Cancer Sniffing Dogs

[Wobbly]

Hi DOL, it's been a long time since I posted, I was diagnosed with a stage 4 Kidney Cancer (TCC/UCC), so I took some time off from writing. But I missed you guys a lot, especially the people here and in OT. I hope you have all been well, I am well in my head, if not so much my body. I have been thinking (my brain's nearly stopped hurting from it now). About how dogs can smell cancer, they can smell a wide range of cancers at very early stages (as early as stage 0). They can do this for a wide array of cancers types - including those cancers that are often diagnosed too late to be survivable. Dogs can detect cancers from human tissue samples, condensed breath samples, sweat samples, urine samples, and presumably other types of samples too.

Some information is found at these sites.

http://www.cancerdogs.com/

http://www.dogsdetectcancer.org/

http://pennvetwdc.org/

The value of this is that it potentially offers a cheap early diagnostic screening method, something we are sorely missing for many types of cancer. Despite the billions of dollars thrown at cancer research, a cure is still a way off. Currently the only sure cure for any cancer is early diagnosis. By way of example, there are a lot of breast cancer survivors because there is an easy, cheap, early diagnostic method - look and feel for lumps, and there is a high awareness of how to perform this early screening. Unfortunately this is not the case for many other cancers - you can't look at or feel organs sitting in the body so they aren't found till late stage and thus many other cancers, including my own, have a terrible survival rate.

So I have been wondering if I can train Jarrah to detect early stage cancer. She is very motivated to work for a reward, so she's a good candidate from that perspective. I am not a proficient trainer though, so I need some help from people who are, on the internets and in person. It takes apparently a few months to train a dog to detect cancer (it'll probably take me longer) I hope to survive another few years (I'm aiming for 5 years more life, I think that's possible, so I do have some time) to get this done.

I will also need a very large array of different cancer samples. For this I will start at my chemo clinic, there are a huge variety of people who attend with different types and stages of cancer, and all on different medications, of different ages, some with other medical conditions, so I can get a large array of samples here to cover a large amount of variables. How many samples would I need? I believe with judiciously written explanations I can extend the field further and get more samples via my medical teams' clinics as well. In the first clip I linked you'll see the sample bags they are using are compact, easy to manage and presumably not prohibitively expensive, they could work for sweat, saliva or breath condensates, which at this point I think would be my preferred sample types to work with.

My aim is as follows:

- to be able to teach my dog to detect cancer of any type, at any stage, preferably via breath condensate or sweat samples as these are least invasive, and so it's easier to obtain samples for training and easier to ask the public to submit to testing.

-to be able to take Jarrah to some cancer awareness raising events, eg the finish or start of one of the walks or runs for cancer awareness (so many of these). And take samples from the general public and advise them if they are indicated as positive for cancer.

Credit for this idea goes to Dave the Farmer, who mentioned he was training a dog for this when I attended Terry Ryan's "Chicken Camp" seminar, and to Denise Fenzi, who when I attended her Seminar suggested to me that the best exercise I could give Jarrah when I was too sick to be really active with her, was some sniffer training.

Now there are some problems with this, including the fact a dog's nose is far more sensitive than any medical equipment, so while the dog may indicate positive, medical equipment may not be able to detect the cancer the dog has found for some weeks or months later when the lesions are large enough for an MRI or CT to detect. A dog of course is not always 100% reliable, so my aim in testing is to let people know that if my dog has indicated they are positive, they need to investigate further - to tell their GP a dog trained to sniff cancer has indicated they may have cancer and that they require referrals and advice for further diagnostic screening, usually MRI or CT scan, a full body scan is Medicare covered, non invasive and doesn't take long.

I will not be able to do wide scale testing myself, but as the saying goes "You can't save the world, but save one person and you save their world." And who knows seeing this in action might inspire other dog owners to try this with their dogs, and a volunteer force of more than one person could cover a lot of people. Perhaps inspire funding for wider scale awareness and use of this, it's such an easy, non invasive, potentially very cheap and accessible method of early diagnostic screening. Who knows, the potential seems very worthwhile exploring from my perspective. For myself I'd be happy to just get a few people into MRI machines to be diagnosed early enough for complete cures, the more of course, the better.

I would be deeply indebted to anyone here who could offer advice, contacts, help (especially in person help, I can travel regularly anywhere between Newcastle, Sydney, and the Blue Mountains) and any good ideas you might have.

Edited October 9, 2013 by Wobbly

[Nekhbet]

You would need to get pure samples, ie people who are not on any cancer medications which may interfere with the biochemistry of the sample. Getting a hold of medical samples like this is quite difficult due to privacy laws and biohazard.

Otherwise it's exactly like teaching any other form of scent detection.

[Wobbly]

Hi Nekhbet,

Thanks for reading and replying!

Very good point, that's a problem that I was worried about too! Luckily for me, there looks to be a way around it - from: http://www.dogsdetectcancer.org/news/dogs-smell-cancer

(love how it the pic there is a Labrador - it's Lab testing ofc. XD)

The dogs are trained on breath samples from cancer patients, which are collected in plastic tubes with polypropylene wool inside. The tubes actually collect molecules from inside the human body. The dogs are actually smelling the cells, gasses, vapors, and many different volatile organic compounds that are found within the breath condensate. Breath is not just air, it’s actually a bodily fluid.

We used a clicker to mark the moment that each dog sniffed his first cancer sample. Cancer equals reward. It’s extremely important in the early training to use many different samples, so the dogs could generalize the cancer scent. In explosives or narcotics detection, it’s easy to isolate the single target odor for the dog. The dog is introduced to the substance, rewarded, and then the handler can begin to “hide” the substance in suitcases full of other things, boxes, underground, etc..

In cancer detection, things are quite the opposite. In one exhaled breath condensate sample, there are literally thousands of volatile organic compounds, gasses, and smells. How then, can we isolate the correct scent for the dog? We can’t. Each sample contains clues from toothpaste to the last meal the person ate, his age, his general health, ethnicity, and other possible diseases. Cancer, or the gas that cancer may give off, is only one of the many things within the breath.

The best way to teach the target odor to the dog is to have a large library of cancer samples. The larger, the better. We train the dogs on so many different cancer samples until the only common denominator is the cancer scent itself. If a trainer only had five cancer samples, for example, then he’d really be training the dogs to find 5 specific peoples breath, but not cancer. It’s easy for a dog to “memorize” breath samples. By using the same 5 samples over and over again for several weeks, a dog would not be able to alert on a new cancer sample.

The key to successful cancer detection training is to have a large enough number of different samples, to swap them out often, and to know when to move on. Moving on involves introducing breath from healthy people, and also introducing breath from people with diseases other than cancer. By the time you introduce healthy breath controls and disease controls, the dogs should have no problem ignoring them. If the dog shows interest in the new healthy controls, this means he did not learn his target odor of cancer. If the dog shows interest in the disease controls, he also did not learn his target scent. A good trainer knows when to introduce the curve balls, when to hold back or repeat, and when to stop.

I can't isolate the smell of cancer, but I do have access to a huge "library" of cancer samples (via my chemo centre, medical team and support groups).

OFC for me, the big problem is being a good trainer who knows when to introduce the curve ball is the issue. I have signed up to this noseworks course in Box Hill (farmer Dave who originally gave me the idea) at http://www.lovesdogs.com.au/bookings/courses-and-events.html?task=view_event&event_id=142 hopefully that'll make me a bit better at training this. XD

And ofc I am exploring other avenues of support and help to balance out my inexperience with dog training.

Edit for linkies!

Edited October 9, 2013 by Wobbly

[raineth]

I reckon you and Jarrah can do it Wobbly :)

As you say the tricky bit would be getting her to really focus in on that cancer smell and not be confused by any other smell. I think if you approach it like any other dog training task where you outline all the steps you need for her to learn, that would be the way to go.

