Artificial intelligence technologies can identify tumour by high-resolution medical imagery. Scientists have used the technology to analyse similar scans of dinosaur fossils. The study has been published in Frontiers in Earth Science.
Dinosaurs’ fossil relies on the morphology of the preserved remains of the animals. Scientists wanted to study the interior structure of a specimen which needs cutting thin sections. It will destroy the sample in the process. But high-resolution scanning technologies like X-ray computed tomography has changed this scene. This reconstructs internal structures in three dimensions by using radiation and digital software.
CT technology will preserve specimens. It will also generate very useful data. The scans differentiate different materials. This includes fossilized bones versus the rock encasing it. This is based on the absorption of X-ray radiation. Densities will make it difficult to determine where one object begins and another ends. So, scientists have to rely on manual segmentation. This is a labor-intensive process for classifying similar sections of an image.
Putting AI to the test
AI can do image segmentation in minutes. If a computer can classify sections voxel by voxel on par with a trained professional. Scientists tried to find out using different types of deep neural networks. This is a type of AI model that mimics the human brain.
Scientists tested the AI systems using more than 10,000 CT scans of three well-preserved embryonic skulls. This is a smaller relative to the more familiar genus Triceratops. The fossils had been recovered in the 1990s.
But that did not perform quite as well as a human. The accuracy and processing speed showed that deep neural networks can reduce the time to differentiate fossils from rock matrices.
Need for bigger data, better algorithms
The AI also helped establishing research standards. Though there is more work to do before that happens. As Protoceratops test struggled to perform well on other dinosaur fossils.