[Research progress on the application of virtual artificial intelligence in risk assessment and diagnosis of periodontal disease].

Periodontal disease is a common and frequently-occurring disease in China. Early detection, diagnosis, and treatment of periodontal disease are of great significance for achieving universal oral health and even systemic health. Artificial intelligence endows machines with the ability to mimic human intelligent behavior, and is commonly used in medical field with both physical and virtual forms. Virtual artificial intelligence empowers traditional experience in the application of periodontal disease risk assessment and diagnosis, with the potential to develop a variety of oral health screening tools. It helps to provide new evidence for the prognosis of periodontal disease, improve the accuracy and efficiency of diagnosis, reduce technical sensitivity and further promote the periodontal treatment transformation from "treatment-oriented" to "prevention-oriented". This paper reviews the current applications and progresses of virtual artificial intelligence in periodontal risk assessment and diagnosis, as well as its limitations, providing ideas for future researches on the application of virtual artificial intelligence in this field.

[Epidemiological characteristics of early-onset colorectal cancer: a prospective cohort study from a single center].

Objective: To explore the differences in distribution of colorectal cancer-related risk factors between patients with early-onset colorectal cancer (EOCRC) and those with late-onset colorectal cancer (LOCRC) in a Chinese cohort, and to provide reference and guidance for the prevention, diagnosis, and treatment of EOCRC. Methods: Using data from the National Colorectal Cancer Cohort study cohort, 5377 patients with newly diagnosed colorectal cancer (CRC) attending the Department of Colorectal Surgery and Oncology of the Second Affiliated Hospital, Zhejiang University School of Medicine from June 2018 to February 2023 were included in the study cohort. Questionnaires capturing epidemiological features, including lifestyle and dietary habits, were administered. The patients were divided into two groups, the cut-off age being 50 years. Those aged ≥50 years were defined as having LOCRC and those aged <50 years as having EOCRC. Wilcoxon (continuous variates) or χ2 tests (categorical variates) were performed to compare differences in epidemiological features. Results: A total of 3799 people who had completed the questionnaire were included in this study, 491 of whom had EOCRC and 3308 LOCRC. The response rate to the questionnaire was 70.7%. The median ages of patients in the EOCRC and LOCRC groups were 43 and 66 years, respectively. There was a higher proportion of female patients (48.5% [253/491] vs. 35.8% [1184/3308], χ2=28.8, P<0.001) in the EOCRC than the LOCRC group. Patients with EOCRC and lower body mass index (medium 22.1 kg/m2 vs. 22.9 kg/m2, W=744 793, P=0.005) and lower proportion of abdominal obesity (87.2% [428/491] vs. 93.8% [3103/3308], χ2=38.3, P<0.001). Patients with EORC significantly less commonly reported a history of hypertension (5.9% [29/491] vs. 41.6% [1375/3308], χ2=231.8, P<0.001), diabetes (1.4% [7/491] vs. 14.4% [476/3308], χ2=63.6, P<0.001) and cardiovascular and cerebrovascular diseases (0.8% [4/491] vs. 7.3% [241/3308], χ2=28.6, P<0.001). However, the proportion of patients with a family history of CRC was significantly higher (P<0.05) in the EOCRC group (10.2% [50/491] vs. 6.9% [227/3 308], χ2=6.5, P=0.010]. In terms of lifestyle, patients with EOCRC had shorter sleep duration (median: 8.0 hours vs. 8.5 hours, W=578 989, P<0.001), and were less likely to participate in physical exercise (29.5% [145/491] vs. 38.7% [1281/3308] χ2=15.0, P<0.001) or engage in physical work (65.2% [320/491] vs. 74.1% [2450/3308], χ2=16.7, P<0.001). Meanwhile, in the EOCRC group a lower percentage of patients were smokers (29.3% [144/491] vs. 42.7% [1411/3308], χ2=46.9,P<0.001) and they smoked less (median 17.6 pack/year vs. 30.0 pack/year,W=55 850,P<0.001). Fewer patients in the EOCRC group habitually drank alcohol (21.0% [103/491] vs. 38.0% [1257/3308], χ2=57.5, P<0.001) or tea (17.5% [86/491] vs. 28.7% [948/3308], χ2=26.2, P<0.001) than in the LOCRC group. Compared with the LOCRC group, patients with EOCRC had a higher frequency of intake of fresh meat, fresh fruit, eggs, and dairy products and a lower frequency of intake of preserved meat and pickled vegetables; these differences are statistically significant (all P<0.05). There was no statistically significant difference in consumption of fresh vegetables or a high-sugar diet between the two groups (both P>0.05). Conclusions: This study highlights disparities in adverse lifestyle and dietary habits between patients in China with EOCRC versus LOCRC.

