This article has been reviewed according to Science X's editorial process and policies. Editors have highlighted the following attributes while ensuring the content's credibility: A new doctoral dissertation by Omar Anwar Elkadi in the field of Applied Physics on near-infrared spectroscopy for the continuous monitoring of tissue-engineered cartilage will be defended at the Faculty of Science, Forestry and Technology, Kuopio campus, of the University of Eastern Finland on Oct. 30. My doctoral research investigated how near-infrared spectroscopy, combined with machine learning, can monitor the development of tissue-engineered cartilage without destroying the samples.
Cartilage has a limited ability to repair itself, and growing replacement tissue in the laboratory is a promising approach to treating cartilage damage. However, laboratory-grown tissues develop at different rates, even under the same conditions. Conventional tests often require samples to be destroyed, making it difficult to follow the same tissue over time.
The research examined how light-based measurements can provide information about cartilage maturity, with a focus on the reliability of these assessments. Such a reliable monitoring approach is important for the reproducible production of cartilage grafts that could help address the unmet need for effective treatments for cartilage defects. The thesis showed that spectroscopy-based machine learning models can provide information about tissue-engineered cartilage maturity.
Models based on light measurements estimated indicators of cartilage matrix formation and cell content. These estimates enabled the classification of the laboratory-grown tissues according to their relative maturity. The research also addressed a key challenge of this approach: its nonspecific nature.
It demonstrated how variations in construct components other than the developing tissue can affect the measurements and bias the models' predictions of tissue development. Accounting for these effects during model development improved accuracy by reducing the associated bias. This highlights the importance of an analysis approach that considers all components of the construct, not just the developing tissue.
A setup allowing repeated measurements was also developed and tested as a proof of concept for monitoring changes in the mechanical properties of cell-free hydrogels. Spectroscopy-based models predicted these properties more accurately than models based only on culture time. They also captured changes during incubation that are difficult to follow using conventional tests performed at separate time points.
Extract — continue reading at the source.
