SNAPP: first medical decision-support tool for snake identification

“The first medical decision-support mobile application for snake identification based on artificial intelligence and remote collaborative expertise in herpetology”.

-Snapp team, 2018

Snakebites remain a major problem, as the second neglected tropical killer it accounts for over 100’000 human deaths, and claims 400’000 victims of disability and disfigurement globally every year [1].  Poor and rural communities in developing countries are mostly affected, due to the high diversity of venomous snakes, high burden of snakebites, and limited medical expertise and access to anti-venoms [1].

The Precision One-Health unit at the Institute of Global Health, University of Geneva will officially launch their project “Snapp” September 2018, hoping to tackle the existing problem of snakebites [2]. Snake identification is challenging due to snake diversity and incomplete or misleading information provided by snakebite victims or bystanders to clinicians, who generally lack the knowledge or resources in herpetology. To reduce potentially erroneous and/or delayed healthcare actions, and taking advantage of the expansion of mobile technologies in developing and emerging countries, the team developed Snapp, the first medical decision-support mobile app for snake identification based on artificial intelligence and remote collaborative expertise in herpetology. The app will combine computer vision with the expertise from a network of herpetologists to identify photos of snakes, particularly supporting victims and clinicians when urgent and reliable snake identification is needed

The project objectives are as follows:

  • Building a massive and global photo repository of venomous and non-venomous snakes
  • Developing a computer system based on machine learning and computer vision capable of identifying snakes
  • Establishing an international working group of experts in herpetology
  • Facilitating and accelerating snake identification in the field

Whilst the app itself is still in development, it shows great potential. The project integrates major global health challenges with smart technical solutions, perfectly exemplifying what Precision Global Health is all about.

 

 

Screen Shot 2018-05-16 at 13.47.33.png

Graphical representation of the medical decision-support tool for snake identification (Diagram created by V. Macalupu, F. Amrouche, XL-Zhang and C. Zhou with the supervision by Dr. R. Ruiz de Castaneda, Dr. I. Bolon during the SDG Summer School 2017 supported by Geneva-Tsinghua Initiative)

 

If you want to contribute to this humanitarian and scientific project, and you own or have access to photos of snakes (venomous, non venomous, identified or not), please contact the principle investigators of this project for collaboration: Dr R. Ruiz de Castañeda or Dr I. Bolon.

Please join us in the quest to  build a massive and global photo repository of venomous and non-venomous snakes! 

 

References 

[1] Gutiérrez JM, Calvete JJ, Habib AG, Harrison RA, Williams DJ, Warrell DA. Snakebite envenoming. Nature Reviews Disease Primers. 2017;3:17063.

[2] Isabelle Bolon . (2018). Snapp: First medical decision‐support tool for snake identification based on artificial intelligence and remote collaborative expertise. Available: https://www.unige.ch/medecine/isg/en/research/environmental-health-and-health-promotion/one-health/snapp-first-medical-decisionsupport-tool-for-snake-identification-based-on-artificial-intelligence-an. Last accessed 16/05/2018 .

Written by Nefti-Eboni Bempong

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Create a website or blog at WordPress.com

Up ↑

%d bloggers like this: