CenterNet++ for Object Detection.

There are two mainstream approaches for object detection: top-down and bottom-up. The state-of-the-art approaches are mainly top-down methods. In this paper, we demonstrate that bottom-up approaches show competitive performance compared with top-down approaches and have higher recall rates. Our approach, named CenterNet, detects each object as a triplet of keypoints (top-left and bottom-right corners and the center keypoint). We first group the corners according to some designed cues and confirm the object locations based on the center keypoints. The corner keypoints allow the approach to detect objects of various scales and shapes and the center keypoint reduces the confusion introduced by a large number of false-positive proposals. Our approach is an anchor-free detector because it does not need to define explicit anchor boxes. We adapt our approach to backbones with different structures, including 'hourglass'-like networks and 'pyramid'-like networks, which detect objects in single-resolution and multi-resolution feature maps, respectively. On the MS-COCO dataset, CenterNet with Res2Net-101 and Swin-Transformer achieve average precisions (APs) of 53.7% and 57.1%, respectively, outperforming all existing bottom-up detectors and achieving state-of-the-art performance. We also design a real-time CenterNet model, which achieves a good trade-off between accuracy and speed, with an AP of 43.6% at 30.5 frames per second (FPS).

Improving the Product Specificity of Maltotetraose-Forming Amylase from Pseudomonas saccharophila STB07 by Removing the Carbohydrate-Binding Module.

Maltotetraose (G4) is composed of four glucose units linked by the α-1,4-glycosidic bond, which has excellent adaptability in food processing and specific physiological functions. Maltotetraose-forming amylases (MFAses) are used in the industry as a promising tool for G4 production. The MFAse from Pseudomonas saccharophila STB07 (MFAPS), which belongs to the GH13, can preferentially hydrolyze substrates to G4. MFAPS contains a carbohydrate-binding module (CBM). In this study, we removed the CBM to obtain the mutant MFAPS-ΔCBM. We explored the aspects affecting the catalytic performance of enzymes through structural simulations and molecular docking. Results showed that when the CBM was removed, the thermal stability of MFAPS was slightly reduced, and its catalytic ability for long-chain substrates, such as corn starch, was significantly reduced. However, the catalytic ability and product specificity of the substrates with shorter chain length, such as maltodextrin (DE 7-9), were improved. The G1-G7 (glucose (G1), maltose (G2), maltotriose (G3), maltotetraose (G4), maltopentaose (G5), maltohexaose (G6), and maltoheptaose (G7)) contents and G4 proportion of the mutant MFAPS-ΔCBM reaction at 24 h were 11.1 and 11.6% higher than those of MFAPS, respectively. The results also showed that the forces of MFAPS on the substrate near the -4, -1, +1, and +3 subsites were critical for its product specificity.

A Real-world Evidence-based Management of HIV by Differential Duration HAART Treatment and its Association with Incidence of Oral Lesions.

BACKGROUND: The efficacy of highly active antiretroviral therapy (HAART) can be estimated by the immunological response and the incidence of opportunistic infections. OBJECTIVE: This study aimed at evaluating the effectiveness of different durations of HAART in terms of immunological response markers (CD4 count and CD4/CD8 ratio) along with disease progression markers (incidence of oral lesions) in Chinese patients with HIV. METHODS: This single-center, retrospective, and real-world study included patients with HIV, grouped into a treatment group and treatment-naïve group, of which the former was further divided into 6, 12, and 18 months based on the treatment duration. The CD4 and CD8 cell counts were analyzed by the FACSCalibur flow cytometry. Kruskal-Wallis test was applied to determine the outcome of different duration of HAART. Oral examination was carried out according to the WHO type IV examination. RESULTS: In 246 patients with HIV, CD4 counts increased significantly post-HAART compared to pre-HAART in all three treatment groups (P<.001), while CD8 count decreased significantly (P<.05) in all three treated groups. A significant association of HAART with the CD4/CD8 ratio was observed (P<.001). A significant increase in CD4 count was observed between 12-months and 18-months treatment groups (P<.05). The occurrence of oral lesions reduced significantly in the treatment group. CONCLUSION: We observed a better response to the HAART regimen with 18-months of duration than 12-months and 6-months therapies and reduction in oral lesions.

Gingival plasma cell granuloma: a case report.

Plasma cell granuloma is extremely rare in the oral cavity. This paper reports a case of plasma cell granuloma in the oral cavity and reviews the literature. In this case, the plasma cell granuloma occurred in the gingiva, which was bright red and granular. The lesion was resected and diagnosed as gingival plasma cell granuloma by clinical pathological examination. Four months follow-up after operation surgery showed no signs of recurrence. Although gingival plasma cell granuloma is rare, it is easily confused with related diseases and should be identified and followed up closely.

A real-world, cross sectional study of oral lesions and their association with CD4 cell counts and HIV viral load in Yunnan, China.

