• Contact us
• About us
• A-Z topics
• Symptom Checker
• Service Finder
• Subscribe to newsletters
• Linking to us
• Information partners
• Terms of use
• Privacy

Pregnancy, Birth and Baby is funded by the Australian Government and operated by Healthdirect Australia.

Pregnancy, Birth and Baby’s information and advice are developed and managed within a rigorous clinical governance framework.

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

Healthdirect Australia acknowledges the Traditional Owners of Country throughout Australia and their continuing connection to land, sea and community. We pay our respects to the Traditional Owners and to Elders both past and present.

This information is for your general information and use only and is not intended to be used as medical advice and should not be used to diagnose, treat, cure or prevent any medical condition, nor should it be used for therapeutic purposes.

The information is not a substitute for independent professional advice and should not be used as an alternative to professional health care. If you have a particular medical problem, please consult a healthcare professional.

Except as permitted under the Copyright Act 1968, this publication or any part of it may not be reproduced, altered, adapted, stored and/or distributed in any form or by any means without the prior written permission of Healthdirect Australia.

Support this browser is being discontinued for Pregnancy, Birth and Baby

Support for this browser is being discontinued for this site

• Internet Explorer 11 and lower

We currently support Microsoft Edge, Chrome, Firefox and Safari. For more information, please visit the links below:

• Chrome by Google
• Firefox by Mozilla
• Microsoft Edge
• Safari by Apple

You are welcome to continue browsing this site with this browser. Some features, tools or interaction may not work correctly.

Labour pain relief: TENS machine | Labour & birth articles & support

Knowing whether TENS can actually reduce labour pain or is just ‘worth a go’ could help you to know if you want to find a TENS machine. Here’s the lowdown…

What is TENS?

TENS stands for transcutaneous electrical nerve stimulation. It’s a form of pain relief that you can control with a hand-held device.  

How does TENS work?

A TENS machine works by sending mild electrical impulses to sticky pads on your lower back. TENS is thought to work as the electrical pulses stimulate nerves that run to the spinal cord and block the transmission of pain (Doswell et al, 2009). This creates a tingling feeling. You can control the strength of the current yourself (OAA, 2019).

TENS might also work by providing a distraction and restoring a sense of control, both of which may lessen the anxiety that can delay the progress of labour (Doswell et al, 2009).

Does TENS work?

A large number of studies have found that women rated their pain similarly, whether using TENS or not (Doswell et al, 2009). But experts are not totally sure whether TENS is effective because more high-quality research about it is needed (Johnson and Jones, 2016).

Some women find it helpful at the start of labour, particularly for backache. While some use only a TENS machine for pain management during labour, it is common to use some other form of pain relief later on in labour (OAA, 2019).

Is TENS suitable for everyone?

Although the electrical pulses used are very small, some people love it, some hate it. It can't be used at the same time as water (shower, bath or birth pool).

There may be additional considerations or monitoring necessary for the use of a TENS machine if a person has a pacemaker fitted. This can be discussed with your health care provider (Badger, 2017).

What do women say about their experiences of TENS?

While most women who use TENS do so to cope with early labour at home, the research has not looked at whether it helps at this point (Doswell et al, 2009). Despite the lack of evidence for its effectiveness, many women said they would be willing to use TENS again in a future labour (Doswell et al, 2009). It’s interesting too that women said they were coping with the pain but felt more confident doing so with the distraction of a machine (Doswell et al, 2009).

Does TENS have any side effects?

While some women dislike the sensation or find the device irritating, there are no side effects (OAA, 2016).

Can TENS slow down labour?

TENS does not seem have an effect on the length of labour, interventions in labour, or the wellbeing of mothers and babies (Doswell et al, 2009).

Does TENS affect breastfeeding?

It’s not known whether there’s any impact on breastfeeding.

Does TENS have any long-term effects?

No.

What happens if I don’t like TENS?

If you don’t like the sensation or want to try a more effective method such as water instead, it’s quick and easy whip off the TENS pads and move on.

Where can I get a TENS machine?

Some hospitals provide TENS machines. If yours doesn’t, you can hire or buy one. 

Where can I use TENS?

TENS is used by labouring women at home, midwife-led units, travelling in the car, and on the labour ward.

When during labour is TENS used?

It’s usually used in early labour, and it can even be used in a car on the way to hospital (OAA, 2016).  It is not thought to be beneficial to start using TENS during strong labour (NICE, 2007).

How can my birth partner help with TENS?

