首页|University of Wisconsin Madison Reports Findings in Machine Learning (Machine le arning-assisted mid-infrared spectrochemical fibrillar collagen imaging in clini cal tissues)
University of Wisconsin Madison Reports Findings in Machine Learning (Machine le arning-assisted mid-infrared spectrochemical fibrillar collagen imaging in clini cal tissues)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News – New research on Machine Learning is th e subject of a report. According to news reporting originating from Madison, Wis consin, by NewsRx correspondents, research stated, “Label-free multimodal imagin g methods that can provide complementary structural and chemical information fro m the same sample are critical for comprehensive tissue analyses. These methods are specifically needed to study the complex tumor-microenvironment where fibril lar collagen’s architectural changes are associated with cancer progression.” Our news editors obtained a quote from the research from the University of Wisco nsin Madison, “To address this need, we present a multimodal computational imagi ng method where mid-infrared spectral imaging (MIRSI) is employed with second ha rmonic generation (SHG) microscopy to identify fibrillar collagen in biological tissues. To demonstrate a multimodal approach where a morphology-specific contra st mechanism guides an MIRSI method to detect fibrillar collagen based on its ch emical signatures. We trained a supervised machine learning (ML) model using SHG images as ground truth collagen labels to classify fibrillar collagen in biolog ical tissues based on their mid-infrared hyperspectral images. Five human pancre atic tissue samples (sizes are in the order of millimeters) were imaged by both MIRSI and SHG microscopes. In total, 2.8 million MIRSI spectra were used to trai n a random forest (RF) model. The other 68 million spectra were used to validate the collagen images generated by the RF-MIRSI model in terms of collagen segmen tation, orientation, and alignment. Compared with the SHG ground truth, the gene rated RF-MIRSI collagen images achieved a high average boundary -score (0.8 at 4 -pixel thresholds) in the collagen distribution, high correlation (Pearson’s 0.8 2) in the collagen orientation, and similarly high correlation (Pearson’s 0.66) in the collagen alignment.”
MadisonWisconsinUnited StatesNorth and Central AmericaCollagenCyborgsEmerging TechnologiesExtracellular Ma trix ProteinsFibrillar CollagensMachine LearningProteinsScleroproteins
NETL
NSTL
