Screening for colorectal cancer remains a challenge, even with recent developments that have been decreasing the rate of colorectal cancer incidences. However, the proportion of young people affected by colorectal cancer has increased. Because screening for colorectal cancer remains difficult, especially with the discrepancies in the quality of screening and the potential for the screening to be invasive, researchers have been seeking non-invasive screening methods for the detection of colorectal cancer. This challenge to develop effective screening techniques is crucial, considering that the survival rate for colorectal cancer, if caught early, can be as high as 90%.

One of the non-invasive screening methods used for colorectal cancer is the odor screening test. The odor screening test is what it sounds like—a test for the presence of a disease, based on a disease’s distinct odor. This odor is a collection of volatile organic compounds (VOCs) that are released out into the bloodstream. Volatile organic compounds are byproducts of metabolism, which involves the body carrying out its functions, breaking down molecules, and forming new molecules. Examples of volatile organic compounds include isoprene, alcohols, acids, and alkenes; often times, VOCs can damage the environment and, especially, the human body (Kesselmeier and Staudt). Each individual releases a unique set of VOCs; the VOCs emitted depend on external factors that influence a person’s metabolism (De Boer). As cancer is an “external factor”, it can influence a set of VOCs and the odor associated with itself.

A combination of volatile organic compounds can indicate the presence of colorectal cancer because it is influenced by changes in metabolism.

With the distinct odor of colorectal cancer, VOCs can be used as an indication of the presence of colorectal cancer; we can use scent detection to screen for colorectal cancer (De Boer). Utilizing this method, we can detect colorectal cancer through canine scent detection; because canines have an incredible sense of smell, they have the ability to distinguish between odors that are seemingly unnoticeable, like the “scent” of diseases (Murphy). Canine scent detection is highly accurate; compared to the conventional colonoscopy, canine scent detection, in analyzing breath samples for specific odors associated with various cancers, has the potential to be more sensitive (Sonoda, In fact, researchers found the ability of canine scent detection to accurately analyze odor samples, from 33 groups of clinical trial participants, to be 0.99, indicating extremely high precision (Sonoda, Additionally, C.elegans, a type of nematode, can also detect certain odors released due to cancer (Hirotsu). However, these methods are not foolproof, as mistakes can still occur. For this reason, companies are looking to develop odor analysis technologies that accurately diagnose cancers. For example, the company Owlstone has invested in such technologies that can diagnose different cancers based on their odors (“Pharmaceutical Applications”). In particular, it has been working on a machine that has gold nanoparticle and carbon nanotube sensors. These sensors can be manipulated to be attracted to specific odor molecules, so that the machine can analyze a patient’s breath and tell us whether these odor molecules “match” with a certain cancer (Murphy). The specific VOCs generally associated with colorectal cancer include benzene compounds and 1-iodo-nonane (Malagu, et. al). While we can detect the presence of these VOCs through odor analysis technologies, we can also detect them through gas sensors that examine how the environment influences the body’s makeup (Malagu, With developments in technology, institutions can now diagnose patients without an invasive means.

Owlstone, Inc. is developing pharmaceutical applications that involve odor analysis to screen for colorectal cancer.

And, ultimately, these advancements mean that we are advancing our screening methods to the extent in which we can begin to see some changes in the diagnosis of colorectal cancer. It is important to emphasize that colorectal cancer does not just happen among older adults; contrary to a common belief, young adults are also affected by colorectal cancer. We must destigmatize the screening of colorectal cancer, so that more people can be screened for this disease. With odor sensing technologies, we can really start seeing a change in the way we approach non-invasive screening methods that are enhanced by technology to increase efficiency. And these technologies can be used for the detection of other cancers, such as breast cancer. Current challenges like the DREAM Challenge are encouraging people to develop software that accurately interprets mammograms, so we can screen for breast cancer more efficiently (Schaffter and Stolovitzky). Examples like this truly show the potential to make screening methods more applicable, especially with the growth of the technology sector.

Machine learning plays a crucial role in cancer screening and is expanding the realms of non-invasive cancer screening.


De Boer, N K, et al. “The Scent of Colorectal Cancer: Detection by Volatile Organic Compound Analysis.” Clinical Gastroenterology and Hepatology : the Official Clinical Practice Journal of the American Gastroenterological Association., U.S. National Library of Medicine, 13 May 2014,

“Deep Machine Learning: Can It Lead to Cancer Screening Improvements?” Medium, KP Washington Health Research Institute, 25 May 2017,

Hirotsu, Takaaki, et al. “A Highly Accurate Inclusive Cancer Screening Test Using Caenorhabditis Elegans Scent Detection.” PLOS ONE, Public Library of Science, 11 Mar. 2015,

Kesselmeier, J., and M. Staudt. “Biogenic Volatile Organic Compounds (VOC): An Overview on Emission, Physiology and Ecology.” SpringerLink, Kluwer Academic Publishers, May 1999,

Malagù, Cesare, et al. “Chemoresistive Gas Sensors for the Detection of Colorectal Cancer Biomarkers.” Sensors (Basel, Switzerland), MDPI, 13 Oct. 2014,

Murphy, Kate. “One Day, a Machine Will Smell Whether You’Re Sick.” The New York Times, The New York Times, 1 May 2017,

“Pharmaceutical Applications.” Owlstone, Owlstone, Inc.,

Schaffter, Thomas, and Gustavo Stolovitzky. “DREAM Challenge Results: Can Machine Learning Help Improve Accuracy in Breast Cancer Screening?” IBM Blog Research, IBM Corporation, 14 June 2017,

Sonoda, Hideto, et al. “Colorectal Cancer Screening with Odour Material by Canine Scent Detection.” Gut, BMJ Group, June 2011,