The woman will need help to put on the sticky pads. It is a good idea for the birth partner to check the instructions before labour starts, and make sure you have fresh batteries. The pads will mean it’s not possible to massage the back, so you can try massaging her head, arms or legs instead if you wish. 

Can I use TENS with other pain relief?

As TENS is electrical, you can’t use it at the same time as water, and it might interfere with some hospital monitoring.  You can use TENS at the same time as Entonox or pethidine/diamorphine.

Would I need any extra procedures if I use TENS?

No.

This page was last reviewed in April 2021.

Our support line offers practical and emotional support with feeding your baby and general enquiries for parents, members and volunteers: 0300 330 0700.

We also offer antenatal courses which are a great way to find out more about pregnancy and life with a new baby.

Research from the Royal College of Obstetricians and Gynaecologists (RCOG) about the use of water as pain relief in labour.

Machine translation - a global economic reformer

Machine learning has significantly influenced the modern world and our vision of the near future. Self-driving cars, smart assistants on smartphones, and video analytics are all examples of how far technology has come.

However, of all the areas where machine learning is applied, few places have the potential to radically reform the global economy as machine translation does. Translation from one natural language to another is an ideal task for machine learning, where high-quality data at the input to the learning system plays a crucial role. The success of the ImageNet project for recognizing objects in photos and videos was preceded by many months of manual data preparation for training, i.e. markup of photo and video materials. In the case of machine translation, data – original texts and their digital translations – arose long before the practical application of neural networks. Today, by combining advanced machine learning technologies and high-quality data, automatic translation results are produced that, in many cases, are not inferior in quality to human translation. Considering that machines translate at a fantastic speed, hundreds of times greater than human capabilities, it can be argued that overcoming the language barrier is turning from a dream and an advertising slogan into reality.

Strategic objective

Translation can be called one of the most important strategic tasks in the world – authorities and large companies have always had to spend significant funds to solve it. Let's look at the European Union. Single currency, common borders, but different languages. The EU brings together citizens from 28 countries and is obliged to provide access to official information in all EU languages. To solve this problem, a special department has been created - the European Commission's Directorate General for Translation (Directorate-General Translation - DGT). Each year, this organization is allocated an impressive budget - in 2018, it amounted to more than 20 million euros for translation alone.

The EU allocates even more substantial budgets for research and technology development programs. For example, in 2016, funding under the largest EU framework program to support and encourage research Horizon 2020 (the total program budget for 7 years was 80 billion euros) was received by the DANTE project, which brought together 14 technology companies and representatives of the Ministry of Internal Affairs of a number of European countries. The goal of the project is to create a platform for searching, collecting and analyzing multimedia and multilingual data from the Internet, including the Deep Web and Dark nets, aimed at promoting and financing terrorist activities. One of the most important components of the platform is the service for machine translation of text and automatically recognized speech, powered by the technologies of the Russian company PROMT. Thanks to Horizon 2020, many IT companies working in the field of linguistics - machine translation, speech recognition, analysis of unstructured data - received funding and support for research and development of new solutions.

Commercial organizations are also actively investing in the translation industry. According to the rating of the research firm Nimdzi, the revenue of each of the hundred largest translation companies in the world is from 10 to 700 million dollars, and according to analysts' forecasts, the market for translation services will only grow in the future. According to Global Market Insights, Inc. By 2021, the volume of the global market for language services will be $56 billion.

Business English, difficult Chinese and other languages ​​

The query "Business English" appears in the Google search engine more than 8 billion times. Business English is a good product, on which many earn money, and, what is important, they will always earn money because trade, cooperation between countries and regions will only expand, which means that the need for knowledge of English will only grow. Proficiency in oral business English is a useful and important skill, but even more often written business English is needed for correspondence, understanding technical and marketing documentation, localizing the content of online stores, products, websites and much more. It is here that machine translation not only helps to solve communication issues, improve document flow and localization, but also provides a competitive advantage.

English is also traditionally used as an intermediary language when it is not the first language of both parties in business negotiations. Translation agencies not only in Moscow, St. Petersburg, but also in the Far East claim that requests for translation from and into Chinese are minimal, the main foreign language is English. And this is despite the fact that in Russia China is the absolute economic leader in both imports and exports: according to the Russian Export Center (REC), the trade turnover between our country and the Celestial Empire in 2018 amounted to $108,244. 7 million, exports - 56 019.4 million, and imports - 52,225.4 million.

