Read the article." />
Skip to Content


MDS makes every effort to publish accurate information on the website. "Google Translate" is provided as a free tool for visitors to read content in one's native language. Translations are not guaranteed to be 100% accurate. Neither MDS nor its employees assume liability for erroneous translations of website content.

International Parkinson and Movement Disorder Society
Main Content

Can dogs diagnose Parkinson's disease?

November 14, 2022
Dr. Sarah Lidstone interviews Dr. Chang-Qing Gao & Prof. Ji-fung Guo about their recent research on the sensitivity of sniffer dogs for diagnosing Parkinson's disease. Read the article.

[00:00:00] Dr. Sarah Lidstone: Welcome to the MDS Podcast, the official podcast of the International Parkinson and Movement Disorder Society.

Today we'll be speaking to Dr. Chang-Qing Gao and Professor Ji-fung Guo from Shiang-Ya Hospital in Central South University, Changsha, China. They are the first and senior authors of a remarkable prospective multi-center double blind study, the sensitivity of sniffer dogs for diagnosis of Parkinson's disease diagnostic accuracy study.

So welcome Dr. Gao and Professor Guo.

View complete transcript

[00:00:36] Dr. Chang-Qing Gao: You're welcome.

[00:00:37] Professor Ji-fung Guo: You're welcome.

[00:00:38] Dr. Sarah Lidstone: Very happy to speak with you today, all the way in China.

[00:00:40] Dr. Chang-Qing Gao: We are very happy too.

[00:00:42] Dr. Sarah Lidstone: Excellent. So can you give us a summary of why your group did this study and what this study adds to the field.

[00:00:49] Dr. Chang-Qing Gao: In the past 10 years, our team has been focusing on training dogs to detect human diseases such as cancer. We [00:01:00] also have been wondering how we can improve the diagnostic procedures of the Parkinson's disease. In 2016 there is a story came from the journal of neurology, seeing that joy, a human's mother who can identify the t-shirts of the people with Parkinson's disease from those of normal people.

This attracted our attention immediately. Three years later, a group in University of Manchester published an article saying that they have found some molecules, which was possible for this smell of PD. This inspired us to collaborate to do this study, the other has been added to the field. Maybe it is more appropriate for [00:02:00] professor Guo or to answer this.

[00:02:02] Professor Ji-fung Guo: Yeah. Lecturing. We have new tools to help us to make a diagnosis? Now we have diagnosis teacher criteria for Parkinson's disease, which was written by professor Posthuma and the other professor. But it's too complicating and there's difficult to remember for non movement disorder doctors. These criteria are difficult to use in the diagnosis or area of prodromal Parkinson's disease. Therefore, those fears, there needs something new, something easy to use to help the diagnosis. Take procedures. Now, one more test to use when there are diagnosed disease, more important our studies suggest that patients with Parkinson's disease do [00:03:00] have unique smell, just pretty wide. We may be able to develop the equipment to detect the disease and may, find new drugs to treat this disease.

[00:03:12] Dr. Sarah Lidstone: I think it's, it's so important and so interesting, and I think I shared with you separately that I, I have a dog and I'm a dog lover as well, so this study is, certainly of great interest. Can you tell us a little bit more about how the dogs were trained to detect Parkinson's? Is there also something unique about that particular breed I saw it was the Belgian Malinois Shepherd that was used for this type of work.

[00:03:37] Dr. Chang-Qing Gao: Yes. The method we used. Are very similar to what the police are using for training their dogs. I mean, the positive feedback better. We train them when they find the target we'll give them the food or a piece of toy [00:04:00] to learn to reward them and the stages of the training can be artificially divided into three stages.

First, the obedience training just teaches them to come sit down and second stage, teach them to identify the tanks that contains the food on the set stage to teach the dogs to distinguish the tank containing the samples of the patients with PD from those without the disease. After this, we evaluate the dogs.

Only these dogs with the diagnostic accuracy, over 95% are qualified to go to the clinical trial. I have to say to emphasize that in the clinical [00:05:00] trial, the samples used in training have never been used in the clinical trial and during the test both my assistant and myself do not know which Tank contains the target. So it's a double blind study.