We encouraged Del's natural tendency to sniff out our daughter when her BSL was low. But it's quite a bit different, and I'm a bit confused about how to translate it to what you're thinking of doing. In the end with del we stopped encouraging her as she was constantly checking on Her many times a night and waking her up in the process and the poor munchkin was dead tired with broken sleep. But that won't be a problem with the job you want Jarrah to do :)

And wobbly, it's great to hear from you :)

[BC Crazy]

Sorry Wobbley am of no help to you regarding your cancer scent work with Jarrah but you have another "Yes" you can do it vote from me as well. Lovely to hear from you as I have missed you to :)

[huski]

Wobbly, sorry to hear you are sick.

Steve teaches scent detection training, let me know if you'd like more info :)

Nose work is becoming quite popular of late, there are a different ways to train it but there are some ways that are better than others (I personally would avoid any classes that train it by mixing the reward and target odour together.)

[Weasels]

Pretty sure when some colleagues of mine wanted dogs to sniff out invasive animal spor they went straight to Steve Austin :)

[bedazzledx2]

You could also contact Karen Prior

[Nekhbet]

Fun sports and pinpoint accuracy are two different things. For an operational dog it can literally mean the difference between life and death.

I train detection and currently have an operational explosives detection. I know cancer detection is still in it's infancy and you're right, it's about training the dog so much that you end up in a way unwittingly training the animal to detect cancer.

We used a clicker to mark the moment that each dog sniffed his first cancer sample. Cancer equals reward. It’s extremely important in the early training to use many different samples, so the dogs could generalize the cancer scent. In explosives or narcotics detection, it’s easy to isolate the single target odor for the dog.

Actually it's very easy to stuff up even on the easy stuff. Because the dogs are basically on a constant reward schedule, many a dog has accidentally been taught to detect an odour + another, or certain quantities, etc. There are drug detection dogs that can sniff out a small, hidden sample but would saunter past 5kg of Marijuana without blinking an eye. Dogs are so literal :laugh:

If you need a hand with anything or anything explained just email me or message me through facebook. We can bounce ideas around. The only way to do it is to try, so you can start with something easy first to get the dog learning how to concentrate. A dog with it's face plastered to the ground will not necessarily jump into scent detection/tracking easily. In fact I find the opposite because they're so used to just doing things at their own erratic pace and find concentration exhausting! Bare bones basic first, then go to cancer detection. Otherwise you're just pushing the wheelbarrow uphill overloaded and you wont get where you want.

... lab testing ... lol

Edited October 10, 2013 by Nekhbet

[~Anne~]

Not all dogs have the capacity or ability to detect cancer I believe. Have you read the book by Alexandra Horowitz? She covers this in her book and there maybe other publications of hers that delve more deeply into it.

[PuddleDuck]

Wobbly I've got no idea about the training but had to pop in with one of these hope things are going ok, I've been thinking about you xx

[Wobbly]

Hi guys thanks so much for all the kind replies! Sorry I am so late to reply (broke my computer…. )

Good to see you too Raineth! Thanks for the support. :D I am not entirely sure how I am going to go, but it will be fun trying, and so I figured I may as well aim high. It'll be good for Jarrah, really stretching her brain tires her out enough to let her be relaxed and sleep soundly when I can't take on long walls and play frisbee with her. Yeah, definitely not giving her the idea she'll get rewarded for indicating actual people at all, I wouldn't want to risk her indicating when that person hasn't specifically requested, otherwise it would be intrusive and I don't want that. Definitely scented items only! Though I wonder if you couldn't remedy the situation with your dog checking your daughter's sugar levels with some professional help from someone knowledgeable and experienced in detection work, so it's nonintrusive for your daughter? The idea is a good one, it might be doable?

Thanks for the well wishes BC Crazy! i got excellent news today, MRI scan says tumours have shrunk A LOT since chemo began! Yay I better get a crackle on with this while I still visit the chemo clinic every week to get samples! I wonder how fresh the samples have to be. I believe the smell stays for a very long time (at least in the bags) so I'm OK there.