But in some situations, an intermediary language is not the best option. For example, tourism. Chinese tops the top 10 most spoken languages ​​in the world with over a billion speakers. However, this language is difficult to learn and even listed in the Guinness Book of Records as the most difficult from the point of view of Europeans. This means that here, too, the use of technology - machine translation of text and speech on mobile devices for ordinary tourists, as well as police, doctors, emergency services and any other related to the service sector - could simplify interaction and provide new opportunities for economic development.

There are now over 6,500 living languages ​​in the world. One third of all languages ​​are languages ​​of poor or developing regions that are outside the global market. In 2012, Boston Language Institute researchers conducted a survey of entrepreneurs and found that 64% of respondents consider the language barrier a serious barrier to access to foreign markets. Most businessmen believe that the elimination of the language barrier can increase income several times and significantly expand the customer base.

Neural network machine translation

Machine translation is not as expensive as the work of professional translators, not to mention the fact that there are many translation tasks that, due to the volume of content and tight deadlines, cannot be solved with the help of human labor. But the main thing is that the quality of machine translation is reaching a fundamentally new level thanks to neural networks.

Translation technologies that were used before neural networks (RBMT, statistical machine translation) have reached a sufficiently high level mainly for gisting tasks (in linguistics, the term "gisting" means a translation sufficient for a general understanding of the text). Before the advent of neural network translation, the most common scenario for using machine translation was the ability to quickly translate text, for example, a letter, presentation, news, in order to understand the meaning or to check that the meaning was understood correctly. Although there were exceptions here too - IT companies, for example, PayPal, PTC, Kaspersky Lab, CIsco, Adobe and others have long and successfully used machine translation for translating documentation. The localization process in this case consists of several stages: first, machine translation, then post-editing and verification by a translator-editor. This helps to reduce the time for preparing documentation and save from 15 to 30% on the cost of localizing product documentation.

But neural network technologies help to achieve even more. With their help, you can get translations of the so-called publishing quality with virtually no human intervention, i.e. there should be human verification of the result, but the amount of post-editing is significantly reduced or not needed at all.

Neural networks, although a fashionable topic, are not so new anymore - specialists have been developing them for more than 70 years. However, the first turning point happened really not so long ago: in 2007, deep learning algorithms for multilayer neural networks were presented at the University of Toronto. Five years later, researchers at the same university taught deep neural networks to recognize objects in photographs and videos with a minimum of errors. This is how the boom of neural networks began in the world, new technologies got to machine translation. It is important to note that the boom in research became possible, among other things, thanks to large international open source projects in machine learning, such as, for example, TensorFlow from Google. TensorFlow, a software library for building and training neural networks on various types of data, was originally developed for internal use, but was transferred to open source in 2015. TensorFlow, OpenNMT, Sockeye, Marian, Nematus and many other open source tools have given a huge boost to the development of neural network machine translation all over the world at the same time.

The new trend is confirmed in practice. Every year, the Association for Computational Linguistics (ACL) hosts a machine translation workshop where developers train their systems on the data provided and demonstrate the results. Basically, the developers presented statistical machine translation systems, and last year most of the systems were already working on neural networks.

Neural networks made it possible to make a colossal breakthrough in overcoming language barriers. Even today, machine translation without a “machine accent” is a reality; thanks to the training capabilities of neural networks, it is possible to translate the text of any subject area with high accuracy, respect for terminology and preservation of style.

System training is one of the most important stages for obtaining a high-quality translation of complex texts. Previously made translations are used to train neural models. It is important that the data for training correspond to the topics of those texts that are subsequently planned to be translated. During training, the system "remembers" highly specialized vocabulary and style - it is this vocabulary and syntax that will be in the translation.

How to measure quality?

There are several generally accepted methods in the translation industry for measuring the quality of machine translation. The first method is called BLEU, and it is based on the automatic comparison of text translated by a machine with a reference translation performed by a human. The closer the machine-translated text is to the reference text, the higher the BLEU scores and the better the result. PROMT experts made more than a hundred such comparisons for translation from English into Russian in customer texts and obtained the following data: neural technologies - from 30 to 58 points.

Another method not only evaluates the quality of the translation, but also simultaneously estimates the amount of work that a specialist must perform when editing machine translation results - the so-called "editing distance". The smaller this distance, the closer the translation is to the reference and the less work the editor has to do. According to PROMT experts, the editorial distance in machine translation based on neural networks is reduced by one and a half to two times.

Assessing the quality of machine translation

In practice, this means that the time needed to prepare the translation of product descriptions, product catalogs, project and other documentation is significantly reduced. At the same time, the cost of translation as a whole is also reduced, because the editor will have less work to correct the machine translation and prepare the final versions.

For other tasks, where a good quality translation is enough, but grammatical and stylistic inaccuracies are acceptable, the editor's work is not needed at all. A user of machine translation - an engineer, lawyer, IT specialist, analyst, sales manager and any other employee of the company - can get a translation of really high quality and sufficient for his work tasks in a matter of seconds or minutes, if we are talking about multi-page documents.

A new economy without language barriers

Exchange of documents and electronic messages, appeals to government agencies for clarifications or supporting documents, legal assistance, localization of goods and services, customs - everywhere, fast and accurate translation is extremely important. Given the number of natural languages ​​in general (English, Russian, French, Chinese…) and professional options in particular (the language of lawyers, marketers, financiers, doctors…), we get a Tower of Babel of unprecedented proportions.

But neural network machine translation has everything to reduce or even destroy this tower and, ultimately, eliminate the language barrier, just as the Internet managed to eliminate the geographic barrier in its time.

Thanks to the Internet and general globalization, new opportunities for competition have appeared - more and more companies are gaining access to the global market, regardless of capital, resources, political influence, geographical location. And, if now this access is limited by a language barrier, then in the very near future - thanks to neural network machine translation technologies - this restriction will disappear.

About the author: Yulia Epifantseva - Business Development Director at PROMT. Graduated from the philological faculty of St. Petersburg State University. Developer of machine translation technologies.

Follow our Telegram channel so you don't miss the most important things!

Share:

Digital footage. Innovative solutions for work with personnel. We measure labor.

All articles

08/16/2022

To increase labor productivity, you must first measure and evaluate this work. And there is a solution in this area - this is an artificial intelligence-based system for monitoring physical labor from the SOLUT company, which is part of the LANIT group. In a short time, the startup has turned into a dynamically developing company, moving from "pilots" to successful implementation projects with the largest representatives of the construction industry, retail and other industries. According to the results of the implemented projects, an increase in personnel output by 20-25% was recorded. Alexander Belousov, CEO and founder of SOLUT, will talk about a unique product that gives not only a quantitative, but also a qualitative assessment of an employee's work.

HOW IT WORKS

The system, using AI models built using machine learning methods, recognizes by the movements of the leading hand of the employee what specific actions he performs throughout the entire shift.

To do this, the worker is equipped with a wearable device in the form of a wristwatch, where the SOLUT application is installed, which collects data from the built-in sensors: accelerometer, gyroscope, barometer, magnetometer and heart rate monitor.

Alexander Belousov, director and founder of SOLUT: "These sensors, in fact, describe the movement of the hand in space. Then the data is transmitted to the server, where AI recognizes a variety of human actions. Baseline: worked or not worked. On at a more detailed level, elementary actions are recognized: working with a hammer, a wrench, a welding machine, power tools, laying out goods on a shelf, drilling, changing price tags, twisting a cable, winding with a stretch film, carrying heavy loads and much more. for each specific specialty or job assignment.

According to him, the recognition accuracy is more than 90%, which is confirmed by comparing the readings of the sensors in the watch with the corresponding video sequence. High accuracy is important so that the customer (employer) can explain to his employees why he trusts the system reports, on the basis of which he often forms the payroll.

An additional option is to determine the location of the employee. This is important for retail, where it is often necessary to know not only what the employee is doing, but also where he is doing it: for example, on which rack the goods are displayed.

WHY

The most important result for the customer of the implementation of a monitoring system is an increase in employee productivity. This happens by reducing downtime, improving the organization of production, and better motivating employees.

"The system detects downtime in the work of thousands of people every day, for a long time, and the task is to figure out why the downtime occurs. Workers are far from always to blame, it often happens that they were not given material, a tool - there can be a million different reasons. For this the watch has an option to select the reason for downtime, where the worker can select it from the menu. This also goes into a report, on the basis of which our customer's management can optimize production processes, expand bottlenecks, "says the founder of SOLUT.

Alexander Belousov, director and founder of SOLUT: "For example, one team repairs an oil drill in three hours, and the other in five, and so on each time. And two hours of idle drill is a loss of a lot of money."

