1.A joint distillation model for the tumor segmentation using breast ultrasound images.
Hongjiang GUO ; Youyou DING ; Hao DANG ; Tongtong LIU ; Xuekun SONG ; Ge ZHANG ; Shuo YAO ; Daisen HOU ; Zongwang LYU
Journal of Biomedical Engineering 2025;42(1):148-155
The accurate segmentation of breast ultrasound images is an important precondition for the lesion determination. The existing segmentation approaches embrace massive parameters, sluggish inference speed, and huge memory consumption. To tackle this problem, we propose T 2KD Attention U-Net (dual-Teacher Knowledge Distillation Attention U-Net), a lightweight semantic segmentation method combined double-path joint distillation in breast ultrasound images. Primarily, we designed two teacher models to learn the fine-grained features from each class of images according to different feature representation and semantic information of benign and malignant breast lesions. Then we leveraged the joint distillation to train a lightweight student model. Finally, we constructed a novel weight balance loss to focus on the semantic feature of small objection, solving the unbalance problem of tumor and background. Specifically, the extensive experiments conducted on Dataset BUSI and Dataset B demonstrated that the T 2KD Attention U-Net outperformed various knowledge distillation counterparts. Concretely, the accuracy, recall, precision, Dice, and mIoU of proposed method were 95.26%, 86.23%, 85.09%, 83.59%and 77.78% on Dataset BUSI, respectively. And these performance indexes were 97.95%, 92.80%, 88.33%, 88.40% and 82.42% on Dataset B, respectively. Compared with other models, the performance of this model was significantly improved. Meanwhile, compared with the teacher model, the number, size, and complexity of student model were significantly reduced (2.2×10 6 vs. 106.1×10 6, 8.4 MB vs. 414 MB, 16.59 GFLOPs vs. 205.98 GFLOPs, respectively). Indeedy, the proposed model guarantees the performances while greatly decreasing the amount of computation, which provides a new method for the deployment of clinical medical scenarios.
Humans
;
Breast Neoplasms/diagnostic imaging*
;
Female
;
Ultrasonography, Mammary/methods*
;
Image Processing, Computer-Assisted/methods*
;
Algorithms
;
Neural Networks, Computer
;
Breast/diagnostic imaging*
2.Effect and mechanism of LncRNA EFRL on homocysteine-induced atherosclerosis in macrophage efferocytosis.
Jiaqi YANG ; Zhenghao ZHANG ; Fang MA ; Tongtong XIA ; Honglin LIU ; Jiantuan XIONG ; Shengchao MA ; Yideng JIANG ; Yinju HAO
Chinese Journal of Cellular and Molecular Immunology 2025;41(7):577-584
Objective To investigate the effect and mechanism of Efferocytosis Relatived LncRNA (EFRL) on homocysteine-induced atherosclerosis in macrophage efferocytosis. Methods RAW264.7 cells were cultured in vitro, and the Control group (0 μmol/L Hcy) and Hcy intervention group (100 μmol/L Hcy) were set up. After GapmeR transfection of macrophages with Hcy intervention, EFRL knockdown negative control group (Hcy combined with LNA-NC) and EFRL knockdown group (Hcy combined with LNA-EFRL) were set up. High-throughput sequencing was applied for different expression of LncRNA MSTRG. 88917.16 (EFRL), UCSC was used to analyze its conservation, CPC and CPAT were used to analyze its ability to encode proteins, and GO and KEGG were used to analyze related biological functions. The localization of LncRNA EFRL in macrophages was analyzed by nucleoplasmic separation and RNA-FISH. Quantitative real-time PCR was used to detect the expression levels of LncRNA EFRL and its target gene SPAST in Hcy-treated macrophages. The apoptosis rate of Jurkat cells induced by UV was detected by flow cytometry. In vitro efferocytosis assay combined with immunofluorescence technique was used to analyze macrophage efferocytosis. ELISA was used to detect the levels of interleukin 1β(IL-1β) and IL-18. Results The new LncRNA MSTRG.88917.16 was identified and named EFRL(Efferocytosis Relatived LncRNA). UCSC, CPC and CPAT analyses showed that LncEFRL is highly conserved and does not have the ability to encode proteins. GO and KEGG analyses suggested that LncEFRL may be involved in macrophage efferocytosis. LncRNA EFRL was localized in the nucleus of macrophages as determined by nucleoplasmic separation and RNA-FISH. In comparison to the Control group, the expression levels of LncRNA EFRL and its target gene SPAST in the Hcy group were increased. In comparison to the Control group (0 min), the apoptosis rate of the experimental group (15, 30 min) Annexin V is more than 85%. Compared with Hcy combined with LNA-NC group, Hcy combined with LNA-EFRL group had enhanced macrophage efferocytosis and reduced levels of inflammatory factors. Compared with Hcy combined with LNA-NC group, the expression level of SPAST in Hcy combined with LNA-EFRL group was decreased. Conclusion Inhibition of EFRL expression can alleviate the process of Hcy inhibiting macrophage efferocytosis, and the mechanism is related to the regulation of the downstream target gene SPAST by EFRL.
