Can AI Fight the Deadly Fungus Among Us?
CDC calls Candida auris a global health threat.
Posted May 30, 2019
In a geriatric hospital in Tokyo ten years ago, an inpatient 70-year-old Japanese woman had mysterious discharge from her external ear canal. Samples were swabbed and the DNA of the discharge was sequenced. This led to an unexpected discovery of an entirely new species of the sac fungus (ascomycetes) of the genus Candida that grows as a yeast. Examples of “friendly” sac fungi include the yeast used for brewing and baking, as well as morels and truffles. However, this particular strain, Candida auris (C. auris)—named auris, the Latin word for ear—is far from friendly, and can lead to fatalities. The Centers for Disease Control and Prevention (CDC) calls Candida auris a “serious global health threat” that “can cause invasive infections associated with up to 40% in-hospital mortality." The CDC reported 613 clinical cases in the U.S. as of the end of March 2019. Today, C. auris can be found globally on five continents in over 30 countries.
Can innovative technologies thwart the deadly fungal threat?
Artificial intelligence (AI) is being currently used by forward-thinking organizations to fight antimicrobial resistance (AMR). AMR accounts for more than 700,000 deaths globally each year, and is projected to increase to more than 10 million deaths by 2050, according to a report issued by Wellcome Trust and the UK government.
Using AI to fight antibiotic resistance is a growing trend. For example, earlier this month, Google awarded a $1.3 millon grant to the Doctors Without Borders/MSF Foundation to create a mobile smartphone app that will use AI to analyze antibiotic resistance tests and make recommendations in low-resource environments. Virginia Tech researchers have developed an AI deep learning solution (DeepARG) that predicts ARGs (antibiotic resistance genes) with a high degree of precision. And last October research scientists at the University of California San Diego announced their AI machine learning solution to identify and predict which genes cause infectious tuberculosis-causing bacteria to become resistant to antibiotics.
According to the latest CDC report, over 90 percent of C. auris infections are resistant to at least one antifungal and an estimated one-third are resistant to two or more medications.
But what, if any AI is being used for antifungal resistance?
The worldwide antifungal drugs market is projected to reach $18.2 billion by 2022, and the most growth is expected in the candidiasis segment—the segment that treats Candida infections— according to Stratistics MRC.
It is worth noting that the market dynamics for antifungal drug development is completely different from antibiotics. Large pharmaceutical companies such as Sanofi and Novartis are exiting the antibiotic markets. Antifungal drug development, however, is a growing opportunity.
For example, fungalAi, led by Dr. Michelle Ananda-Rajah at Alfred Health in Melbourne, Australia, in collaboration with Monash University, produced the world’s first AI-based platform technology (fungalAi) that can detect and help diagnose fungal diseases in hospitals in real-time.
The platform technology for fungalAi includes natural language processing, an artificial neural network deep learning algorithm that identifies fungal diseases from chest CT images, and an expert system for clinical data integration for fungal prediction. Currently, fungalAi is in clinical trials in seven health networks throughout Australia.
In addition to diagnosing and detecting fungal diseases, AI can be used for antifungal drug discovery. Deep learning can be used to shorten the drug development lifecycle in a number of areas.
For example, the pattern-recognition capabilities of deep learning can be used to predict organic reaction outcomes, drug performance in testing, and toxicity prior to clinical trials. AI can help automate molecule design, and accelerate both computer-aided synthesis and retrosynthesis based on molecular similarity.
Using innovative technologies like artificial intelligence to combat the rising deadly fungal threat is a rising opportunity for start-ups, venture capital investors, biotech companies, and large pharmaceutical companies—it is currently a wide-open playing field.
Copyright © 2019 Cami Rosso All rights reserved.
Rosso, Cami. “How Artificial Intelligence Is Accelerating Life Sciences.” Psychology Today. July 5, 2018.
Rosso, Cami. “How AI and Genomics Can Help Fight Antibiotic Resistance.” Psychology Today. November 15, 2018.
Rosso, Cami. “How AI and Smartphones May Help Fight Antibiotic Resistance.” Psychology Today. May 20, 2019.
Spivak, Emily S., Hanson, Kimberly E. “Candida auris: an Emerging Fungal Pathogen.” Journal of Clinical Microbiology.” February 2018.
CDC. “Candida auris.” Retrieved 5-25-2019 from https://www.cdc.gov/fungal/candida-auris
Encyclopaedia Britannica. “Ascomycota.” Retrieved 5-25-2019 from https://www.britannica.com/science/Ascomycota
Wellcome Trust, UK Government. “Review on Antimicrobial Resistance. Antimicrobial Resistance: Tackling a Crisis for the Health and Wealth of Nations." 2014. https://amr-review.org/
University of California – San Diego. “Machine learning identifies antibiotic resistance genes in tuberculosis-causing bacteria.” Phys.org. October 25, 2018.
fungalAI. Retrieved 5-30-2019 from https://www.fungalai.com/
Langreth, Robert. "Antibiotics Aren’t Profitable Enough for Big Pharma to Make More." Bloomberg. May 3, 2019.