Close
Explore more Centres, Projects and Groups
Welcome
Welcome Banner
Banner
A bio-detection dog takes part in training. Credit: MDD/BexArts/Nigel Harper

Using dogs to detect COVID-19

Investigating whether dogs could be trained to identify unique odours associated with coronavirus infection

Bottom Content
Logo List Links
Intro Blocks List
About

The Arctech Innovation team at the London School of Hygiene & Tropical Medicine, Medical Detection Dogs and Durham University ran a study training dogs to identify unique odours associated with coronavirus infection.

Who we are

Meet the interdisciplinary group of researchers involved in this exciting study, all of whom have extensive experience working with volatile compounds and human infection.

Meet the dogs

Meet the six dogs that were trained to detect if someone has COVID-19 and could play a vital role in preventing further spread of the pandemic in future.

Why use dogs?

Dogs are great biosensors, capable of detecting odours associated with drugs, explosives, and food and are now being used in practice for public health to detect diseases such as malaria.

Donations
Twitter Embedded Code
Facebook Embedded Code
About
About dogs COVID 2 columns
About dogs COVID 2 columns left
Paragraph

Background

This study was built on the scientifically proven ability of dogs to act as great biosensors, capable of detecting odours associated with human health, as well as drugs, explosives and food.

For example, it has been scientifically proven that dogs can sense small changes in volatile organic compounds (VOCs), which can be produced by the human body when it is diseased. Dogs have already been trained to diagnose cancer in samples from patients, to alert people with diabetes that their blood sugar is low and detect many other diseases.

In a study published in the Lancet Infectious Diseases last year and funded by the Bill and Melinda Gates Foundation, our team of partners demonstrated that dogs can detect the presence of malaria in human odour with an effectiveness greater than 90%, even at an early stage of infection, which is above the World Health Organization’s standards for a diagnostic test. Importantly, the dogs can detect this in individuals who have no symptoms.  We are now furthering this method to be used on the ground in sub-Saharan Africa where malaria is of serious concern.   

We hypothesised that dogs would also be able to detect VOCS in people who are infected with COVID-19. The research involved training six available bio-detection dogs to distinguish between positive and negative odour samples, collected from up to 3,250 asymptomatic adults (who were independently tested for COVID-19 infection). Our vision is to develop a fast, effective and non-invasive diagnostic, with the bio-detection dogs working to detect individuals infected with coronavirus at ports of entry and large gatherings.

Our world-leading team of scientists is highly experienced and perfectly positioned to perform the work required to demonstrate that dogs can detect COVID-19, deploy and up-scale. 

The implications of this study

This study could have a profound impact on how quickly we return to normality. By demonstrating that the six dogs are capable of accurately detecting the odour of COVID-19, this non-invasive testing approach can then be scaled up significantly and rapidly.

Dogs could screen travellers at airports, identifying individuals who may require an additional test for confirmation or advised to self-isolate. This would allow the UK to further re-open its borders to international travel, with authorities confident that asymptomatic passengers can be quickly and effectively identified on arrival.  

There is also potential for trained dogs to be used to screen commuters at travel hubs, visitors at museums, theatres, cinemas and sporting venues, which could help the entertainment sector. And just as importantly, a fast, effective and non-invasive COVID-19 diagnostic tool would ensure other more limited healthcare testing resources are used where they are truly needed.

Ultimately, we intend to scale up operations on a global scale, by working with dog training agencies, and the military and police forces, to deploy dogs in as many countries as possible. We are already in discussion with local governments and border agencies in several countries regarding this.

Who we are
Who we are dogs COVID 2 columns
Who we are dogs COVID 2 columns left
Paragraph

Our expertise

LSHTM has the expertise and knowledge to make a significant impact on COVID-19 diagnostic testing. We have extensive experience working with volatile organic compounds (VOCs) and human infection. Previous work from our group has shown that people infected with malaria parasites produce a body odour detectable by mosquitoes, which results in malaria mosquitoes preferentially feeding on asymptomatic, malaria-infected individuals. This was published in the high impacting journal PNAS. Together with Medical Detection Dogs and Durham University we later showed that dogs, with their highly advanced sense of smell, could be trained to detect people carrying malaria parasites. This was published in the journal Lancet Infectious Diseases.

Our research team is an interdisciplinary group of researchers with extensive experience working with VOCs and human infection, comprising experts in dog training and behavioural science, chemical ecology, infectious diseases, diagnostics, virology, biological chemistry, epidemiology, statistics and clinical trials. 

Study researchers

Professor James Logan

Professor James Logan is the Director of Arctech Innovation and Head of the Department of Disease Control at the London School of Hygiene & Tropical Medicine.

James is the Principal Investigator of a large research portfolio investigating chemical ecology approaches for the surveillance and control of diseases. This includes using sensors, insects and other animals as biosensors for the identification of VOCs that can be used as diagnostic markers for disease, or as attractants and repellents for arthropods that transmit pathogens of medical importance, including malaria, Zika, dengue, trachoma, and Lyme disease.

