Browsing by Author "Liu Z"

Now showing 1 - 5 of 5
  • Thumbnail Image
    Item
    An in-depth survey on Deep Learning-based Motor Imagery Electroencephalogram (EEG) classification
    (Elsevier BV, Netherlands, 2024-01) Wang X; Liesaputra V; Liu Z; Wang Y; Huang Z
    Electroencephalogram (EEG)-based Brain–Computer Interfaces (BCIs) build a communication path between human brain and external devices. Among EEG-based BCI paradigms, the most commonly used one is motor imagery (MI). As a hot research topic, MI EEG-based BCI has largely contributed to medical fields and smart home industry. However, because of the low signal-to-noise ratio (SNR) and the non-stationary characteristic of EEG data, it is difficult to correctly classify different types of MI-EEG signals. Recently, the advances in Deep Learning (DL) significantly facilitate the development of MI EEG-based BCIs. In this paper, we provide a systematic survey of DL-based MI-EEG classification methods. Specifically, we first comprehensively discuss several important aspects of DL-based MI-EEG classification, covering input formulations, network architectures, public datasets, etc. Then, we summarize problems in model performance comparison and give guidelines to future studies for fair performance comparison. Next, we fairly evaluate the representative DL-based models using source code released by the authors and meticulously analyse the evaluation results. By performing ablation study on the network architecture, we found that (1) effective feature fusion is indispensable for multi-stream CNN-based models. (2) LSTM should be combined with spatial feature extraction techniques to obtain good classification performance. (3) the use of dropout contributes little to improving the model performance, and that (4) adding fully connected layers to the models significantly increases their parameters but it might not improve their performance. Finally, we raise several open issues in MI-EEG classification and provide possible future research directions.
  • Thumbnail Image
    Item
    Biochar and soil properties limit the phytoavailability of lead and cadmium by Brassica chinensis L. in contaminated soils
    (Springer Nature on behalf of the Shenyang Agricultural University, 2022-12) Houssou AA; Jeyakumar P; Niazi NK; Van Zwieten L; Li X; Huang L; Wei L; Zheng X; Huang Q; Huang Y; Huang X; Wang H; Liu Z; Huang Z
    The current study investigated the effect of biochars derived from cinnamomum woodchip, garden waste and mulberry woodchip on soil phytoavailable lead (Pb), cadmium (Cd) pools, and their uptake by Chinese cabbage (Brassica chinensis L.). The biochars were produced at 450 °C of pyrolysis temperature. The contaminated soils were collected from Yunfu (classified as Udept), Jiyuan (Ustalf) and Shaoguan (Udult) cities in China at the depth of 0–20 cm and amended with biochars at the rate of 3% w/w. After mixing the soil with biochar for 14 days, the Chinese cabbage was planted in the amended soils. Then, it was harvested on the 48th day after sowing period. In Udult soil, Chinese cabbage died 18 days after sowing period in control and soils amended with cinnamomum and mulberry biochars. Although only plants grown with the garden waste biochar treatment survived in Udult soil, amendment of garden waste or mulberry biochars at 3% w/w (450 °C) to Udult soil significantly increased (4.95–6.25) soil pH compared to other biochar treatments. In Udept and Ustalf soils, the application of garden waste and mulberry biochars significantly improved plant biomass compared to control, albeit it was dependent on both biochar and soil properties. Garden waste biochar significantly decreased soil Cd phytoavailable concentration by 26% in the Udult soil, while a decrease of soil Cd phytoavailable concentration by 16% and 9% was observed in Ustalf and Udept soils, respectively. The available phosphorus in biochar and soil pH were important factors controlling toxic metal phytouptake by the plant. Thus, the amendment of soil with biochar at 3% can effectively reduce the mobility of Cd and Pb in soil and plant uptake. However, biochar and soil properties should be well-known before being used for soil toxic metal immobilization.
  • Thumbnail Image
    Item
    Inactivation of salmonella enterica serovar enteritidis on chicken eggshells using blue light
    (MDPI (Basel, Switzerland), 2021-08-10) Hu X; Sun X; Luo S; Wu S; Chu Z; Zhang X; Liu Z; Wu J; Wang X; Liu C; Wang X; Santini A
    Salmonella enterica serovar Enteritidis (S. Enteritidis) is a pathogen that poses a health risk. Blue light (BL), an emerging sanitization technology, was employed for the first time in the present study to inactivate S. Enteritidis on eggshell surfaces and its influence on maintaining eggshell freshness was investigated systematically. The results showed that 415 nm-BL irradiation at a dose of 360 J/cm2 reduced 5.19 log CFU/mL of S. Enteritidis in vitro. The test on eggshells inoculated with S. Enteritidis showed that a BL dose at 54.6 J/cm2 caused a 3.73 log CFU reduction per eggshell surface and the impact of BL inactivation could be sustained in post-5-week storage. The quality of the tested eggs (weight loss, yolk index, Haugh unit (HU) and albumen pH) demonstrated that BL treatments had negligible effects on the albumen pH of eggs. However, compared to the control, BL-treated eggs showed lower weight loss and higher HU after 5 weeks of storage at 25◦C and 65% humidity and yolk index in the control group could not be determined after 5 weeks of storage. Besides, the total amino acid content of the BL-treated egg was higher than the control, exhibiting an advantage of BL irradiation in maintaining the nutrient quality of whole eggs. The current study determined the efficacy of BL against S. Enteritidis on eggshell and suggested that BL could be an effective application in maintaining the freshness and quality of eggs.