[Research progress on the impact of diet on the early-onset colorectal cancer].

The incidence of early-onset colorectal cancer has been gradually increasing in recent years. Studies have shown that early-onset CRC is closely related to modifiable risk factors such as diet, but there is still a lack of consistent conclusions and a systematic review of relevant research results. In this review, we comprehensively summarized the association between diet and the early-onset CRC, clarified the association and relative risk between different dietary patterns, common food types and nutrients and the occurrence of early-onset CRC, and elaborated the underlying physiological mechanisms. Enhancing the understanding of dietary risk factors, which are modifiable exogenous risk factors, is expected to serve as a reference for the formulation of primary prevention strategies for early-onset CRC.

[Updates of colonoscopy surveillance guidelines after screening and polypectomy, and related research progress].

With the widespread application of colorectal cancer screening, the surveillance and management of the increasing number of screened population has become a pivotal aspect in preventing and controlling colorectal cancer. In recent years, researches have been conducted on the risk of colorectal cancer incidence and mortality in the population after screening. At the same time, various organizations in Europe and the United States have continuously updated colonoscopy surveillance after screening and polypectomy based on the latest research evidence. In this review, we summarized the current progress of studies on colorectal cancer risk in post-screening colorectal cancer populations and the key points of relevant guideline updates, in order to provide a reference for conducting relevant studies and formulating surveillance guidelines or consensus in China.

In vitro evaluation of different protective techniques to reduce titanium particle contamination during implantoplasty.

OBJECTIVES: Our aim is to study titanium remains in a bone model during standardized implantoplasty under different isolation and protective modalities. MATERIAL AND METHODS: Forty implants were placed in artificial spongy bone blocks mimicking a horizontal bone loss and implant neck protrusion of 5 mm. Samples were randomly divided into four groups (n = 10), which were treated as follows: rubber dam (A), a dental adhesive paste (B), bone wax (C), and an unprotected positive control (D). Implantoplasty was performed using carbide and diamond burs under strict water cooling and standardized suction. After removal of the respective isolation materials, the bone blocks were thoroughly rinsed with tap water for 3 min and titanium chips were collected using a filter integrated in the model. The filter paper was removed and dissolved in 37% hydrochloric acid for 2 h at 120 °C and the titanium remnants were quantified using atomic absorption spectrometry. RESULTS: None of the test groups were able to completely prevent titanium particle contamination. Rubber dam (691 ± 249 µg) and bone wax (516 ± 157 µg) were found to be significantly more protective than the positive control (2313 ± 747 µg) (p < 0.001) with respect to the amount of titanium particles that remained in the bone model after implantoplasty. The adhesive paste group (1863.5 ± 538 µg) was not significantly different from the positive control (p = 0.19). CONCLUSIONS: Despite some limitations of the present study, titanium particles resulting from a standardized implantoplasty can be assumed to be significantly reduced when the tissues/bone were protected with rubber dam and bone wax, or a combination, depending on individual accessibility. CLINICAL RELEVANCE: Tissue protective measures to reduce or avoid particle contamination during implantoplasty is possible and should be considered and further clinically assessed to avoid iatrogenic inflammatory reactions.

Lactobacillus coryniformis subsp. torquens inhibits bone loss in obese mice via modification of the gut microbiota.

