A new publication from a lab member
We're thrilled to share a recent publication of Michael F. Adamer, our Lead of Paediatric Applications Sarah Brüningk, Dexiong CHEN and Karsten Borgwardt.
The authors showcase a powerful new tool for feature selection 𝐒𝐩𝐈𝐧𝐎𝐩𝐭-𝐌𝐌𝐃, which exploits the concept of maximum mean discrepancy.
Imagine being able to identify key differences in data more accurately and efficiently—this is exactly what 𝐒𝐩𝐈𝐧𝐎𝐩𝐭-𝐌𝐌𝐃 achieves! From medical images to genetic data, this method excels in selecting the most informative features, leading to superior performance in various classification tasks.
Why is this important?
🔹 Better Disease Detection: We give an example to pinpoint Alzheimer's disease relevant regions in brain scans.
🔹 Enhanced Data Analysis: Outperforms traditional methods across multiple datasets, making it a versatile tool for researchers.
🔹 Future Potential: Can be expanded to analyze structured data like graphs and time series, opening new avenues in data science.
This innovation will support both data scientists and medical researchers, paving the way for more accurate and reliable analyses.
Read more about this fascinating study!
🔗 external page Link to publication
𝗔𝘂𝘁𝗵𝗼𝗿𝘀 𝗮𝗿𝗲 𝗴𝗿𝗮𝘁𝗲𝗳𝘂𝗹 𝘁𝗼 𝘁𝗵𝗲 𝗔𝗹𝘇𝗵𝗲𝗶𝗺𝗲𝗿’𝘀 𝗗𝗶𝘀𝗲𝗮𝘀𝗲 𝗡𝗲𝘂𝗿𝗼𝗶𝗺𝗮𝗴𝗶𝗻𝗴 𝗜𝗻𝗶𝘁𝗶𝗮𝘁𝗶𝘃𝗲 (𝗔𝗗𝗡𝗜) 𝗰𝗼𝗻𝘀𝗼𝗿𝘁𝗶𝘂𝗺. Data collection and sharing for this project was funded by ADNI (National Institutes of Health Grant U01 AG024904) and DOD ADNI (Department of Defense award number W81XWH-12-2-0012). external page https://adni.loni.usc.edu/