With regard to rationing, it is traditionally carried out with the help of a timekeeper who records the duration of operations. And here a distorted picture is often obtained: some workers begin to work faster under supervision, while others deliberately work more slowly so that they do not receive an increase in the norm. When using the SOLUT system, the customer receives the maximum possible sample for all workers at once and for any period, and in natural conditions, which gives a more reliable picture.


The third case is quality control of operations within the process. The system makes sure that the operations included in the technical process are necessarily performed, and in strict sequence and in the right quantity. Safety violations and unauthorized actions are also monitored: from simple smoking in the wrong place to complex industry cases when employees violate technology.

"For example, to repair the unit, you need to perform a series of operations: unscrew 20 screws, remove the cylinder block cover, fill in oil, change filters. The system will make sure that all operations are completed. Or: the part needs to be polished for ten minutes, and the worker is too lazy, I polished for five minutes, then it doesn’t fit, and he hammers it in. The system recognizes such process violations,” says Belousov.

CONTRACTORS AND SUB-CONTRACTORS

The system allows you to monitor the work of several contractors or subcontractors within a large project at once. The main customer of the work can receive daily accurate data (percentage of work, downtime, etc.) for each worker, contractor and subcontractor, rank them and make important management decisions at the moment.

Belousov describes an interesting example of the different use of the system by two construction companies that work on the same large facility. At the same time, one company builds on its own, and the second - with the help of subcontractors.

Alexander Belousov, director and founder of SOLUT: "The first to tie the salaries of workers to the data of our system, motivating them to increase productivity. This is a great example: it's good that most customers accept our paradigm, when the system should be of value not only for the company itself but also for workers. And they already have a productivity increase of 15-20%, and it continues to grow."

The second company, which signed contracts for construction work with subcontractors at a fixed price, uses the SOLUT system to rank these subcontractors: those who work faster and better, put them in the most critical areas.

"It happens that a subcontractor fails to meet deadlines, which means fines that the customer will impose on the general contractor. Or, let's say you are building on loan money, and every extra day is important. Therefore, the system issues a schedule of performance of subcontractors, which allows the contractor to foresee, one of them will go wrong. Such a subcontractor is given recommendations for optimization, and, if necessary, the amount of work is taken away from him, "says the head of SOLUT.

DEVELOPMENT AND EXPANSION

The system is constantly being improved, interfaces are improving, processes are accelerating, new options are appearing. With the expansion of the circle of customers, new specialties are included in the recognition base. Also, the company, based on the experience of implementation and the best practices of its customers, has developed a methodology for working with the system: customers are advised how to improve certain processes based on the reports received.

The scope of the system is also expanding. So far, the leading enterprises in the construction industry and large grocery retail. The next range of clients, where there are already confirmed "pilots", is oilfield service, mining, metallurgy, and power grid companies. Also, the application of the system is possible in housing and communal services, in ports, at large warehouse and logistics enterprises, in agricultural holdings - wherever there is still a lot of manual labor.

The company also began to actively develop exports, promoting its product to the markets of the Middle East, South America and Southeast Asia. "The process of entering foreign markets will be lengthy, since potential customers need to be explained a lot, because we are the only ones in the world so far, and the market for such products simply does not exist, it needs to be created," concludes Belousov.

Savings and flexibility: what digital ecosystems give to business Own ecosystem allows companies not only to quickly respond to new market trends, but also to maintain stability in a crisis.
Maria Reshetnikova 04/17/2023

Figures recalculated by platform In 2023, the interest of companies in optimizing business processes using specialized digital platforms will increase, market participants expect. This is facilitated by the departure from Russia of many foreign players, active reorientation to domestic solutions and support from the state. Under these conditions, domestic digital platforms become the technological basis for the development of ecosystems in various sectors of the economy, interviewed experts from the LANIT group of companies told Kommersant. Ivan Perevalov 29.03.2023

How can a brand improve the accuracy of customer referrals? Imagine that you know in advance which product will interest the client, as well as where and when it is better for him to make an offer. Such ultra-precise personalization is possible. How to achieve it, says Anastasia Semenov (CleverData company, part of LANIT Group) Anastasia Semenova 05.

Learn more

• 
  How to care for autism child
• 
  Is it safe to eat cottage cheese while pregnant
• 
  Newborn body temp
• 
  Why do you get heartburn in pregnancy
• 
  At what week do you start showing
• 
  Babies spit up smells like vomit
• 
  Baby head coming out
• 
  Miscarriage gush of fluid
• 
  When is the miscarriage stage over
• 
  How long can you store colostrum
• 
  How to tell pregnancy