High-fat diet (HFD)-induced obesity results in bone loss associated with an imbalanced gut microbiota and altered immune status. Probiotics are live microorganisms that are beneficial to the host and are important in maintaining bone health and gut homeostasis. In this study, the probiotic Lactobacillus coryniformis subsp. torquens (T3L) was isolated from traditional yak milk cheese produced in Lhasa and showed distinct acid and bile salt resistance as potential probiotics. Our data indicated that T3L not only reversed HFD-induced gut dysbiosis, as indicated by decreased Firmicutes-to-Bacteroidetes ratios but also reduced bone loss. The anti-obesity, microbiome-modulating, and bone-protective effects were transmissible via horizontal faeces transfer from T3L-treated mice to HFD-fed mice. The protective effects of T3L on bone mass were associated with regulatory T (Treg) cell-mediated inhibition of osteoclast differentiation. Our data indicate that T3L is a regulator of the gut microbiota and bone homeostasis in an animal model.

[Principle and routine operation of laser assisted periodontal surgery].

Periodontitis is one of the most common infectious oral diseases, which can cause destruction of periodontal supporting tissues and even tooth mobility and loss. Controlling infection, eliminating inflammation, restoring the physiological shape of periodontal tissues, and meeting functional and aesthetic needs are the main goals of periodontal treatment. When periodontitis develops to a more severe stage, surgical treatment is necessary to handle soft and hard tissues for good treatment results. Since the development of the first Nd:YAG laser dedicated to dental medicine by Myers in 1990, over 30 years of clinical and basic research have shown that lasers have tremendous potential in assisting periodontal surgery. This article summarizes the principles and operational routines of laser-assisted periodontal surgery, aiming to provide clinical reference for diagnosis and treatment.

Bone defect development in experimental canine peri-implantitis models: a systematic review.

PURPOSE: To provide a systematic overview of preclinical research regarding bone defect formation around different implant surfaces after ligature-induced peri-implantitis models in dogs. Two focused questions were formulated: 'How much bone loss can be expected after a certain time of ligature induced peri-implantitis?' and 'Do different implant types, dog breeds and study protocols differ in their extent of bone loss?' MATERIALS AND METHODS: A systematic literature search was conducted on four databases (MEDLINE, Web of Science, EMBASE and Scopus). Observations, which consisted of bone defects measured directly after ligature removal in canine models, were included and analysed. Two approaches were used to analyse the relatively heterogeneous studies that fulfilled the inclusion criteria. First, separate simple linear regressions were calculated for each study and implant surface, for which observations were available across multiple time points. Second, a linear mixed model was specified for the observations at 12 weeks after ligature initiation, and assessing the potential influencing factors on defect depth was explored using lasso regularisation. RESULTS: Thirty-six studies with a total of 1082 implants were included after. Bone loss was determined at different time points, either with clinical measurements radiographically or histologically. Different implant groups [e.g. turned, sand-blasted-acid-etched (SLA), titanium-plasma-sprayed (TPS) and other rough surfaces] were assessed and described in the studies. A mean incremental defect depth increase of 0.08 mm (SD: -0.01-0.28 mm) per week was observed. After 12 weeks, the defect depths ranged between 0.7 and 5 mm. Based on the current data set, implant surface could not be statistically identified as an essential factor in defect depth after 12 weeks of ligature-induced peri-implantitis. CONCLUSION: Expectable defect depth after a specific time of ligature-induced peri-implantitis can vary robustly. It is currently impossible to delineate apparent differences in bone loss around different implant surfaces.

[Interpretation of updated guidelines for colorectal cancer screening in average-risk individuals in the United States].

In recent years, due to changes in the epidemiology of colorectal cancer in the United States, the update of evidence-based medical evidence for screening, and the emergence of various new screening methods, various organizations in the United States, such as American College of Gastroenterology、Preventive Services Task Force, have updated guidelines for colorectal cancer screening in average-risk individuals. These guidelines have different recommendation levels in terms of starting and ending age, screening methods and frequency for colorectal cancer screening. A comprehensive understanding of the key points of these guideline updates and the similarities and differences recommended by different guidelines has important reference value for the colorectal cancer screening in China.

[Establishment of a platelet production model by bone marrow cavity transplantation of mouse primary megakaryocytes].

