Machine Learning Could Speed Healing Process for Tissue Injuries

September 22, 2020 - Using machine learning techniques, researchers from Rice University were able to predict the quality of bioscaffold materials used to help tissue injuries heal, according to a study published in Tissue Engineering Part A.

Bioscaffolds are bonelike structures that serve as placeholders for injured tissue. These structures are porous to support the growth of cells and blood vessels that turn into new tissue and ultimately replace the implant.

The research team has been developing bioscaffolds to improve techniques to heal craniofacial and musculoskeletal wounds. This work has progressed to include sophisticated 3D printing that can make a biocompatible implant custom-fit to the site of a wound.

To improve the development process, researchers set out to determine the most important metrics of printing quality. The group found that print speed was the most important of the five metrics measured, followed in descending order of importance by material composition, pressure, layering, and spacing.

“We were able to give feedback on which parameters are most likely to affect the quality of printing, so when they continue their experimentation, they can focus on some parameters and ignore the others,” said Lydia Kavraki, a renowned authority on robotics, artificial intelligence and biomedicine and director of Rice’s Ken Kennedy Institute.

Researchers used a mass of data from a 2016 study on printing scaffolds with biodegradable propylene fumarate to develop machine learning algorithms, and then figured out what more was needed to train the models.

“This was a way to make great progress while many students and faculty were unable to get to the lab,” said Antonios Mikos, a bioengineer at Rice University.

While the researchers had previously considered bringing machine learning into the process, the COVID-19 pandemic provided the team with a unique opportunity to pursue the project.

“The students had to figure out how to talk to each other, and once they did, it was amazing how quickly they progressed,” Kavraki said.

The team explored two machine learning modeling approaches. The first was a classification method that predicted whether a given set of parameters would produce a low- or high-quality scaffold. The second was a regression-based approach that estimated the values of print-quality metrics to come to a conclusion. Both methods relied on a classic supervised learning technique that builds multiple decision trees and merges them for a more accurate and stable prediction.

The results showed that the approach could speed the development of 3D-printed bioscaffolds that help injuries heal.

The method could ultimately lead to better ways to quickly print a customized jawbone, kneecap, or a bit of cartilage on demand.

“A hugely important aspect is the potential to discover new things. This line of research gives us not only the ability to optimize a system for which we have a number of variables — which is very important — but also the possibility to discover something totally new and unexpected. In my opinion, that’s the real beauty of this work,” said Mikos.

“It’s a great example of convergence. We have a lot to learn from advances in computer science and artificial intelligence, and this study is a perfect example of how they will help us become more efficient.”

These models have the potential to someday dramatically improve tissue injury healing.

“In the long run, labs should be able to understand which of their materials can give them different kinds of printed scaffolds, and in the very long run, even predict results for materials they have not tried,” Kavraki said.

“We don’t have enough data to do that right now, but at some point we think we should be able to generate such models.”

The study demonstrates the continually evolving role of AI and machine learning in healthcare and engineering.

“Artificial intelligence has a role to play in new materials, so what the institute offers should be of interest to people on this campus,” said Kavraki. “There are so many problems at the intersection of materials science and computing, and the more people we can get to work on them, the better.”