  • Thumbnail Image
    Item
    Integrative analysis identifies two molecular and clinical subsets in Luminal B breast cancer
    (Elsevier Inc, 2023-09-15) Wang H; Liu B; Long J; Yu J; Ji X; Li J; Zhu N; Zhuang X; Li L; Chen Y; Liu Z; Wang S; Zhao S
    Comprehensive multiplatform analysis of Luminal B breast cancer (LBBC) specimens identifies two molecularly distinct, clinically relevant subtypes: Cluster A associated with cell cycle and metabolic signaling and Cluster B with predominant epithelial mesenchymal transition (EMT) and immune response pathways. Whole-exome sequencing identified significantly mutated genes including TP53, PIK3CA, ERBB2, and GATA3 with recurrent somatic mutations. Alterations in DNA methylation or transcriptomic regulation in genes (FN1, ESR1, CCND1, and YAP1) result in tumor microenvironment reprogramming. Integrated analysis revealed enriched biological pathways and unexplored druggable targets (cancer-testis antigens, metabolic enzymes, kinases, and transcription regulators). A systematic comparison between mRNA and protein displayed emerging expression patterns of key therapeutic targets (CD274, YAP1, AKT1, and CDH1). A potential ceRNA network was developed with a significantly different prognosis between the two subtypes. This integrated analysis reveals a complex molecular landscape of LBBC and provides the utility of targets and signaling pathways for precision medicine.
  • Thumbnail Image
    Item
    KMT-2021-BLG-1547Lb: Giant microlensing planet detected through a signal deformed due to source binarity
    (EDP Sciences, France, 2023-10) Han C; Zang W; Jung YK; Bond IA; Chung S-J; Albrow MD; Gould A; Hwang K-H; Ryu Y-H; Shin I-G; Shvartzvald Y; Yang H; Yee JC; Cha S-M; Kim D; Kim D-J; Kim S-L; Lee C-U; Lee D-J; Lee Y; Park B-G; Pogge RW; Monard B; Qian Q; Liu Z; Maoz D; Penny MT; Zhu W; Abe F; Barry R; Bennett DP; Bhattacharya A; Fujii H; Fukui A; Hamada R; Hirao Y; Ishitani Silva S; Itow Y; Kirikawa R; Kondo I; Koshimoto N; Matsubara Y; Miyazaki S; Muraki Y; Olmschenk G; Ranc C; Rattenbury NJ; Satoh Y; Sumi T; Suzuki D; Tomoyoshi M; Tristram PJ; Vandorou A; Yama H; Yamashita K
    Aims. We investigate the previous microlensing data collected by the KMTNet survey in search of anomalous events for which no precise interpretations of the anomalies had been suggested. From this investigation, we find that the anomaly in the lensing light curve of the event KMT-2021-BLG-1547 is approximately described by a binary-lens (2L1S) model with a lens possessing a giant planet, but the model leaves unexplained residuals. Methods. We investigated the origin of the residuals by testing more sophisticated models that include either an extra lens component (3L1S model) or an extra source star (2L2S model) on top of the 2L1S configuration of the lens system. From these analyses, we find that the residuals from the 2L1S model originate from the existence of a faint companion to the source. The 2L2S solution substantially reduces the residuals and improves the model fit by δ x 2 = 67.1 with respect to the 2L1S solution. The 3L1S solution also improves the fit, but its fit is worse than that of the 2L2S solution by δ x 2 = 24.7. Results. According to the 2L2S solution, the lens of the event is a planetary system with planet and host masses (Mp/MJ, Mh/M·) = (1.47-0.77+0.64, 0.72-0.38+0.32) lying at a distance DL = 5.07-1.50+0.98 kpc, and the source is a binary composed of a subgiant primary of a late G or an early K spectral type and a main-sequence companion of a K spectral type. The event demonstrates the need for sophisticated modeling of unexplained anomalies if one wants to construct a complete microlensing planet sample.