High-speed imaging and artificial intelligence are helping us understand how insect wings work

About 350 million years ago, our planet witnessed the evolution of the first flying creatures. They are still around, and some of them continue to annoy us with their buzzing. While scientists classify these creatures as winged insects, the rest of the world simply calls them winged insects.

There are many aspects of insect biology, especially flight, that remain a mystery to scientists. The first is simply how they move their wings. The insect wing joint is a specialized joint that connects an insect's wings to its body. They are composed of five interconnected sheet-like structures called sclerites. When these plates are moved by the underlying muscles, it makes the insect's wings flap.

Until now, it has been difficult for scientists to understand the biomechanics that govern scleral movement even using advanced imaging techniques. “The sclera within the wing joint is so small and moves so quickly that its mechanical action in flight has not been accurately captured despite efforts using stroboscopic imaging, high-speed videography, and X-ray tomography,” said Michael Dickinson, Zarem Professor of Biology and Bioengineering at the Institute. California Tech (Caltech), for Ars Technica.

As a result, scientists are unable to visualize exactly what happens on the small scale within the wing joint as they fly, preventing them from studying insect flight in detail. However, a new study by Dickinson and his team has finally revealed the workings of the sclera and hinges of an insect's wing. They captured the wing movement of fruit flies (Black-bellied fruit fly) analyzed 72,000 recorded wing beats using a neural network to decode the role that individual sclerites play in shaping insect wing movement.

Understand the detail of an insect's wing

The biomechanics governing insect flight are quite different from those in birds and bats. This is because wings in insects did not develop from limbs. “In the case of birds, bats and pterosaurs, we know exactly where wings came from evolutionarily because all of these animals fly with their forelimbs. They mainly use their arms to fly. As for insects, it's completely different. They evolved from six-legged creatures and retained all the legs.” All six, however, have added flapping appendages to the dorsal side of their bodies, and it is unknown where those wings came from.

Some researchers suggest that insect wings came from Gill-like appendages Found in ancient aquatic arthropods. Others believe that the wings originated from “Cloves“A special growth found on the legs of ancient crustaceans that were the ancestors of insects. This debate is still ongoing, so its evolution can't tell us much about how the hinge and sclera function.”

Understanding the mechanics of arthropods is crucial because this is what makes insects such effective flying creatures. It enables them to fly at amazing speeds for their body sizes (some insects can fly at 33 miles per hour) and to display great maneuverability and stability in flight.

“The insect wing joint is arguably among the most sophisticated and evolutionarily important skeletal structures in the natural world,” according to the study authors.

However, imaging the activity of four of the five sclerites that make up the hinge was impossible due to its size and the speeds at which it moves. Dickinson and his team used a multidisciplinary approach to overcome this challenge. They designed a device equipped with three high-speed cameras that record the activity of tethered fruit flies at a rate of 15,000 frames per second using infrared light.

They also used a calcium-sensitive protein to track changes in the activity of the insects' guiding muscles as they flew (calcium helps stimulate muscle contractions). “We recorded a total of 485 flight sequences from 82 flies. After excluding a subset of wing beats from the sequence when the fly stopped flying or flew with an abnormally low wing beat frequency, we obtained a final dataset of 72,219 wing beats. NB.

Next, they trained a machine learning-based convolutional neural network (CNN) using 85% of the dataset. “We used a CNN model to study the transformation between muscle activity and wing movement by performing a set of virtual manipulations, exploiting the network to perform experiments that would be difficult to perform on actual flies,” they explained.

In addition to the neural network, they also developed an encoder-decoder neural network (an architecture used in machine learning) and fed it data related to guide muscle activity. While the CNN model can predict wing movement, the encoder/decoder can predict the action of individual rigid muscles during wing movement. Now, it's time to check if the data they predicted is accurate.