Abstract-Detecting melanoma with a terahertz spectroscopy imaging technique

Dandan Li, Zhongbo Yang, Ailing Fu, Tunan Chen, Ligang Chen, Mingjie Tang, Hua Zhang, Ning Mu, Shi Wang, Guizhao Liang,  Huabin Wang

https://www.sciencedirect.com/science/article/abs/pii/S1386142520302079

Transmission mode terahertz time-domain spectroscopy system was employed to image BALB/c mouse skin tissue slices containing melanoma. The melanoma was unambiguously identified in the frequency region of 0.6–1.8 THz because melanoma has a higher refractive index as well as a higher absorption coefficient than the normal region of the skin tissue. Based on the results of hematoxylin-eosin staining and mass weighing, it was further suggested that the higher density of nucleic acids, higher water content, and lower fat content in the melanoma compared to the normal region are major factors responsible for melanoma's higher refractive index and absorption coefficient than normal tissue. The present work validates that terahertz time-domain spectroscopy imaging technique is possible to be used for the diagnosis of melanoma.

Abstract-Automatic evaluation of traumatic brain injury based on terahertz imaging with machine learning

Jia Shi, Yuye Wang, Tunan Chen, Degang Xu, Hengli Zhao, Linyu Chen, Chao Yan, Longhuang Tang, Yixin He, Hua Feng, and Jianquan Yao

https://www.osapublishing.org/oe/abstract.cfm?uri=oe-26-5-6371

The imaging diagnosis and prognostication of different degrees of traumatic brain injury (TBI) is very important for early care and clinical treatment. Especially, the exact recognition of mild TBI is the bottleneck for current label-free imaging technologies in neurosurgery. Here, we report an automatic evaluation method for TBI recognition with terahertz (THz) continuous-wave (CW) transmission imaging based on machine learning (ML). We propose a new feature extraction method for biological THz images combined with the transmittance distribution features in spatial domain and statistical distribution features in normalized gray histogram. Based on the extracted feature database, ML algorithms are performed for the classification of different degrees of TBI by feature selection and parameter optimization. The highest classification accuracy is up to 87.5%. The area under the curve (AUC) scores of the receiver operating characteristics (ROC) curve are all higher than 0.9, which shows this evaluation method has a good generalization ability. Furthermore, the excellent performance of the proposed system in the recognition of mild TBI is analyzed by different methodological parameters and diagnostic criteria. The system can be extensible to various diseases and will be a powerful tool in automatic biomedical diagnostics.

© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement