Improving the upper atmospheric temperature accuracy of the ground-based instrument by eliminating noise ways.

In view of the special properties of the upper atmosphere at the altitude of 80-120 km, a ground-based passive remote sensing instrument ground-based airglow volume emission rate and temperature imaging interferometer (GBAVTII) is built to detect the atmospheric temperature used O 2(0-1) spectral line of night airglow at the altitude of 94 km. In the process of photographing the upper atmosphere airglow with the GBAVTII, the stray light (white noise) such as moonlight, city lights, and starlight will be affected. In this paper, the theoretical expression of denoising is derived based on the rotational line temperature measurement of diatomic O 2(0-1) airglow. Through a slight adjustment of different parameters in the forward equation of the GBAVTII and noise reduction in laboratory flat-field fine calibration, and other denoising methods in the GBAVTII image processing process, the maximum accuracy of the GBAVTII detection of the upper atmospheric temperature is enhanced to 2.4 K. Also, the minimum error of the GBAVTII detecting data with the satellite instrument sounding of atmosphere using broadband emission radiometry is 0.4 K. Thus, the absolute accuracy of the GBAVTII in detecting the upper atmospheric temperature can be improved to ±(0.4-2.4)K through the theory and method studied in this paper.

Topology-enhanced molecular graph representation for anti-breast cancer drug selection.

BACKGROUND: Breast cancer is currently one of the cancers with a higher mortality rate in the world. The biological research on anti-breast cancer drugs focuses on the activity of estrogen receptors alpha (ER[Formula: see text]), the pharmacokinetic properties and the safety of the compounds, which, however, is an expensive and time-consuming process. Developments of deep learning bring potential to efficiently facilitate the candidate drug selection against breast cancer. METHODS: In this paper, we propose an Anti-Breast Cancer Drug selection method utilizing Gated Graph Neural Networks (ABCD-GGNN) to topologically enhance the molecular representation of candidate drugs. By constructing atom-level graphs through atomic descriptors for each distinct compound, ABCD-GGNN can topologically learn both the implicit structure and substructure characteristics of a candidate drug and then integrate the representation with explicit discrete molecular descriptors to generate a molecule-level representation. As a result, the representation of ABCD-GGNN can inductively predict the ER[Formula: see text], the pharmacokinetic properties and the safety of each candidate drug. Finally, we design a ranking operator whose inputs are the predicted properties so as to statistically select the appropriate drugs against breast cancer. RESULTS: Extensive experiments conducted on our collected anti-breast cancer candidate drug dataset demonstrate that our proposed method outperform all the other representative methods in the tasks of predicting ER[Formula: see text], and the pharmacokinetic properties and safety of the compounds. Extended result analysis demonstrates the efficiency and biological rationality of the operator we design to calculate the candidate drug ranking from the predicted properties. CONCLUSION: In this paper, we propose the ABCD-GGNN representation method to efficiently integrate the topological structure and substructure features of the molecules with the discrete molecular descriptors. With a ranking operator applied, the predicted properties efficiently facilitate the candidate drug selection against breast cancer.

Temporal information-guided dynamic dual-tracer PET signal separation network.

PURPOSE: The difficulty of dynamic dual-tracer positron emission tomography (PET) technology is to separate the complete single-tracer information from mixed dual-tracer. Traditional methods cannot separate single-injection single-scan dynamic dual-tracer PET images. In this paper, we propose a deep learning framework based on gated recurrent unit (GRU) network and evaluate its performance with simulation experiments and realistic monkey data. METHODS: The proposed single-scan dynamic dual-tracer PET image separation network consists of three parts, including encoder, separation, and decoder module. Encoder part is to map the mixed time activity curves (TACs) from the low-dimensional space to the high-dimensional space to get mixed weight vector matrix. Separation part is to capture the temporal information of mixed weight vector matrix using bi-directional GRU (bi-GRU) layer to obtain the single-tracer masks, and the decoding part remaps the high-dimensional single-tracer weight vector matrix to the low-dimensional space to obtain two separated single tracers. RESULTS: In the simulation experiments under different tracers, phantoms, noise levels, arterial input function (AIF), and k-parameter with Gaussian random, compared to the stacked auto encoder network and traditional background subtraction method, GRU-based network has better performance with low bias and mean squared error. In the realistic study, the image results of GRU network have higher mean structural similarity and peak signal to noise ratio. CONCLUSIONS: This study demonstrates the feasibility of temporal information-guided neural network in single-injection single-scan dynamic dual-tracer PET images separation. The GRU-based network uses TAC temporal information without AIFs to make the separation results more robust and accurate, which significantly outperforms state-of-the-art method qualitatively and quantitatively.

Ballistic imaging through an intense scattering medium using a supercontinuum with a roundabout spatial gate.

We propose a new ballistic imaging method that is capable of imaging an object through an intense scattering medium. In this method, a femtosecond supercontinuum and a roundabout spatial gate were used to suppress speckles and filter background noise, respectively. The roundabout spatial gate extracts ballistic light and avoids low-pass spatial filtering to ensure the high resolution of images. The experimental results showed that even when the optical depth of the scattering medium reached 17, the images extracted by the method had improved identifiability and contrast.

Speckle-suppressed full-field imaging through a scattering medium using a supercontinuum.

We demonstrate speckle-suppressed full-field imaging through a scattering medium using incoherent supercontinuum (SC) illumination. The patterns in images obtained using SC illumination were found to be more easily identifiable than those in images acquired using coherent direct laser illumination due to the speckle suppression. Even when the optical depth reached 12.3, the patterns remained identifiable. As one of the potential applications, we also demonstrated the imaging for a high-pressure diesel spray using SC illumination.

Long-working-distance microscopic imaging in a turbid medium by use of an ultrafast optical Kerr gate.

We demonstrate a long-working-distance microscopic imaging of hidden objects in a turbid medium by use of an ultrafast optical Kerr gate (OKG). The results show that the working distance and the spatial resolution of the long-working-distance microscopic imaging system have been increased simultaneously compared with those of the conventional 4f OKG imaging systems. A compound lens consisting of a long-focus achromatic doublet and a microscope objective is used to increase the long working distance and ensure the sufficient spatial resolution. The microscopic OKG imaging system with a working distance of 245 mm and a maximal spatial resolution of approximately 7 µm has been performed.

Optimum heterodyning angle for heterodyned optical Kerr gated ballistic imaging.

We perform heterodyned optical Kerr gated (HOKG) ballistic imaging of an object hidden behind a turbid medium using a femtosecond laser. The experimental results show that an optimum heterodyning angle should be selected to acquire the highest spatial resolution of the HOKG imaging system. The optimum heterodyning angle depends on the scattering parameters of the turbid media, and it decreases with increasing optical density or decreasing thickness of the turbid medium.

Sharpness-enhanced ultrafast imaging by using a biased optical Kerr gate.

We proposed a novel biased optical Kerr gated imaging (BOKGI) method for ultrafast imaging. The imaging performance of the BOKGI system has been investigated. Experimental results showed that by using the BOKGI, the high spatial frequency components of the detected object could be effectively retrieved, which are often filtered by the photo-induced soft aperture in a conventional OKGI system. Comparing with the conventional OKGI method, the BOKGI method could enhance the sharpness of images and provide a higher spatial resolution of the imaging system. In addition, the influence of the biased angle on the BOKGI performance has been also investigated.