Objective: To establish an intramedullary transplantation model of primary megakaryocytes to evaluate the platelet-producing capacity of megakaryocytes and explore the underlying regulatory mechanisms. Methods: Donor megakaryocytes from GFP-transgenic mice bone marrow were enriched by magnetic beads. The platelet-producing model was established by intramedullary injection to recipient mice that underwent half-lethal dose irradiation 1 week in advance. Donor-derived megakaryocytes and platelets were detected by immunofluorescence staining and flow cytometry. Results: The proportion of megakaryocytes in the enriched sample for transplantation was 40 to 50 times higher than that in conventional bone marrow. After intramedullary transplantation, donor-derived megakaryocytes successfully implanted in the medullary cavity of the recipient and produce platelets, which showed similar expression of surface markers and morphology to recipient-derived platelets. Conclusion: We successfully established an in vivo platelet-producing model of primary megakaryocytes using magnetic-bead enrichment and intramedullary injection, which objectively reflects the platelet-producing capacity of megakaryocytes in the bone marrow.

[DALYs for breast cancer in China, 2000-2050: trend analysis and prediction based on GBD 2019].

Objective: Based on the data of Global Burden of Disease 2019 data, to analyze the past, current, and future burden of disability-adjusted life years (DALYs) in China and compare with the international status. Methods: The total number of DALYs, age-standardized DALY rate, and the composition of different subgroups were extracted and described to analyze the time trend in 2000-2019 and the current situation in 2019 for Chinese female breast cancer. The burden of DALYs in 2050 was predicted by Joinpoint using average annual percent change (AAPC). Results: In 2000-2019, the ranking of DALYs caused by female breast cancer in China rose from the fourth to the second in all female cancers. The total DALYs increased by 48.4%, of which the years lived with disability increased from 4.8% to 8.8%. The age-standardized DALY rate only slightly decreased (AAPC=-0.3%; which increased during 2016-2019, AAPC=1.6%). In 2019, the age-standardized DALY rate for breast cancer in China was 278.0/100 000. The DALYs were 2.88 million (accounting for 14.2% of the global burden and 12.1% of all female cancers burden in China), 26.5% of which attributed known risk factors (overweight and obesity were the largest: 0.34 million DALYs, but some common breast cancer risk factors were not available on the platform, such as menstruation and fertility). In 2050, the prediction suggests that the total DALYs caused by female breast cancer in China will reach 3.80 million person-years-5.16 million person-years, increasing 32.1%-79.4% over 2019. From 2000 to 2019, the peak age of DALYs and DALY rate became older, and the DALYs among females aged 65 years and above increased faster than those younger than 65 years (AAPC were 4.8% and 1.3%, respectively). In 2019, females aged 45-74 (the starting age recommended by local guidelines for breast cancer screening) contributed 74.3% of the total DALYs. Conclusions: Over the past 20 years, the age-standardized DALY rate for breast cancer in female populations in China has not changed obviously. Without the continuous expansion of effective intervention and population aging, the burden of DALYs for female breast cancer in China will increase. DALYs for breast cancer attributed leading risk factors were still limited.

Investigation of the association between gall bladder wall thickness and hypoalbuminaemia in dogs.

OBJECTIVES: To assess the correlation between gallbladder wall thickness and serum/plasma albumin concentrations in dogs. MATERIALS AND METHODS: Retrospective searches of digital medical record databases were conducted to identify dogs that had either severely low serum/plasma albumin concentration (<1.5 g/dL) or ultrasonographic evidence of gallbladder wall thickening (>2 mm). Analysis of covariance models were used to analyze gallbladder wall thickness with sample type (serum vs plasma), age, etiology, albumin, and albumin ' etiology as the covariates. RESULTS: A total of 216 dogs met inclusion criteria. One-hundred and forty-six dogs had a thickened gallbladder wall (Group 1). Median serum/plasma albumin concentration for dogs in this group was 2.2 g/dL (1 to 5 g/dL), and 84 dogs (57.5%) had hypoalbuminemia (<2.5 g/dL). The search for dogs with severe hypoalbuminemia (< 1.5 g/dL) identified 70 dogs (Group 2). In this group, median gallbladder wall thickness was 1.3 mm (0.2 to 6.1 mm) and 17 dogs (24.3%) had a thickened gallbladder wall. Serum/plasma albumin concentration and gallbladder wall thickness were not significantly correlated for Group 1 (r = 0.0044, p = 0.9580) or Group 2 (r = -0.1137, p = 0.3487). A moderate negative correlation (-0.64) was identified between gallbladder wall thickness and albumin concentration in dogs with immune-mediated diseases (p = 0.03). CLINICAL SIGNIFICANCE: Gallbladder wall thickness and serum/plasma albumin concentration are independent variables in dogs.

