![BugBitten How AI may help you depend your (mosquito) eggs earlier than they hatch BugBitten How AI may help you depend your (mosquito) eggs earlier than they hatch](https://keiseronlineuniversity.com/wp-content/uploads/sites/11/2023/11/iStock-811236334-2-620x342.jpg)
[ad_1]
A brand new device for counting mosquito eggs developed by Javed et al. exhibits the potential purposes AI can have in parasitology analysis.
![© metamorworks / Getty Images / iStock](https://blogs.biomedcentral.com/bugbitten/wp-content/uploads/sites/11/2023/11/iStock-811236334-2-620x342.jpg)
AI(Synthetic Intelligence) idea, 3D rendering, summary picture visible *** Native Caption *** © metamorworks / Getty Pictures / iStock
With the discharge of ChatGPT late final yr, discussions about synthetic intelligence moved from the realms of Large Tech and Science Fiction to mainstream headlines. Since then, debates in regards to the potential makes use of and risks of those instruments haven’t left the general public consciousness.
In a single ironic episode, a outstanding Science Fiction journal needed to shut down its submission portal final winter after being inundated with AI-generated quick tales. Extra not too long ago, simply this week US President Joe Biden issued an government order laying out new guidelines for AI.
It’s not all unhealthy, although. AI and deep studying strategies may help with simplifying and innovating. For instance, within the sciences it could assist help with processing massive datasets and different info.
Which brings me to the principle matter of immediately’s BugBitten weblog: a paper not too long ago printed within the journal Parasites & Vectors, EggCountAI: a convolutional neural network-based software program for counting of Aedes aegypti mosquito eggs by Javed et al. On this article, the authors focus on a free automated synthetic intelligence (AI)-based device they developed for counting the eggs of Aedes aegypti mosquitoes.
Why develop one thing like that?
As a result of, as Javed et al. report, “Aedes aegypti mosquitos are the first vector of a number of medically vital viruses, together with dengue, Zika, yellow fever, and chikungunya.” And though there was loads of analysis into how you can restrict and even get rid of mosquito-transmitted pathogens, these illnesses are nonetheless a serious downside everywhere in the world. What’s extra, sooner or later they could even grow to be an even bigger concern, as local weather change worsens and we grow to be extra interconnected.
However nonetheless, why the eggs particularly?
As a result of researchers can truly glean loads of info simply from them. For instance, the variety of eggs from can be utilized to estimate the mosquito inhabitants in a area. Moreover, sure pathogens can truly change the conduct of the mosquitoes, together with their fecundity (aka, what number of eggs they lay).
Nevertheless, as you may think, manually counting tiny mosquito eggs could be fairly troublesome, particularly as samples can have excessive ranges of eggs. That’s the place the software program is available in.
![](https://blogs.biomedcentral.com/bugbitten/wp-content/uploads/sites/11/2023/11/Pic-PARV-blog-300x225.png)
Are you able to depend what number of eggs are current in these figures? From Javed, N., López-Denman, A.J., Paradkar, P.N. et al. EggCountAI: a convolutional neural network-based software program for counting of Aedes aegypti mosquito eggs. Parasites Vectors 16, 341 (2023). https://doi.org/10.1186/s13071-023-05956-1
Javed et al. created EggCount AI and educated it on egg samples that they had collected. They used two completely different datasets for the coaching, one for “micro” photos (with egg counts starting from 13 to 167) and one for “macro” photos (starting from 133 to 624). Afterwards, they put the device to the check to see the way it in comparison with related software program, validating the counts with a handbook depend (fortunately aided by ImageJ software program and never a magnifying glass!).
General, EggCount AI carried out remarkably nicely. For micro photos, it had an accuracy of 98.88%, and for macro photos, 96.06%, outperforming related instruments. As Javed et al. report, by producing correct egg counts, it could lead researchers “to a deeper understanding of mosquito behaviour and health traits.” What’s extra, for anybody curious in regards to the device, EggCount AI is free and obtainable on the article web page.
So, what does the long run maintain for AI for science? It’s powerful to say for certain, however clearly it is going to play a big function transferring ahead for analysis. And that’s why Parasites & Vectors not too long ago launched a brand new article assortment, Synthetic intelligence, parasites, and parasitic illnesses, welcoming articles reporting on the makes use of of AI and deep studying in parasitology analysis, resembling EggCount AI. We encourage anybody concerned about AI purposes in science to verify in periodically as extra papers are added to the gathering.
[ad_2]