Hebbian and corrector factors makes 1045-9227/$20.00 © 2005 IEEE

Quality of care and midwifery services to meet the needs of women and newborns

  • Upload
    dt

  • View
    212

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Quality of care and midwifery services to meet the needs of women and newborns

Quality of care and midwifery services to meetthe needs of women and newbornsF McConville,a DT Lavenderb

a Department of Maternal, Newborn, Child and Adolescent Health, World Health Organization, Geneva, Switzerland b School of Nursing,

Midwifery and Social Work, University of Manchester, Manchester, UK

Correspondence: F McConville, Technical Officer, Midwifery, WHO Policy, Planning and Programmes, Department of Maternal, Newborn,

Child and Adolescent Health, World Health Organization, 20 Avenue Appia, Geneva 27, Switzerland. Email [email protected]

Accepted 28 January 2014.

Please cite this paper as: McConville F, Lavender DT. Quality of care and midwifery services to meet the needs of women and newborns. BJOG 2014; 121

(Suppl. 4): 8–10.

Introduction

Globally, maternal mortality has declined by 47% (from

400 to 210 maternal deaths per 100 000 live births) over

the last two decades (1990–2010), but considerably greater

progress will be needed to reach Millennium Development

Goal 5, especially in countries experiencing conflict and

high HIV incidence.1 There has been a corresponding pro-

motion of childbirth in facilities in low-income countries,

and access to skilled birth attendants (SBAs) in health

facilities has risen from 55% in 1990 to 66% in 2011. Yet,

as the Millennium Development Goals Report 2013 points

out, 46 million of the 135 million women who had live

births in 2011 delivered alone or without adequate quality

of care.

Improving the quality of care has dominated maternity

strategies in the last decade, motivated by the desire to

offer care that not only leads to improved outcomes, but

also offers a good experience for women and their families.

Multiple definitions have attempted to conceptualise the

various attributes of quality of care, most of which have

core elements. The United Nations (UN) Committee on

Economic, Social and Cultural Rights states that care

should be ‘Available, Accessible, Acceptable and of good

Quality’.2 However, care should also be provided by com-

petent, respectful practitioners, who are not only able to

carry out specific tasks, but who also have the knowledge

to support their practice. In relation to midwifery care,

providing a good quality service also needs to be viewed in

the context of wider reproductive health services.

Quality of care is a priority for the World Health

Organization (WHO). Globally, the focus on quality (not

just coverage) is essential because quality is critical for

impact. Under the UN Secretary-General’s Global Strategy

for Women’s and Children’s Health, the Commission on

Information and Accountability (CoIA) called for a process

to ensure global reporting, oversight and accountability on

women’s and children’s health. The CoIA recommended

the establishing of an Independent Expert Review Group

(IERG).3 At the annual World Health Assembly at WHO

headquarters in Geneva, the 197 UN Member States debate

and agree global health priorities. Central to the future

direction of WHO, and the ongoing post-2015 discussion,

is Universal Health Coverage (UHC) and the message of

equity of access for all.4 Underpinning UHC is the need for

quality of care and the unique role of community and

facility-based midwifery services. This brings an unprece-

dented opportunity to bring together UHC, quality of care

and the unique role of community-based and facility-based

midwifery services in providing universal access to

improved maternal and newborn health.

Skilled care at birth

We should not confuse ‘midwives’ with ‘midwifery’—they

are not the same. Midwifery is what matters to women

and newborns. A midwife is the professional best qualified

to provide midwifery, but others (nurses, doctors) also

provide aspects of midwifery, varying across settings.

Where the practice of midwives is constrained, or where

there are no midwives, terms lose their meaning. In 2004,

WHO–International Confederation of Midwives (ICM)–International Federation of Gynecology and Obstetrics

(FIGO) developed a Joint Statement, which defined the

SBA and agreed this to be ‘a midwife, nurse, doctor’, or

other care provider. The 2004 Joint Statement clearly sets

8 ª 2014 Royal College of Obstetricians and Gynaecologists

DOI: 10.1111/1471-0528.12799

www.bjog.orgCommentary

Page 2: Quality of care and midwifery services to meet the needs of women and newborns

out that an SBA is someone who has (at a minimum) 23

‘core’ midwifery skills (ICM definition).5 In low-income

countries where civil registration and vital statistics are

lacking, access to an SBA (one of several proxy measures

for reduction in maternal mortality) is measured through

demographic health surveys. These surveys ask women

whether or not they gave birth in a facility, and if so

whether a midwife, nurse or doctor was with her during

childbirth. Although this indicator has enabled the tracking

of increased births at the health facility level, there are two

major concerns: how does a woman know the profession

of the persons who supported her in childbirth, and how

do we know that SBAs in each setting have the 23 ‘core’

skills and abilities? A recent study in Gujarat, India, for

example, showed that degree and diploma nurse-midwives

are graduating with virtually no midwifery experience

because they have been unable to gain practical training

experience in the critical 24 hours around childbirth.6 So,

are women and newborns really getting ‘skilled’ care? We

think they deserve better.

The global agenda for improvingquality of care

The good news is that governments, UN agencies, donors

and advocates are engaged in a series of global initiatives that

aim to improve the quality of midwifery services. The second

Global Midwifery Symposium (May 2013 in Kuala Lumpur)

united a broad-based constituency and identified seven pri-

orities to strengthen: investment; education; deployment;

being woman-centred; regulation; support to midwives and

nurses associations; and the evidence base to justify increased

investment.7 The UN Population Fund (UNFPA), WHO

and the ICM are co-chairing a process to develop the second

State of the World’s Midwifery (SOWMy) Report, 2014.

Driven by UNFPA, the 2011 SOWMy Report was remarkable

as it was the first time that the situation of midwifery in

low-income countries had been summarised.8 2014 will also

bring the much anticipated Lancet Series on Midwifery

(LSM). The LSM will set out a new framework for maternal

and newborn health services, based on the evidence of what

women and newborns need, demonstrating the value of

quality midwifery services in terms of outcomes and

women’s perceptions of satisfaction. The LSM will call for

significantly increased investment in research on the impact

of midwifery services. 2014 saw the launch of ‘Every New-

born’ Action Plan.9 The ICM is engaged in the process and

midwives are part of global consultations that are addressing

the ‘how to’ improve quality of essential newborn care

services at home and in facilities. To ensure that the energy

and momentum continue, we will be reunited in common

purpose at the 30th Triennial Congress of the ICM in June

2014.

Improving the quality of midwiferycare

To assist ministries of health, WHO is at the early stages of

developing Guidance on quality of care in midwifery services

through the Availability, Accessibility, Acceptability and

Quality (AAAQ) lens. A preliminary meeting with experts

at WHO headquarters agreed the need for multiple stake-

holders to recognise the ‘core’ midwifery skills needed by

all SBAs. This is to ensure the availability of quality of care

for the estimated 85% of women experiencing normal

childbirth, as well as the 15% of women undergoing emer-

gency procedures—for example a woman needing a caesar-

ean section still requires the essential care that midwives

provide, and the baby will also require essential newborn

care. To ensure availability of SBAs, the guidance will help

countries to benchmark for, and invest in, the right teams

of SBAs (midwives, nurses, obstetricians, anaesthetists and

paediatricians, others) to ensure that they are in the right

place at the right time with the right skills.

How should midwifery services be best organised to

meet the needs of women and newborns in the most eco-

nomic model? Evidence from a recent Cochrane systematic

review of midwife-led care versus other models of mater-

nity care shows that women and newborns who receive

midwife-led care experience positive maternal outcomes

with fewer interventions, and that there is no statistical dif-

ference between midwifery-led and other models of care in

fetal loss or neonatal death.10

Midwifery services need to be acceptable to women as

well as to those who provide the care. Together, institu-

tions, doctors, nurses and midwives must end the well-

documented disrespect and abuse of women who come to

give birth in facilities.11 Recent research in an area of

Kenya shows that disrespectful and abusive care is the sec-

ond most common reason (after inaccessibility) why

women will choose to avoid a health facility at the time of

childbirth. Women health workers providing 24-hour mid-

wifery care may themselves face considerable challenges.

WHO is carrying out a systematic mapping of barriers to

women working 24-hour shifts in maternal and newborn

care. These include professional barriers (lack of supplies,

equipment etc.) as well as context specific, sociocultural

and economic barriers. A systematic mapping of interven-

tions, and qualitative analysis, will follow. This work origi-

nated at the Women Deliver Conference 2013, where

WHO and ICM jointly hosted a session in which the reality

of midwives’ lives was explored. Speakers from Afghanistan,

Nepal and Papua New Guinea12 highlighted some distress-

ing issues, encapsulated in emerging research on ‘moral

distress’. It is time to acknowledge the physically insecure,

sometimes violent, and all too often economically crippling,

environment in which many women are working while try-

9ª 2014 Royal College of Obstetricians and Gynaecologists

Quality of care midwifery

Page 3: Quality of care and midwifery services to meet the needs of women and newborns

ing to provide care for other women. We must think

beyond training because it is essential to address the all too

frequent gender-based, negative institutional hierarchies

that undermine professionalism and prevent women pro-

viders from giving the quality of humanised care that

women, newborns and families deserve.

The provision of quality of care must not become a simple

mantra or abstract vision; it is a vitally important component

in interventions that save lives. As we move forward we need

to engage with multiple stakeholders to ensure that quality of

care remains high on the maternity agenda. We also need to

ensure that we use appropriate ways to evaluate the care pro-

vided; assessment should be based on pre-defined expecta-

tions and from the perspective of users and providers. 2014

provides an opportunity to really make a difference to the

lives of women and their families; working together to

improve quality of care is the way we can do this.

Disclosure of interestsThe authors alone are responsible for the views expressed

in this commentary and they do not necessarily represent

the views, decisions or policies of the institutions with

which they are affiliated.

FundingThe author received no funding for this paper.&

References

1 United Nations. The Millennium Development Goals Report 2013,

New York: United Nations; 2013. [www.undp.org/content/dam/

undp/library/MDG/english/mdg-report-2013-english.pdf].

2 Economic and Social Council, Committee on Economic, Social and

Cultural Rights (2000). The right to the highest attainable standard

of health (article 12 of the International Covenant on Economic,

Social and Cultural Rights). UN, 22nd Session, Geneva, 25 April - 12

May 2000, General Comment No.14.

3 World Health Organization. Keeping promises, measuring results:

Commission on information and accountability for Women‘s and

Children‘s Health. Geneva: WHO; 2011 [http://www.every

womaneverychild.org/images/content/files/accountability_commission

/final_report/Final_EN_Web.pdf].

4 WHO (2013). 10 facts on Universal Health Coverage. [www.who.int/

features/factfiles/universal_health_coverage/en/index.html].

5 World Health Organization. Making pregnancy safer: the critical role

of the skilled attendant: a joint statement by WHO, ICM and FIGO.

Geneva: WHO; 2004 [www.who.int/maternal_child_adolescent/

documents/9241591692/en/index.html].

6 Sharma B, Ramani KV, Hildingson I, Kyllile C. Confidence in selected

midwifery skills: self-assessment by students just before graduation

in one province of India. Presentation given by Bharati Sharma at

the “Developing WHO Guidance in Quality of Care in Midwifery

Services” meeting, October 23–24, 2013.

7 The Second Global Midwifery Symposium. Strengthening quality

midwifery care. Making strides, addressing challenges. May 26–27,

Kuala Lumpur, Malaysia.

8 UNFPA. State of the World’s Midwifery Report 2011: Delivering

Health, Saving Lives. New York: UNFPA; 2011 [www.unfpa.

org/sowmy/report/home.html].

9 Healthy Newborn Network (2013) Every Newborn Action Plan

(ENAP) [http://www.everynewborn.org/every-newborn-action-plan/].

10 Soltani H, Sandall J. Organisation of maternity care and choices of

mode of birth: a worldwide view. Midwifery 2012;28:146–9.

11 Browser D, Hill K. Exploring Evidence for Disrespect and Abuse in

Facility-based Childbirth. Report of a Landscape Analysis. USAID

TRAction Project, Harvard School of Public Health, September 20

2010. [www.coregroup.org/storage/documents/elluminates/Respectful_

Care_at_Birth_9-20-101_Final.pdf].

12 Brodie P. ‘Midwifing the midwives’: addressing the empowerment,

safety of, and respect for, the world’s midwives. Midwifery 2013;

29:1075–6.

10 ª 2014 Royal College of Obstetricians and Gynaecologists

McConville et al.