[Intelligent prediction of HER2 status based on breast histopathology].

Objective: To study the association between histopathological features and HER2 overexpression/amplification in breast cancers using deep learning algorithms. Methods: A total of 345 HE-stained slides of breast cancer from 2012 to 2018 were collected at the China-Japan Friendship Hospital, Beijing, China. All samples had accurate diagnosis results of HER2 which were classified into one of the 4 HER2 expression levels (0, 1+, 2+, 3+). After digitalization, 204 slides were used for weakly supervised model training, and 141 used for model testing. In the training process, the regions of interest were extracted through cancer detected model and then input to the weakly supervised classification model to tune the model parameters. In the testing phase, we compared performance of the single- and double-threshold strategies to assess the role of the double-threshold strategy in clinical practice. Results: Under the single-threshold strategy, the deep learning model had a sensitivity of 81.6% and a specificity of 42.1%, with the AUC of 0.67 [95% confidence intervals (0.560,0.778)]. Using the double-threshold strategy, the model achieved a sensitivity of 96.3% and a specificity of 89.5%. Conclusions: Using HE-stained histopathological slides alone, the deep learning technology could predict the HER2 status using breast cancer slides, with a satisfactory accuracy. Based on the double-threshold strategy, a large number of samples could be screened with high sensitivity and specificity.

Towards effective machine learning in medical imaging analysis: A novel approach and expert evaluation of high-grade glioma 'ground truth' simulation on MRI.

PURPOSE/OBJECTIVE(S): Gliomas are uniformly fatal brain tumours with significant neurological and quality of life detriment to patients. Improvement in outcomes has remained largely unchanged in nearly 20 years. MRI (magnetic resonance imaging) is often used in diagnosis and management. Machine learning analyses of large-scale MRI data are pivotal in advancing the diagnosis, management and improve outcomes in neuro-oncology. A common challenge to robust machine learning approaches is the lack of large 'ground truth' datasets in supervised learning for building classification and prediction models. The creation of these datasets relies on human-expert input and is time-consuming and subjective error-prone, limiting effective machine learning applications. Simulation of mechanistic aspects such as geometry, location and physical properties of brain tumours can generate large-scale ground-truth datasets allowing for comparison of analysis techniques in clinical applications. We aimed to develop a transparent and convenient method for building 'ground truth' presentations of simulated glioma lesions on anatomical MRI. MATERIALS/METHODS: The simulation workflow was created using the Feature Manipulation Engine (FME®), a data integration platform specializing in the spatial data processing. By compiling and integrating FME's functions to read, integrate, transform, validate, save, and display MRI data, and experimenting with ways to manipulate the parameters concerning location, size, shape, and signal intensity with the presentations of glioma, we were able to generate simulated appearances of high-grade gliomas on gadolinium-based high-resolution 3D T1-weighted MRI (1 mm3). Data of patients with canonical high-grade tumours were used as real-world tumours for validating the accuracy of the simulation. Twenty raters who are experienced with brain tumour interpretation on MRI independently completed a survey, designed to distinguish simulated and real-world brain tumours. Sensitivity and specificity were calculated for assessing the performance of the approach with the binary classification of simulated vs real-world tumours. Correlation and regression were used in run time analysis, assessing the software toolset's efficiency in producing different numbers of simulated lesions. Differences in the group means were examined using the non-parametric Kruskal-Wallis test. RESULTS: The simulation method was developed as an interpretable and useful workflow for the easy creation of tumour simulations and incorporation into 3D MRI. A linear increase in the running time and memory usage was observed with an increasing number of generated lesions. The respondents' accuracy rate ranged between 33.3 and 83.3 %. The sensitivity and specificity were low for a human expert to differentiate simulated lesions from real gliomas (0.43 and 0.58) or vice versa (0.65 and 0.62). The mean scores ranking the real-world gliomas did not differ between the simulated and real tumours. CONCLUSION: The reliable and user-friendly software method can allow for robust simulation of high-grade glioma on MRI. Ongoing research efforts include optimizing the workflow for generating glioma datasets as well as adapting it to simulating additional MRI brain changes.