[00:05:19] Dr. Sarah Lidstone: Yeah, it's a very important point that there was blinding on both sides that occurred, but also that these were fresh samples the dogs had never been exposed to before, and that they were clearly very highly well trained dogs for this particular purpose already. So it wasn't that they were, you know, trained doing other, types of diseases. They were trained specifically for Parkinson's before even participating in this trial.

[00:05:42] Dr. Chang-Qing Gao: So upon the, the breed of the dog, Malinois is a very good dog. It is very smart, very active. They like to play so easy to train. They have very long nose, they have a very good sense of smell. So [00:06:00] it's good and suitable for teaching them to detect something. Good working dog

[00:06:06] Professor Ji-fung Guo: Mm-hmm.

[00:06:06] Dr. Sarah Lidstone: That's wonderful. Can you tell us a bit about this concept of a PD smell, a Parkinson's smell, right? Clearly as human clinicians, we're not able to pick this up diagnostically, do we know what compounds are in the smell? What's responsible for the PD smell and how might it be different from other types of diseases?

[00:06:27] Dr. Chang-Qing Gao: Well Given the research group in the world have reported something first, there is a lab in the University of Manchester, they reported that some compounds might be the metabolites of the bacteria on skin over the back, but then later they suggest something new. Another group in Italy, they did very good work, meticulous work.

They found that among [00:07:00] 51 very important compounds in the sebum of the human being, at least 35 of them in the, quantity, it is different between the people with PD and those without PD.

[00:07:18] Dr. Sarah Lidstone: Hmm.

[00:07:19] Dr. Chang-Qing Gao: Until now. What exactly the compounds is, We don't know. We still have to study.

[00:07:26] Dr. Sarah Lidstone: Hmm. But that's, that's remarkable that already we know that there's 35 different compounds that are different in the sebum of human beings between PD patients and controls. That's remarkable. I also learned from your paper that, other things have to be taken into account when we're thinking about these compounds.

Things like disease duration, the medication status of, of the patient, the patient's age, and how these factors might influence the odor. Can you share with us about that please?

[00:07:55] Dr. Chang-Qing Gao: Yes, we studied in the medicated people [00:08:00] and in the blood naive patients too. Both of them, we found that the dog can identify the PD patients correct. The accuracy nearly the same. Just to say the medication doesn't interfere with the detection because medication, may bring some new smell to the people. The smell of the drug and it's metabolites, but the original smell of PD has never changed. It is always there. The dogs used these, the smell of PD it itself to identify the patient's samples.

So, The medication doesn't interfere the work of the data of the dogs, but ages, maybe a factor to have some roles in the detection because in our study we found that the sensitivity [00:09:00] in the age, of the people, over 60, It needs to be newer than the general of the population studied. It is reasonable because the sebum produced by the age of the people maybe declined this, maybe the reason the dog cannot detect as much as that in the younger people.

[00:09:24] Dr. Sarah Lidstone: I think it raises a lot of questions, and I think we'll talk a little bit later about the role of how the dogs might be helpful in diagnosis, but I think the age, the age question was very appropriately raised in the paper and I think has, clinical relevance for sure. We talk a lot about diagnostic accuracy of Parkinson's and I was just wondering if you could share with us a little bit about how the sniffer dogs would compare to other premortem diagnostic tests that we use now or that are being studied now. So for example, MRI integrity of the nigra zone, or skin biopsies, et cetera.

[00:09:58] Dr. Chang-Qing Gao: Yes. I can say [00:10:00] is really an issue to be addressed. In our study the accuracy of the test is about 95% overall, so compared to the clinical diagnosis. I think it's very, very good. But of course, we used the clinical diagnosis as reference test or reference standards, not the pathologic examination. So the real accuracy of the test. We don't know.

[00:10:27] Dr. Sarah Lidstone: And that would apply to all of our pre-mortem diagnostic tests as well?

[00:10:30] Dr. Chang-Qing Gao: Yes. same for other tests too.

[00:10:33] Dr. Sarah Lidstone: Can you comment on the ability of the dogs to detect this concept of a prodromal PD?