Human immunodeficiency virus (HIV) supresses immune system, primarily cell-mediated immunity. Cluster of differentiation 4 (CD4) cell count, viral load, and oral lesions are the most important laboratory parameters to evaluate the evolution of acquired immunodeficiency syndrome. The present study aims to determine the incidence of HIV-related oral lesions with CD4 cell count and viral load in Yunnan, China.A cross-sectional study was conducted from December 2007 to December 2009, in 1812 HIV positive patients from Department of Infectious Diseases in Kunming Third People's Hospital. CD4, CD8, and viral load data were collected and analyzed statistically using SPSS 11.3.Out of 1812 HIV positive patients, 929 (51.27%) were associated with 1 or more oral lesions. The most common oral lesions observed were Candida Pseudomembranous (13.75%), Candida erythematous (10.93%), Oral hairy leukoplakia (7.95%), Aphthous ulcer (6.18%), Herpes simplex infection (5.58%). In most patients with oral lesions, the CD4 cell count was < 200/µL. The incidence of oral lesions was lower when CD4 count was > 200/µL and with undetectable (P < .01) HIV viral load. Almost no oral lesions was observed when CD4 count > 500/µL (P < .01). With highly active antiretroviral therapy, reduction in HIV-related oral lesions was observed especially in Candida erythematous, Candida Pseudomembranous, Oral hairy leukoplakia, and Aphthous ulcer.The higher incidence of oral lesions with lower CD4 count (<200/µL) in HIV-infected patients indicated importance of CD4 cell count in identifying disease progression.

Current Advances and Limitations of Deep Learning in Anticancer Drug Sensitivity Prediction.

Anticancer drug screening can accelerate drug discovery to save the lives of cancer patients, but cancer heterogeneity makes this screening challenging. The prediction of anticancer drug sensitivity is useful for anticancer drug development and the identification of biomarkers of drug sensitivity. Deep learning, as a branch of machine learning, is an important aspect of in silico research. Its outstanding computational performance means that it has been used for many biomedical purposes, such as medical image interpretation, biological sequence analysis, and drug discovery. Several studies have predicted anticancer drug sensitivity based on deep learning algorithms. The field of deep learning has made progress regarding model performance and multi-omics data integration. However, deep learning is limited by the number of studies performed and data sources available, so it is not perfect as a pre-clinical approach for use in the anticancer drug screening process. Improving the performance of deep learning models is a pressing issue for researchers. In this review, we introduce the research of anticancer drug sensitivity prediction and the use of deep learning in this research area. To provide a reference for future research, we also review some common data sources and machine learning methods. Lastly, we discuss the advantages and disadvantages of deep learning, as well as the limitations and future perspectives regarding this approach.

[The incidence of oral candidiasis in patients with human immunodeficiency virus infection/acquired immunodeficiency syndrome from Yunnan, China].

OBJECTIVE: To assess the incidence of oral candidiasis and its influencing factors in patients with human immunodeficiency virus infection/acquired immunodeficiency syndrome (HIV/AIDS). METHODS: An oral examination was conducted in the 1 566 HIV/AIDS patients in the Third Hospital of Kunming from March 2008 to September 2012 (M/F: 1 062/504, age range: 0.2 to 84.0 years old). The HIV viral load (HIV- RNA) and peripheral blood CD4 count were respectively analyzed by Bayer Q340 fluorescence signal surveying instrument (bDNA method) and flow cytometry analysis. The information on usage of highly active anti-retroviral (HAART) drugs and transmission of HIV were obtained through questionnaires. The incidence of oral candidiasis in patients with different HIV-RNA levels and CD4 count and the use of HAART was analyzed and compared. RESULTS: The total incidence of oral candidosis was 31.0% (486/1 566) and there was no difference in sex. The oral lesions were presented by three types, psudomembranous candidosis (PC), erythematous candidosis (EC) and angular cheilitis (AC), and the morbidity was 13.9% (217/1 566), 17.0% (267/1 566) and 4.9% (77/1 566), respectively. The average level of CD4 count in psudomembranous candidosis, erythematous candidosis and angular cheilitis [81.0 (146.0), 74.0 (152.0) and 69.0 (121.5) cell/µl] showed no significant difference (P > 0.05). The incidence of oral candidiasis in non-HAART and HAART subjects were 36.3% (402/1 107) and 18.3% (84/459), respectively (P = 0.000). The CD4 count and absolute counts of HIV viral load in oral candidiasis patients and non-oral candidiasis patients had significant difference (Z = -10.261, P = 0.000 and Z = -4.762, P = 0.000). CONCLUSIONS: The morbidity of oral candidiasis in HIV/AIDS patients in Yunnan Province was high, including PC, EC and AC and hyperplastic candidosis was not detected. The incidence was related to the degree of immune suppression and HIV viral load.

Correlation of transcription of MALAT-1, a novel noncoding RNA, with deregulated expression of tumor suppressor p53 in small DNA tumor virus models.

Although metastasis-associated lung adenocarcinoma transcript (MALAT)-1 is known to be consistently upregulated in several epithelial malignancies, little is known about its function or regulation. We therefore examined the relationship between MALAT-1 expression and candidate modulators such as DNA tumor virus oncoproteins human papillomavirus (HPV)-16 E6 and E7, BK virus T antigen (BKVTAg), mouse polyoma virus middle T antigen (MPVmTAg) and tumor suppressor genes p53 and pRb. Using suppressive subtractive hybridization (SSH) and real-time reverse transcriptase polymerase chain reaction (RT-PCR) assays, MALAT-1 was shown to be increased in viral oncongene-expressing salivary gland biopsies from humans and mice. The results also indicated that MALAT-1 transcripts and promoter activity were increased in vitro when viral oncongene-expressing plasmids were introduced into different cell types. These same viral oncogenes in addition to increasing MALAT-1 transcription have also been shown to inhibit p53 and/or pRb function. In p53 mutant or inactive cell lines MALAT-1 was also shown to be highly upregulated. We hypothesize that there is a correlation between MALAT-1 over-expression and p53 deregulation. In conclusion, we show that disruption of p53, by both polyoma and papilloma oncoproteins appear to play an important role in the up-regulation of MALAT-1. MALAT-1 might therefore represent a biomarker for p53 deregulation within malignancies.

[Incidence of human herpes virus 1-4 type in saliva of 245 human immunodeficiency virus-infected patients].

OBJECTIVE: To determine the incidence of human herpes virus (HHV) 1-4 type including herpes simplex virus type-1 (HSV-1), herpes simplex virus type-2 (HSV-2), varicella-zoster virus (VZV), Epstein-Barr virus(EBV) in the saliva of human immunodeficiency virus (HIV) -infected patients. METHODS: The incidence of salivary HSV-1, HSV-2, VZV and EBV from 245 HIV-seropositive individuals and control group was used to investigate by polymerase chain reaction(PCR) or nested PCR. The data was analyzed by SPSS 18.0 statistical software. RESULTS: In the 245 HIV-seropositive individuals, the detection rates of HSV-1, HSV-2, VZV, EBV were 29.0%, 3.3%, 4.1%, 82.0%. In the control group, the detection rates of HSV-1, HSV-2, VZV, EBV were 13.3%, 0, 0, 36.7%. Four HHVs were significantly more prevalent in the salivas of HIV-seropositive persons than those in the control group (P < 0.01). The detection rates of HSV-1, HSV-2, VZV and EBV DNA were no difference between the HIV-positive group with highly active antiretroviral therapy (HAART) and HIV-positive group without HAART (P > 0.05). CONCLUSION: There is a high prevalence of HHV infection in HIV-infected people in Yunnan. The most common virus are EBV, followed by HSV-1, but VZV and HSV-2 are rarely detected. HHV co-infection is also observed.

[Incidence of oral hairy leukoplakia in human immunodeficiency virus-seropositive adult patients in Yunnan, China].

OBJECTIVE: To study the incidence rates, clinical characteristics of oral hairy leukoplakia (OHL) and its relation to the immune status in a sample of human immunodeficiency virus (HIV)-infected adults in Yunan, China. METHODS: 1 060 adult patients with HIV from January 2008 to June 2010 were evaluated. The age, gender, education grade, diagnosis time of HIV-infected, route of transmission, xerostomia, oral candidiasis, high active antiretroviral therapy and CD4 lymphocytes counts. The occurrence of OHL was recorded by oral examination. The relationship of CD4 lymphocytes counts and the incidence of OHL were analyzed by statistical methods. RESULTS: There were 94 OHL patients in 1 060 HIV patients (8.9%). The average age of the OHL patients was (39.33 +/- 10.45) years old. 90% OHL was found on the two lateral aspect of the tongue. The CD4 lymphocytes of 70.2% OHL patients were less than 200 mm-3. CONCLUSION: OHL is a frequent finding in patients with indicates severe immunosuppression and associated with the reduction of CD4 lymphocytes.

The intervention research on understanding of the AIDS prevention and occupational safety of the dentist in Kunming and west part of Yunnan.

OBJECTIVE: To investigate the knowledge, attitude of the AIDS prevention and occupational safety in dentist in Kunming and west part of Yunnan and evaluate the effect of education and intervention. METHOD: 165 dentists in Yunnan were tested by KABP questionnaire before education. All of the 165 dentists participated a course on AIDS prevention. After that course the same questionnaire was answered by participants. RESULT: The correct rates obviously increased. The understanding rates of the oral prevention measure were obviously improved. CONCLUSION: The knowledge of AIDS prevention is very low in dentist in Kunming and west part of Yunnan. The education intervention can effectively improve the understanding level of the AIDS prevention in dentist.