Dr Claire Guest

Dr Claire Guest is a behavioural psychologist with more than 30 years’ experience of understanding and training dogs. In 2004 she directed one of the first programmes in the world to train dogs to identify cancer by odour and published the first robust proof of principle study in the BMJ later that year. She went on to co-found Medical Detection Dogs with Dr John Church in 2008.

Over the past 12 years under Claire’s leadership, Medical Detection Dogs has focused on establishing a strong evidence base for its research. Its commitment to rigorous scientific methodology, and to improving our understanding of canine olfactory ability, has led to research into the detection of a number of cancers, Parkinson’s disease, malaria and bacterial infections. It already applies what we know about the science of canine olfaction to benefit people by training Medical Alert Assistance Dogs, which help individuals manage complex, life-threatening medical conditions. The potential for this work to lead to the development of fast, reliable and non-invasive diagnostics is enormous.

In 2016 Claire wrote “Daisy’s Gift”, which tells the story of the formation of the charity and her remarkable dog ‘Daisy’, who indicated her own breast cancer. 

Professor Steve Lindsay

Professor Steve Lindsay is a public health entomologist and Professor in the Department of Biosciences at Durham University.

Steve has considerable experience in public health interventions and has run large clinical trials measuring the impact of mosquito control on malaria in children. He was the principal investigator of the study that demonstrated that trained dogs could identify asymptomatic carriers of malaria. He is also the proud owner of a young working Labrador called Tilly.

Dr Sarah Dewhirst

Dr Sarah Dewhirst is Head of Research & Development at Arctech Innovation at the London School of Hygiene & Tropical Medicine.

Sarah has over 17 years’ experience in entomological and chemical ecology research, including in a post-doctorate position working on identifying odour baits to substantially increase the efficiency of traps and targets used in the control of the tsetse fly population, a major vector of African Sleeping Sickness.

Dr Anna Last

Dr Anna Last is a clinical epidemiologist and Associate Professor in Infectious Diseases at the London School of Hygiene & Tropical Medicine. 

Anna has been working on a network of integrated studies on the integrated elimination of NTDs (including soil-transmitted helminths, lymphatic filariasis, schistosomiasis, yaws, and scabies) and malaria in East and West Africa.

Dr David Allen

Dr David Allen is Associate Professor of Virology at the London School of Hygiene & Tropical Medicine.

David’s research interests are in enteric virus infections - particularly norovirus and enterovirus - and the virological features of these viruses that drive their highly dynamic epidemiology, allowing them to emerge and persist in human populations.

 

Other researchers included:

  • Professor John Pickett CBE FRS, Professor of Biological Chemistry, Cardiff University
  • Professor Steven Morant, Professor of Epidemiology, University of Dundee
  • Dr John Bradley, Assistant Professor of Medical Statistics and Epidemiology, LSHTM
  • Professor Immo Kleinschmidt, Professor of Epidemiology, LSHTM
  • Chelci Squires, Research Scientist and Trials Manager, Arctech Innovation
  • James Hourston, Research Scientist and Trials Manager, Arctech Innovation
  • Ann Rooney, Veterinary Epidemiologist, Lomond Veterinary Clinic
Meet the dogs
Meet the dogs 2 columns
Meet the dogs 2 columns left paragraph
Paragraph
Super six dogs in COVID-19 detection dog trial. Credit: Neil Pollock/MDD
Super six dogs in COVID-19 detection dog trial. Credit: Neil Pollock/MDD

These six dogs were trained to detect the smell of COVID-19 and play a vital role in preventing further spread of the pandemic in future.

Millie, Kyp, Lexi, Marlow, Asher and Tala picked up the scent in 6 to 8 weeks.

The charity Medical Detection Dogs intensively trained the dogs at their dedicated training centre in Milton Keynes, working in partnership with the ARCTEC at the London School of Hygiene & Tropical Medicine (LSHTM), and Durham University.

Millie

Age: 4
Breed: Golden Retriever
Character: Playful, friendly, vocal
At work: Happy with a real spring in her step
Likes: Instigating games with her colleagues and charging around the paddock

Kyp

Age: 4
Breed: Labrador x Golden Retriever
Character: Loving, affectionate, likes to sing!
At work: Bright, quick to learn
Likes: His humans, other dogs, zoomies

Lexi

Age: 5
Breed: Labrador 
Character: Happy, keen, cheerful
At work: Driven, loves to use her nose
Likes: Lunchtime walks, her ball, cuddles

Marlow

Age: 4
Breed: Labrador
Character: Works hard, plays hard
At work: Excited to enter the training room, sociable
Likes: Power naps, exploring, playing with other dogs

Asher

Age: 6
Breed: Working Cocker Spaniel
Character: Eccentric, busy, inquisitive
At work: Focussed, loves an audience
Likes: Using his nose, working, howling to be let through doors

Tala

Age: 4
Breed: Labrador
Character: Soft, gentle, cuddly
At work: Hard worker, very decisive
Likes: Jumping very high, cuddles

Why use dogs?
Why use dogs? 2 columns
Why use dogs? 2 columns paragraph
Paragraph

Dogs have an incredible sense of smell that eclipses our own abilities by orders of magnitude. The complex folded structure of their nasal cavity which results in a huge surface area is home to over 300 million scent receptors compared to 5 milllion in a human. This makes it ideal for detecting odours and combined with their ability to learn, means dogs are great biosensors, capable of detecting odours associated with drugs, explosives, and food, which most of us will have encountered at airports.  

Medical Detection Dogs has more than 10 years of research experience, in training bio detection dogs to detect the odour of disease from samples in a controlled environment, and then applying that knowledge to train them to identify changes in human odour caused by certain medical conditions.

Over the past 10 years Medical Detection Dogs has published in numerous peer reviewed journals, demonstrating that the dogs’ accuracy in detecting the odour of disease is reliably between 85 and 90%. Their research has also demonstrated that dogs can be trained to detect odour at dilutions of parts per trillion, which is the equivalent of one teaspoon of sugar in two Olympic sized swimming pools of water.

We know that dogs can already be trained to indicate the presence of respiratory conditions such as lung cancer through changes in the VOCs emitted from patients. Recent research on flu viruses has shown that they interact with the human immune system and change our VOC profiles. Dogs have already been shown to be able to detect the odours associated with respiratory infections, so we were confident that dogs could do it for COVID-19.

How will we do it?
How will we do it? 2 columns
How will we do it? 2 columns left paragraph
Paragraph
Training dogs to detect COVID-19
A bio-detection dog takes part in training. Credit: MDD/BexArts/Nigel Harper

Proof-of-principle (completed)

First we collected odour samples from people who were infected with COVID-19, and people who are uninfected. Then, we used those odour samples to train six medical detection dogs to sniff out the smell of COVID-19 accurately.

Scale-up

The next stage of research will involve testing out dogs to screen infected people in real-world settings. Then we will work with other experienced partners to scale-up our operations rapidly to deploy COVID-19 detection dogs in airports or other venues to screen large numbers of people, providing a rapid non-invasive screening for COVID-19. One single dog can screen up to 250 people per hour. We are already discussing scale up with international border agencies in several countries around the world.

FAQs
FAQs 2 columns
FAQs 2 columns left paragraph
Paragraph

Why dogs?

We were confident that dogs could be trained to detect COVID–19. Medical Detection Dogs has always adopted a rigorous scientific approach to its work and the results of its work have been published in a number of peer reviewed research papers. These results have further supported our shared belief that diseases have their own unique odour that dogs have the ability to detect.         

In principle, we are sure that dogs have the potential to detect COVID-19.

 

How will our dogs be able to help?

COVID–19 detection dogs would be to passively screen any individual, including those who are asymptomatic, and indicate to our dog handlers whether they have detected the COVID–19 virus. This will then need to be confirmed by a PCR test. The dogs would be fast, effective and non-invasive.

We intend to scale up and deploy COVID-19 detection dogs in public places such as airports, stadia and workplaces. This would help prevent onward transmission of the virus amongst large groups.

 

How many dogs did you train?

Six bio-detection dogs were trained and tested through a randomised double-blind trial. 

 

How did you train them?

Dogs searching for COVID-19 were trained in the same way as those dogs the charity has already trained to detect diseases like cancer, Parkinson’s and bacterial infections – by sniffing odour samples in a training room and indicating when they have identified a positive sample.  

 

Is it safe for the dogs?

We will keep in constant contact with scientists and veterinarians to confirm the latest understandings, however it is currently believed that this is completely safe for dogs to perform this duty. Scientists have found that although COVID-19 is a novel virus that originated from animals and was passed on to humans, dogs cannot contract the disease. The dogs were trained on non-infectious samples and did not need to make contact with the individuals the samples directly.

 

Will dogs spread the virus?

No, our dogs were trained on a dead virus, once deployed, will have no contact with the individuals they are screening. Instead, they will sniff the air around the individual. The dogs are only be permitted to be touched by the handler, and there is only a very low risk of spread of the virus from the dog to their handler.

 

How did you get samples to train the dogs?

The dogs were trained on samples obtained though the London School of Hygiene & Tropical Medicine. 

 

Results
Results Using dogs to detect COVID-19 2 columns
Results Using dogs to detect COVID-19 2 columns left paragraph
Paragraph

We’re delighted to announce that our COVID-19 dogs are now training and being tested in real-life settings such as testing centres, workplaces and events and conferences as part of their final stage of validation (Phase 3).

In this final phase of testing, the dogs will indicate if they smell its odour on a consented individual in a real-world environment. The individual will then take a PCR test to confirm that the dog has correctly indicated. It’s fast, reliable, non-invasive and will enable earlier self-isolation to prevent onward transmission.

This follows submission of the results of the first phase of our study which demonstrated COVID-19 infection does have a distinct smell which dogs can detect with incredible accuracy.

Read the Phase 1 results as published in the Journal of Travel Medicine.

Volunteer
Volunteer Using dogs to detect COVID-19 2 columns
Volunteer Using dogs to detect COVID-19 2 columns left paragraph
Paragraph

Thank you for your interest. This study has closed to further recruitment.