[00:10:39] Dr. Chang-Qing Gao: Oh, we're very happy to share about this. In our study, the dog. Has signaled out that in the control group that some people, the dog say it is positive and we, tested six of them. Five of them have one or another prodromal signs [00:11:00] of PD. So just to say this test, the sniffer dog test may have the ability to detect the people with prodromal PD.

[00:11:11] Dr. Sarah Lidstone: I thought this was absolutely fascinating and a very important point that five of the six people had features, I think you wrote that one had dreaminess or a suggestion of perhaps a rems sleep behavior disorder. One had some stiffness in the upper limbs. I'm not sure if that was rigidity that was documented by a clinician or self-report from the patient.

Another had enlarged substantial nigra echogenicity. So there are perhaps some prodromal features that the dogs are able to detect in the sea bone, which is fascinating.

[00:11:41] Dr. Chang-Qing Gao: Yes, you're right . We're happy with this. Very happy with this.

[00:11:45] Dr. Sarah Lidstone: Clearly very clinically relevant for us in the field, especially for the perhaps non-movement disorder specialists as well. Can you talk to us a little bit about the high false positive rates in these types of diagnostic tests?

[00:11:58] Dr. Chang-Qing Gao: Yes, this is really something [00:12:00] we were concerned about. So in our Chinese population, the prevalence of PD is about 1.7% in the population aged over 65. You all started, We have 4.7% in the medicated studied population and in the drug naive study population even higher. So indeed we have a higher false positive might be a false positive results. However, we have to consider that we discussed about that, in the positive controls, the majority of them, five of six, they have one or another symptoms or, or signs. So in these positive controls, they're indeed some prodromal patients there.

Another thing, maybe other factors may be involved in it too. Some other things like the environment factors and other [00:13:00] factors we don't know, maybe made the dog with some mistakes. So in general there are false positive there, so we should not use the sniff dog test, allowed to make a diagnosis. It is very important.

[00:13:16] Dr. Sarah Lidstone: Yeah, so the fact that we probably have to use other tests and not the sniffer dogs in isolation to diagnose Parkinson's is a very important point. And it's, it's related to my next question, which is based on, on the results of this study, how do you see the role of sniffer dogs in the diagnosis of Parkinson's?

How do you think they should be incorporated into the clinical world?

[00:13:38] Dr. Chang-Qing Gao: Yes, I think this test is cost effective. It's non invasive and easy to use, so it can be used in community screening and in the disease prevention Center in some cases, maybe in the neurology clinic, to help [00:14:00] doctors to make a correct diagnosis.

[00:14:04] Dr. Sarah Lidstone: Are you not advocating that we should replace all of our neurologists with dogs then? Are you?

Surely not, the sniffer dog test is just a tool like MRI, CT.

Well, I can say it's probably a test I would certainly use.

[00:14:21] Dr. Chang-Qing Gao: Yeah, well you're welcome.

We, myself and the Dr. Guo. We check ourselves regularly.

[00:14:31] Dr. Sarah Lidstone: Right. Self screening. Of course. Oh goodness. Well, that's wonderful. Thank you very, very much for sharing with us the background and behind the scenes of this study. It was a really fascinating paper, and I think will be of great value to our listeners. So thank you again.

[00:14:44] Dr. Chang-Qing Gao: Thank you very much.

[00:14:45] Professor Ji-fung Guo: Thank you.

[00:14:46] Dr. Sarah Lidstone: Okay. Thank you very much.

Special thank you to:

Dr. Chang-Qing Gao & Prof. Ji-fung Guo

Sarah C. Lidstone, MD, PhD, FRCP(C) 

Toronto Western Hospital/Toronto Rehabilitation Institute, University Health Network and University of Toronto

Toronto, Canada 

We use cookies to give you the best possible experience with our website. These cookies are also used to ensure we show you content that is relevant to you. If you continue without changing your settings, you are agreeing to our use of cookies to improve your user experience. You can click the cookie settings link on our website to change your cookie settings at any time. Note: The MDS site uses related multiple domains, including and This cookie policy only covers the primary and domain. Please refer to the MDS Privacy Policy for information on how to configure cookies for all other domains on the MDS site.
Cookie PolicyPrivacy Notice