[Disease burden of colorectal cancer in China: any changes in recent years?]

Objective: To update the disease burden of colorectal cancer (CRC) in Chinese population by integrating the latest multi-source evidences. Methods: Groups of data from GLOBOCAN, series of Chinese Cancer Registry Annual Report (annual report), Cancer Incidence in Five Continents (CI5), Global Burden of Disease Project 2017 (GBD), China Death Cause Surveillance Datasets and China Health Statistical Yearbooks (yearbook) were used to extract the information. Data on incidence, mortality, disability-adjusted life year (DALY) and percentage distribution of sub-location of CRC were used to analyze the latest disease burden in China, and age-standardized rates by world standard population were mainly used. Joinpoint Trend Analysis Software 4.7.0.0 was applied for time trend analysis. Data related to the economic burden of CRC in China were gathered by literature review. Results: (1) Current status: according to the latest annual report, the incidence and mortality rates of CRC were 17.1 per 100 000 and 7.9 per 100 000, respectively among the covered registration sites in 2015. The incidence ratios of male to female and that of urban to rural were 1.5 and 1.4, with the mortality ratios were 1.6 and 1.4, respectively. Similar to data from the annual report, the mortality rate was reported as 6.9 per 100 000 in 2017 by the surveillance data sets. Data from the GBD project showed that, the DALYs caused by CRC in China in 2017 was 4.254 million person years (doubled compared with that of 1990), accounting for 22.4% of the global burden of CRC. (2) Time trends: according to the annual reports, from 2009 to 2015, the incidence rate and mortality rate of CRC in China decreased by 10.2% and 9.5%, respectively. The same trend was also observed in urban sites, but was opposite in rural areas (increased 20.0% in incidence and 15.2% in mortality). Results from the Joinpoint analysis showed that the averaged annual percentage change (AAPC) was estimated as -1.6% (P<0.05) in the national mortality rate. Similarly, in the incidence and mortality rates of urban sites appeared as AAPC=-1.5% and -1.4% (all P<0.05), but inversely in the incidence rate from the rural sites as AAPC=3.3% (P<0.05). The yearbook data showed a 9.8% increase in urban and 20.6% increase in rural on the mortality in 2017 when compared with 2004, but the Joinpoint analysis showed no statistical significance (P<0.05). (3) Distribution of sub-location of CRC: the annual report showed that among all the new CRC cases in China in 2015, colon, rectal and anal cancer accounted for 49.6%, 49.2% and 1.2%, respectively, while the proportions were 51.3%, 47.6% and 1.1%, respectively in 2009. The proportion of colon cancer was continuously higher in the urban (>52%) than that in the rural areas (<44%). The CI5 Ⅺ data showed that ascending and sigmoid colons were more commonly seen among all the colon cancers. (4) Economic burden: the average annual growth rate of the medical expenditure per CRC patient in China ranged from 6.9% to 9.2%, and the 1-year out-of-pocket expenditure of a newly diagnosed patient accounted for about 60% of their previous-year household income. Conclusions: In China, the overall disease burden of CRC might have been decreased slightly but generally remained stable in the last several years, however, the rising burden appeared in the rural areas should not be ignored. In consistent with findings from a previous review, men and people from the urban areas are considered the target populations for CRC. The finding of higher proportion of colon cancer in urban areas suggests the impact of development of socioeconomic and medical technologies on CRC development and detection. The economic burden of CRC continued to grow.

[Rates on the acceptance of colonoscopy, fecal immunochemical test and a novel risk-adapted screening approach in the screening programs of colorectal cancer as well as related associated factors].

