Wyniki wyszukiwania - BazTech

Ograniczanie wyników

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Sortuj według:

Ogranicz wyniki do:

A semi-automated annotation algorithm based on weakly supervised learning for medical images

Li Hailiang, Zhang Bin, Zhang Yu, Liu Weiwei, Mao Yijun, Huang Jiacheng, Wei Linfeng

Biocybernetics and Biomedical Engineering

2020

Vol. 40, no. 2

787--802

For medical image recognition, deep learning requires a massive training set, while anno-tation work is a tedious and time-consuming process because of the high technical thresh-old. Furthermore, it is difficult to guarantee annotation accuracy due to the knowledge, skills, and status of the annotator. In this research, we propose a semi-automated annota- tion model based on weakly supervised learning. Moreover, a target-level annotation method is proposed based on weakly supervised learning that is guided by machine learning. The machine learning method is used to screen the regions of interest (RoIs), whose semantic feature vectors are extracted by the deep learning method. Then, the machine learning method is used to cluster them, and the RoIs are finally classified and labeled by a distance comparison. Therefore, this model achieves target-level semi-auto- mated annotation by only using image-level annotations. We applied this method to ultrasound imaging of thyroid papillary carcinoma. The experiments demonstrate the potential of this new methodology to reduce the workload for pathologists and increase the objectivity of diagnoses. We find that 89.8% of papillary thyroid carcinoma regions can be detected automatically, while 82.6% of benign and normal tissue can be excluded without the use of any additional immunohistochemical markers or human intervention.

Direct inversion for sensitive elastic parameters of deep reservoirs

Li Shengjun, Gui Jinyong, Gao Jianhu, Wang Shangxu, Li Hailiang

Acta Geophysica

2019

Vol. 67, no. 5

1329--1340

The deep reservoir is usually a type of tight reservoir with high pressure, high stress, low permeability and low porosity. The elastic parameters including Poisson’s ratio and Young’s modulus are important sensitive parameters to the tight reservoir, and the Gassmann fuid term is frequently used in the feld of fuid identifcation as a highly sensitive fuid factor. Such parameters can be obtained by the common prestack seismic inversion method, but not directly. It must frst invert for other elastic parameters and then convert them into the Poisson’s ratio, Young’s modulus and Gassmann fuid term by some formula. The errors will be accumulated in the conversion step, and the inversion results will have a large deviation. We propose a one-step inversion method to solve this problem. Firstly, a new form of P-wave refection coefcient equation in terms of Poisson’s ratio, Young’s modulus and Gassmann fuid term is derived which can directly establish the functional relationship between the P-wave refection coefcient and these elastic parameters. Considering seismic data of deep reservoir generally have a lower signal-to-noise ratio (S/N) and the partial angle stack gather has a higher S/N than single angle gather, we then derive a stack impedance equation which is suitable for the partial angle stack gather. By using three stacked impedance inversion data with diferent angle stack ranges, we can directly get the Poisson’s ratio, Young’s modulus and Gassmann fuid term simultaneously. Model and real data tests both prove that the one-step direct inversion method can reduce the cumulative errors efectively and has higher inversion accuracy.