The deterioration of alpine grassland has great impact on ecosystem services in the alpine region of Qinghai-Tibetan Plateau. However, the effect of grassland degradation on ecosystem services and the consequence of grassland deterioration on economic loss still remains a mystery. So, in this study, we assessed four types of ecosystem services following the Millennium Ecosystem Assessment classification, along a degradation gradient. Five sites of alpine grassland at different levels of degradation were investigated in Guoluo Prefecture of Qinghai Province, China. The species composition, aboveground biomass, soil total organic carbon (TOC), and soil total nitrogen (TN) were tested to evaluate major ecological services of the alpine grassland. We estimated the value of primary production, carbon storage, nitrogen recycling, and plant diversity. The results show the ecosystem services of alpine grassland varied along the degradation gradient. The ecosystem services of degraded grassland (moderate, heavy and severe) were all significantly lower than non-degraded grassland. Interestingly, the lightly degraded grassland provided more economic benefit from carbon maintenance and nutrient sequestration compared to non-degraded. Due to the destruction of the alpine grassland, the economic loss associated with decrease of biomass in 2008 was $198/ha. Until 2008, the economic loss caused by carbon emissions and nitrogen loss on severely degraded grassland was up to $8 033/ha and $13 315/ha, respectively. Urgent actions are required to maintain or promote the ecosystem services of alpine grassland in the Qinghai-Tibetan Plateau.
Background
Pathological diagnosis of glioma subtypes is essential for treatment planning and prognosis. Standard histological diagnosis of glioma is based on postoperative hematoxylin and eosin stained slides by neuropathologists. With advancing artificial intelligence (AI), the aim of this study was to determine whether deep learning can be applied to glioma classification.
Methods
A neuropathological diagnostic platform is designed comprising of a slide scanner and deep convolutional neural networks (CNNs) to classify five major histological subtypes of glioma to assist pathologists. The CNNs were trained and verified on over 79,990 histological patch images from 267 patients. A logical algorithm is used when molecular profiles are available.
Results
A new model of the squeeze-and-excitation block DenseNet with weighted cross-entropy (named SD-Net_WCE) is developed for the glioma classification task, which learns the recognizable features of glioma histology CNN-based independent diagnostic testing on data from 56 patients with 17,262 histological patch images demonstrated patch level accuracy of 86.5% and patient level accuracy of 87.5%. Histo-patholgical classifications could be further amplified to integrated neuropathological diagnosis by two molecular markers (IDH and 1p/19q).
Conclusion
The model is capable of solving multiple classification tasks and can satisfactorily able to classify glioma subtypes. The system provides a novel aid for the integrated neuropathological diagnostic workflow of glioma.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.