Objective: To compare the rates of acceptance of colonoscopy, fecal immunochemical test (FIT), or a novel risk-adapted screening approach in the colorectal cancer (CRC) screening program. Related risk factors were also studied. Methods: The study has been based on an ongoing randomized controlled trial on colorectal cancer screening programs in six centers of research since May 2018. The involved participants were those who presented at the baseline screening phase. All the participants were randomly allocated into one of the following three intervention arms in a 1∶2∶2 ratio: colonoscopy group, FIT group, and a novel risk-adapted screening group. All the participants underwent risk assessment on CRC by an established risk score system. The subjects with high-risk were recommended to undertake the colonoscopy while the low-risk ones were receiving the FIT. Detailed epidemiological data was collected through questionnaires and clinical examinations. Rates of participation and compliance in all three groups were calculated. Multivariate logistic regression models were used to explore the potential associated factors related to the acceptance of screening. Results: There were 19 546 eligible participants involved in the study, including 3 916 in the colonoscopy group, 7 854 in the FIT group, and 7 776 in the novel risk-adapted screening group, respectively. Among the 19 546 participants, the mean age was 60.5 years (SD=6.5), and 8 154 (41.7%) were males. The rates of participation in the colonoscopy, FIT and the novel risk-adapted screening groups were 42.5%, 94.0% and 85.2%, respectively. In the novel risk-adapted screening group, the participation rate was 49.2% for the high-risk participants who need to undertake colonoscopy and was 94.0% for the low-risk ones who need to undertake FIT. Results from the multivariate logistic regression models demonstrated that there were several factors associated with the rates of participation in CRC screening, including age, background of education, history of smoking cigarettes, previous history of bowel examination, chronic inflammatory bowel disease and family history of CRC among the 1(st)-degree relatives. Conclusions: FIT and the novel risk-adapted screening approach showed superior participation rates to the colonoscopy. Further efforts including health promotion campaign for specific target population are needed to improve the engagement which ensures the effectiveness of CRC screening programs.

[Diagnostic performance of quantitative fecal immunochemical test in detection of advanced colorectal neoplasia].

Objective: To evaluate the diagnostic performance of quantitative fecal immunochemical testing (FIT) and to provide reference for designing effective colorectal cancer (CRC) screening strategy in China. Methods: Based on an ongoing randomized controlled trial comparing the colorectal cancer screening strategies, this current study involved 3 407 participants aged 50-74 years who had undergone colonoscopies. All the feces samples were collected from the participants prior to receiving the colonoscopy. Fecal hemoglobin (Hb) was tested by FIT following a standardized operation process. Diagnosis-related indicators of FIT were calculated using the colonoscopy results as the gold standard. Results: Among the 3 407 participants, the mean age (SD) as 60.5 (6.3) years and 1 753 (51.5%) were males. The participants involved 28 (0.8%) CRCs, 255 (7.5%) advanced adenomas, 677 (19.9%) nonadvanced adenomas, and 2 447 (71.8%) benign or negative findings. With an overall positivity rate of 2.8% (96/3 407) at the recommended cutoff value of 20 µg Hb/g, the sensitivities of FIT for both CRC and advanced adenoma were 57.1% (95%CI: 37.2%-75.5%) and 11.0% (95%CI: 7.4%-15.5%), respectively, with the corresponding specificity as 98.4% (95%CI: 97.8%-98.8%). At a decreased cut-off value of 5 µg Hb/g, the sensitivities for detecting CRC and advanced adenoma increased to 64.3% (95%CI: 44.1%-81.4%) and 16.5% (95%CI: 12.1%-21.6%), respectively, but the specificity reduced to 95.2% (95%CI: 94.4%-95.9%). The areas under the ROC curve for CRC and advanced adenoma were 0.908 (95%CI: 0.842-0.973) and 0.657 (95%CI: 0.621-0.692), respectively. Of the diagnostic performance, there were no significant differences noticed by different sex and age groups. Conclusions: In our study, the quantitative FIT showed modest sensitivity in detecting CRC but limited sensitivity in detecting advanced adenoma. In population-based CRC screening programs, the quantitative FIT had the advantage of adjusting the positive threshold based on the targeted detection rate and available resource load of colonoscopy.

Ferromagnetic frustration in ternary nitride ZnFe3N.

As a new antiperovskite nitride, ZnFe3N was synthesized and characterized by almost completely substituting iron atoms at corner positions of γ'-Fe4N. The magnetic interactions of the system with the space group Pm3[combining macron]m are fully investigated. The critical behavior was investigated based on the measured magnetic data around the ferromagnetic phase transition temperature. In this work, the values of critical exponents (ß, γ and δ) were obtained systematically using the Kouvel-Fisher method in the critical region. The Widom scaling law (δ = 1 + γß-1) and the scaling equation (m = f±(h)) were used to reveal the reliability of these values. The values of the critical exponents (ß = 0.325, γ = 1.228, and δ = 4.778) are different from those predicted by the three-dimensional (3D) Heisenberg model and mean-field model, and are very close to those of the 3D-Ising model. Combined with ESR analysis, the spin clusters induced by changes in chemical bonds are considered to be the cause for the existence of an anisotropic short-range ordered state in this ferromagnetic system.

