A revolutionary non-invasive method has been developed by researchers at Tokyo University of Science (TUS) to detect and visualize the risk of fatty liver disease using machine learning and near-infrared hyperspectral imaging (NIR-HSI). Fatty liver disease, also known as steatotic liver disease (SLD), affects approximately 25% of the global population and is the most common liver disorder. SLD can progress without noticeable symptoms and lead to severe conditions such as cirrhosis and liver cancer.
Traditionally, liver biopsy, an invasive procedure that involves extracting liver tissue samples, has been used to test for SLD. However, the research team at TUS has introduced NIR-HSI as a non-invasive method to visualize the total lipid content in the liver. NIR light with longer wavelengths than ultraviolet and visible light can identify fat distribution in the liver by showing absorption attributed to various organic substances.
In a new study published in Scientific Reports, the research team enhanced this method by using a machine learning model to differentiate the type of lipids present in the liver at a pixel-by-pixel level. The model can differentiate lipids based on hydrocarbon chain length (HCL) and degree of saturation (DS) of fatty acids, helping to estimate the risk of SLD progression, steatohepatitis (NASH), and SLD/NASH-associated liver cancer.
By training a support vector regression machine learning model using gas chromatography analysis of liver samples from mice fed different diets, the researchers successfully differentiated the fatty acid composition in mice livers. The researchers found correlations between the fatty acid composition and the fat content in the diets of the mice. This information was then depicted as a color map, offering a unique visual representation of fat distribution in the liver, simplifying the diagnosis of fatty liver conditions.
The non-invasive nature of this method could potentially serve as an alternative to invasive liver biopsy procedures for identifying fatty liver disease in a large population segment. In addition to healthcare applications, the framework developed by the research team could find uses in pharmacological research, metabolic imaging, and the personalization of nutritional strategies through biomarker identification and treatment response prediction.
This breakthrough has the potential to revolutionize healthcare and related research by providing a rapid and label-free technique for detecting fatty liver disease. Further research and development in this field could significantly impact liver care and improve patient outcomes.
References:
– Akino Mori, Masakazu Umezawa, Kyohei Okubo, Tomonori Kamiya, Masao Kamimura, Naoko Ohtani, Kohei Soga. DOI: https://doi.org/10.1038/s41598-023-47565-z
– Tokyo University of Science (TUS). Revolutionary Non-Invasive Method Detects and Visualizes Fatty Liver Disease Risk with Machine Learning and Near-Infrared Hyperspectral Imaging. Retrieved from [newsapi:link]