Hi Huski! Hope you and Wis are well. :D I'd love more info, Steve made Jarrah a good dog! I had already definitely decided we need to see Steve at some point for this! Why do you think pairing with food is bad? Jarrah can get a bit of excitement induced neural static when the toys come out, which can make it hard for her to learn new things for a toy reward, I've always taught behaviours with food initially because of that.

Hey LBD, yes Steve Austin is really well known for scent detection work. Does he do individual work with private clients? For some reason I thought he was more about consults for official enterprises; quarantine, bomb detection and the like? I think he owns Terrigal Pet Resort, where I send Jarrah for boarding when needed, so his staff there knows Jarrah well., maybe they can put in a good word for her! Thank you for mentioning, I hadn't considered the idea he may take on private type jobs with essentially clueless pet owners like myself. "D

Hi Bedazzled, I really need hands on help! Thank you for the tip, I will look into that too!

Hey Nekhbet, they are so literal aren't they! Sometimes it's great, sometimes not so much…

Actually it's very easy to stuff up even on the easy stuff.

Yep, this is exactly my concern and why I realise I need help! Lots of help! I am such a rookie dog trainer that I don't even know the potential pitfalls, let alone how to avoid them. You're right we are definitely still at bare bones level here, as novice as novice can possibly be, and she does find concentration exhausting, especially learning new things.

I see a lot of potential for me to set us back due to my inexperience, which would be a silly thing for me to do when someone knowledgeable and experienced could easily identify elementary mistakes we might be developing and help me avoid them. I am very terrible at Facebook, but I do have an account, so I will definitely contact you through there! Thank you so much for the offer. "D

And I agree - this is the sort of situation where like explosives, it is serious, in the case of cancer a false negative could potentially give people the idea they don't need to do a radiology screen, and even if they only put it off 3 months, that time could well be the difference between curable and incurable. Absolutely can't be giving people an indication they are in the clear when hey are not. As such an inexperienced handler, assuming I manage to get a good accuracy rating in training (not a given!) it doesn't mean that will necessarily translate to the field in the way I am imagining right now, or even at all. Consulting a variety of people with real knowledge and experience in detection will be really, really necessary - you, Steve Courtenay, Steve Austin, Karen Pryor, and more.

Hi Anne, thank you for the recommendation, I haven't read her book, but that's something I hadn't considered! I will look for her books. Thank you so much for mentioning! If it doesn't work out that's OK, but I am really keen to give it a go.

Hey Puddleduck! Thank you and hugs back! I have been thinking of you too, and hoping the warmer weather is helping you feel a lot better than winter cold.

Thankyou everyone for all your input and advice it is appreciated! I anticipate this will take me some time to do, and a lot of new information for me, so i won't be able to act on all suggestions immediately, baby steps for me first. I am going to need a LOT of help, so all these suggestions will likely be put in practice. I am honestly not sure how far I will get, but even worst case scenario, the effort i put in will at the very least raise awareness of the fact it IS possible, get people talking about i, and thinking about it. The further I get of course, the more the scope for the idea to inspire others too, so I am aiming high at this point!

Edited October 14, 2013 by Wobbly

[huski]

Hi Huski! Hope you and Wis are well. :D I'd love more info, Steve made Jarrah a good dog! I had already definitely decided we need to see Steve at some point for this! Why do you think pairing with food is bad? Jarrah can get a bit of excitement induced neural static when the toys come out, which can make it hard for her to learn new things for a toy reward, I've always taught behaviours with food initially because of that.

Sorry I didn't mean don't use food rewards, I will PM you :)

Edited October 15, 2013 by huski

[Sparden]

Hi, I am just starting out in this sport and have already seen a few different training methods.

I just read your comment below. Can you please expand on why?

What exactly do you mean by mixing the reward with the target odour?

Thanks.

Nose work is becoming quite popular of late, there are a different ways to train it but there are some ways that are better than others (I personally would avoid any classes that train it by mixing the reward and target odour together.)