Application of machine learning techniques in the diagnostic approach of PTSD using MRI neuroimaging data: A systematic review.

Background: At present, the diagnosis of post-traumatic stress disorder（PTSD） mainly relies on clinical symptoms and psychological scales, and finding objective indicators that are helpful for diagnosis has always been a challenge in clinical practice and academic research. Neuroimaging is a useful and powerful tool for discovering the biomarkers of PTSD,especially functional MRI (fMRI), structural MRI (sMRI) and Diffusion Weighted Imaging（DTI）are the most commonly used technologies, which can provide multiple perspectives on brain function, structure and its connectivity. Machine learning (ML) is an emerging and potentially powerful method, which has aroused people's interest because it is used together with neuroimaging data to define brain structural and functional abnormalities related to diseases, and identify phenotypes, such as helping physicians make early diagnosis. Objectives: According to the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) declaration, a systematic review was conducted to assess its accuracy in distinguishing between PTSD patients, TEHC(Trauma-Exposed Healthy Controls), and HC(healthy controls). Methods: We searched PubMed, Embase, and Web of Science using common words for ML methods and PTSD until June 2023, with no language or time limits. This review includes 13 studies, with sensitivity, specificity, and accuracy taken from each publication or acquired directly from the authors. Results: All ML techniques have an diagnostic accuracy rate above 70%，and support vector machine（SVM） are the most commonly used techniques. This series of studies has revealed significant neurobiological differences in key brain regions among individuals with PTSD, TEHC, and HC. The connectivity patterns of regions such as the Insula and Amygdala hold particular significance in distinguishing these groups. TEHC exhibits more normal connectivity patterns compared to PTSD, providing valuable insights for the application of machine learning in PTSD diagnosis. Conclusion: In contrast to any currently available assessment and clinical diagnosis, ML techniques can be used as an effective and non-invasive support for early identification and detection of patients as well as for early screening of high-risk populations.

Transcriptional signature of durable effector T cells elicited by a replication defective HCMV vaccine.

Human cytomegalovirus (HCMV) is a leading infectious cause of birth defects and the most common opportunistic infection that causes life-threatening diseases post-transplantation; however, an effective vaccine remains elusive. V160 is a live-attenuated replication defective HCMV vaccine that showed a 42.4% efficacy against primary HCMV infection among seronegative women in a phase 2b clinical trial. Here, we integrated the multicolor flow cytometry, longitudinal T cell receptor (TCR) sequencing, and single-cell RNA/TCR sequencing approaches to characterize the magnitude, phenotype, and functional quality of human T cell responses to V160. We demonstrated that V160 de novo induces IE-1 and pp65 specific durable polyfunctional effector CD8 T cells that are comparable to those induced by natural HCMV infection. We identified a variety of V160-responsive T cell clones which exhibit distinctive "transient" and "durable" expansion kinetics, and revealed a transcriptional signature that marks durable CD8 T cells post-vaccination. Our study enhances the understanding of human T-cell immune responses to V160 vaccination.

[Cancer survival analysis in Tianjin, 2010 to 2016].

Objective: Survival analysis of cancers' incidence data in Tianjin from 2010 to 2016 was conducted to provide the basis for formulating and evaluating regional health policies on cancer prevention and treatment. Methods: Registration data in Tianjin were used between January 1, 2010 to December 31, 2016 and patients were followed-up till 31 December, 2021. Life-table method was used to calculate the observed survival rate and Edered Ⅱ was used to calculate the relative survival rate. The data were stratified by year, gender, age group and cancer sites. Difference in survival curves between group was analyzed by Kaplan-Meier method and Log rank test. Joinpoint regression model was used to analyze the trend change. Results: The 5-year relative survival rates of cancer were 41.92% to 53.65% from 2010 to 2016 for residents in Tianjin, with an increasing trend (t=4.81, P=0.005), and the average was 48.56%. The survival rate of females was higher than that of males (57.71%vs. 39.20%), and the survival rate of urban residents was higher than that of rural residents (49.38% vs. 47.24%). The 5-year relative survival rates were 63.14%, 78.39%, 58.25% and 32.67% in 0-14, 15-44, 45-64 and 65 and above age groups, respectively. The median relative survival times of all cancer were 2.34 to 6.00 years from 2010 to 2016 in Tianjin, with an increasing trend (t=3.86, P=0.012). The average of median relative survival times was 4.11 years. The median survival time of females was longer than that of males (11.99 years vs. 2.03 years), and the time of urban residents were longer than that of rural residents (4.60 years vs. 3.43 years). The median relative survival time were 12.07, 11.92 and 1.34 years in 15-44, 45-64 and 65 and above age groups, respectively. Conclusions: The cumulative survival rate of cancer increased significantly from 2010 to 2016 in Tianjin, indicating that the prevention and treatment effect of cancer is obvious. The focus should be on male, rural areas, higher age group, and targeted prevention and treatment measures should be taken to lung, esophagus, liver, gallbladder and pancreatic cancer.