[Pathological diagnosis of lung cancer based on deep transfer learning].

Objective: To establish an artificial intelligence (AI)-assisted diagnostic system for lung cancer via deep transfer learning. Methods: The researchers collected 519 lung pathologic slides from 2016 to 2019, covering various lung tissues, including normal tissues, adenocarcinoma, squamous cell carcinoma and small cell carcinoma, from the Beijing Chest Hospital, the Capital Medical University. The slides were digitized by scanner, and 316 slides were used as training set and 203 as the internal test set. The researchers labeled all the training slides by pathologists and establish a semantic segmentation model based on DeepLab v3 with ResNet-50 to detect lung cancers at the pixel level. To perform transfer learning, the researchers utilized the gastric cancer detection model to initialize the deep neural network parameters. The lung cancer detection convolutional neural network was further trained by fine-tuning of the labeled data. The deep learning model was tested by 203 slides in the internal test set and 1 081 slides obtained from TCIA database, named as the external test set. Results: The model trained with transfer learning showed substantial accuracy advantage against the one trained from scratch for the internal test set [area under curve (AUC) 0.988 vs. 0.971, Kappa 0.852 vs. 0.832]. For the external test set, the transferred model achieved an AUC of 0.968 and Kappa of 0.828, indicating superior generalization ability. By studying the predictions made by the model, the researchers obtained deeper understandings of the deep learning model. Conclusions: The lung cancer histopathological diagnostic system achieves higher accuracy and superior generalization ability. With the development of histopathological AI, the transfer learning can effectively train diagnosis models and shorten the learning period, and improve the model performance.

[Population's acceptance and attitude toward a novel fecal immunochemical test for colorectal cancer screening: a multi-center survey in China].

Objective: To investigate the acceptance and attitude toward a novel fecal immunochemical test (FIT) in colorectal cancer screening among populations in China. Methods: From May 2018 to May 2019, 2 474 people aged 50-74 years were recruited from five provinces of China (Zhejiang, Anhui, Jiangsu, Hunan and Yunnan). The general demographic characteristics, acceptance of the new FIT technology and operational difficulties through the whole screening process were obtained through questionnaire survey. Multivariate logistic regression model was used to analyze the factors related to difficulties encountered in sampling stool, reading and uploading results. Results: The subjects were (60.0±6.4) years old, and female, high school of above educated, unemployed/retired/other, married and with medical insurance status of "new rural cooperative medical care (NRCMC)" accounted for 61.7% (1 526), 29.0%(718), 34.3% (849), 92.7% (2 293) and 31.3%(775), respectively. The population's acceptance of the FIT technology was 94.8%. In the process of FIT screening, the percentage of occurred difficulties in sampling stool, reading and uploading results were 33.1% (819), 46.4% (1 147) and 62.9% (1 557), respectively. The main difficulties were the uncertainty about whether the sampling operation was standard (28.0%), the inability to accurately judge the result displayed (32.5%) and the need for help without using a smartphone (44.2%). The results of multivariate logistic regression model analysis showed that people aged 65-74 years old and with medical insurance status of "NRCMC" were more likely to encounter difficulties in sampling, and those who were unemployed/retired/other and living with 3 or more family members were less likely to encounter difficulties in sampling. Those aged 65-74 years old, farmers or migrant workers, and those with "NRCMC" were more likely to encounter difficulties in readingresults, and those with 3 or more family members were less likely to encounter difficulties in reading result. Those with "NRCMC" were more likely to encounter difficulties in uploading results, and those with education level of high school or above, living with more than 3 family members were less likely to encounter difficulties in uploading results. Conclusion: The acceptance of the new FIT technology is relatively high among the subjects. Age, education level, occupation, number of family members living together and medical insurance status might be related to difficulties encountered in sampling stool, reading and uploading results, and it can be further strengthened in terms of the technology and characteristics of sub-populations.