RNA, Long Noncoding/physiology*
;
Animals
;
Homocysteine
;
Mice
;
Macrophages/drug effects*
;
Humans
;
RAW 264.7 Cells
;
Atherosclerosis/chemically induced*
;
Apoptosis/genetics*
;
Phagocytosis/genetics*
;
Jurkat Cells
;
Interleukin-1beta/genetics*
;
Efferocytosis
3.Expression of lncRNA GTSE1-AS1 in prostate cancer tissues and its effect on proliferation and invasion of LNCaP cells
LU Shuaiqi ; LI Xiaohui ; HAO Tongtong, ; HAN Xingtao, ; ZHANG Han
Chinese Journal of Cancer Biotherapy 2021;28(1):17-22
[Abstract] Objective: To explore the expression of long non-coding RNA (lncRNA) GTSE1-AS1 in prostate cancer tissues and the mechanism that affects the proliferation and invasion of LNCaP cells. Methods: From November 2017 to December 2018, 68 pairs of prostate cancer tissue and para-cancerous tissue specimens were resected from prostate cancer patients at the Department of Urology of Luoyang Central Hospital Affiliated to Zhengzhou University; in addition, prostate cancer cell lines LNCaP, PC-3, C4-2B, 22Rv1, DU-145 and normal prostate follicular epithelial RWPE-1 cells were also chosen for this study. qPCR was used to detect the expression level of GTSE1-AS1 in cancer tissues and cell lines. The GTSE1-AS1 over-expression plasmid (experimental group) and negative control plasmid (control group) were respectively transfected into LNCap cells. MTT assay and Transwell chamber method were used to detect the effect of GTSE1-AS1 over-expression on the proliferation and invasion ability of LNCaP cells, respectively. The targeting relationship among GTSE1-AS1 and miR-324-3P as well as FBXW7 (F-frame/WD repeat domain protein 7) was verified by bioinformatics tools and dual-luciferin reporter gene assay. The effect of GTSE1-AS1 over-expression on downstream gene and protein expression was detected by qPCR and WB assay. Results: The expression level of GTSE1-AS1 in prostate cancer tissues was significantly lower than that in para-cancerous tissues (P<0.01), and the expression of GTSE1-AS1 in prostate cancer cell lines was significantly lower than that in RWPE-1 cells (P<0.05 or P<0.01). Over-expression of GTSE1-AS1 significantly inhibited the proliferation and invasion (P<0.05 or P<0.01) of LNCaP cells. Dual-luciferin reporter gene assay confirmed the complementary binding between GTSE1-AS1 and miR-324-3p as well as between miR-324-3p and FBXW7. Over-expression of GTSE1-AS1 significantly reduced the expression of miR-324-3p in LNCaP cells (P<0.01), and promoted the mRNA and protein expressions of FBXW7 (all P<0.01). Conclusion: GTSE1-AS1 is under-expressed in prostate cancer tissues and cell lines. Over-expression of GTSE1-AS1 can inhibit the proliferation and invasion of LNCaP cells, the mechanism of which may be related with the inhibition of miR-324-3p to further promote FBXW7 expression.
4. Investigate thenovel coronaviruses in external environmentof COVID-19
Zerun XUE ; Yang LUAN ; Hailong CHEN ; Quanli DU ; Yuewen HAN ; Yalan ZHANG ; Tongtong YU ; Xiaoguang WEI ; Hao LI ; Yujie YANG ; Yangni DENG ; Kaixin LI ; Ruihua TIAN ; Yongfeng WU ; Jie WEI ; Xian MA ; Chaofeng MA ; Rui WU ; Mingjun HU
Chinese Journal of Experimental and Clinical Virology 2020;34(0):E011-E011
Objective To understand the distribution of novel coronaviruses in the external environment of confirmed COVID-19 cases. Methods Environmental surface swab specimens such as bed rails, doorknob, closestool, hand washing sink, table, locker,ward pager, mobile phone, cup, clothes, were collected from the sentinel hospital of COVID-19, and samples were collected for the nucleic acid detection by RT-PCR. Results A total of 150 environmental samples were collected from 30 confirmed COVID-19 cases, 6 samples were determined to be novel coronaviruses postive (positive rate 4.00%). The total 14 mobile phone showed 3 novel coronaviruses positive.Among the 30 confirmed COVID-19 cases, 6 cases (positive rate 20.00%)were found novel coronaviruses in the external environment. Conclusions Novel coronaviruses exists in external environment of confirmed COVID-19 cases, which indicates the potential risk of COVID-19 infection.
5.Nursing study in abdominal breathing training ameliorating sleep status in elderly patients with chronic heart failure
Ying SHEN ; Jingrong HAO ; Wenjuan LIU ; Haiyan BAI ; Peiying ZHANG ; Tongtong HE ; Hui NI
Chinese Journal of Practical Nursing 2017;33(12):904-907
Objective To investigate the effects of abdominal breathing training on sleep disorders in elderly patients with chronic heart failure. Methods Total of 100 patients with chronic heart failure complain of sleeping disorders and Pittsburgh Sleep Quality Index (PSQI)>7 points were assigned into two groups by random digits table method, 50 cases in each group. The observation group and the control group were nursed in the same way except that abdominal breathing was adapted to the observation group. Sleep status, heart rate, blood pressure, SpO2 and brain natriuretic peptide (BNP) were evaluated before training, one week and eight weeks after training respectively. Statistics was used to analyze the differences between two groups. Results After training one week, the sleep status of the observation group was ameliorated, but without significant difference compared to the control group (P>0.05). And after training eight weeks, the PSQI, BNP and heart rate were (9.21 ± 6.38) points, (193.78 ± 152.16) μg/L, (63.5 ± 10.8) times/min in the observation group, and (12.92 ± 0.33) points, (417.55 ± 262.47) μg/L, (70.7 ± 8.5) times/min in the control group, and there was significant differences between 2 groups (t=3.627, 2.041, 2.767, all P < 0.05), while the blood pressure, SpO2 did not change obviously(P>0.05). Conclusions Abdominal breathing training could ameliorate sleep status in elderly patients with chronic heart failure.
6.Clinicopathologic Characteristics of the Patients in the Elderly Lung Carcinoma
CHEN JIE ; HAO XUEZHI ; CHENG FANG ; ZHANG TONGTONG ; XING PUYUAN ; LI JUNLING
Chinese Journal of Lung Cancer 2016;19(10):675-678
Background and objective As the prevalence of tobacco and the aging of the population, the incidence of lung cancer in the elderly rises. However, few elderly patients (older than 70 years old) with lung squamous cell carcinoma were involved into the clinical trials, which offered insuffcient clinical evidence for these patients. Lung squamous cell carcino-ma patients older than 80 years old were included in our study to analyze the clinical characteristics, treatment and prognostic factors, and to explore the optimal treatment choices for these patients.Methods We retrospectively analyzed the clinical fea-tures of 38 elderly patients with lung squamous carcinoma and summarized the treatment under the clear diagnosis and clini-cal staging.Results Elderly patients with squamous cell carcinoma can choose surgery, radiotherapy and chemotherapy based on diagnosis and clinical staging when their physical condition is permitted.Conclusion Because of the short life expectancy of patients more than 80 years old, fewer of them could receive completed and effective treatment, comparing with patients between 70 and 80 years old.

Result Analysis
Print
Save
E-mail