Mouse model of Graves' orbitopathy induced by the immunization with TSHR A and IGF-1R α subunit gene.

PURPOSE: The study aimed to establish a mouse model of Graves' disease (GD) with Graves' orbitopathy (GO; GD + GO) that can represent the clinical disease characteristics. METHODS: A eukaryotic expression plasmid of insulin-like growth factor 1 receptor (IGF-1R) α subunit (pcDNA3.1/IGF-1Rα) and a thyrotropin receptor (TSHR) A subunit plasmid (pcDNA3.1/TSHR-289) were injected in female BALB/c mice followed by immediate electroporation to induce a GD + GO model. Grouping was performed according to the frequency of injection (2- to 4-week intervals) and type of injected plasmids: T: pcDNA3.1/TSHR-289( +), I: pcDNA3.1/IGF-1Rα( +), or co-injection T + I: pcDNA3.1/TSHR-289( +) and pcDNA3.1/IGF-1Rα( +). Serum TSH, T4, TSAb, TSBAb, body weight, and blood glucose levels were evaluated. Thyroid 99mTcO4- imaging and retrobulbar magnetic resonance imaging (MRI) were performed, and bilateral eye muscle volumes were measured. Immunohistochemistry and hematoxylin-eosin staining were performed on the relevant tissues, and semi-quantitative analysis was performed. RESULTS: A total of 60% of mice (3/5, one mouse died) in the T group developed GD + GO. In the T + I group, 83.3% of mice (5/6) developed GD + GO. Mice in the I group did not develop GD. Compared with the control group, serum T4, TSAb, and TSBAb of the mice in the GD + GO model groups were increased to varying degrees (P < 0.05), and serum TSH and body weight were significantly lower compared to the control group (P < 0.05). The thyroid uptake capacity of 99mTcO4- and the volume of eye muscle of mice in the GD + GO group were significantly higher compared to the control group (P < 0.05). The thyroid and retrobulbar muscles of these mice showed varying inflammatory infiltration and interstitial muscle edema. The severity of GD + GO in the co-injection group was not related to injection frequency; however, GD and ocular signs in co-injection mice were more severe compared to the T group. CONCLUSIONS: We successfully induced a GD + GO mouse model by a repeated co-injection of pcDNA3.1/IGF-1Rα and pcDNA3.1/TSHR-289 plasmids. Injection of pcDNA3.1/IGF-1Rα alone failed to induce GD. Co-injection of two plasmids induced more severe GD + GO than pcDNA3.1/TSHR-289( +) alone.

Identification and validation of a novel cuproptosis signature for stratifying different prognostic, immune, metabolic, and therapeutic landscapes in pancreatic adenocarcinoma.

OBJECTIVE: Pancreatic adenocarcinoma (PAAD) is a highly malignant cancer that urgently needs more effective therapeutic strategies. The discovery of cuproptosis brings great inspiration for the treatment and clinical assessment of cancers. MATERIALS AND METHODS: A novel cuproptosis-related (CR) risk signature was constructed using the Lasso regression analysis. Its prognostic value was assessed via a series of survival analyses and validated in four GEO cohorts. The effects of CR risk signature on tumor immune microenvironment (TIM) were explored through CIBERSORT, ESTIMATE, and ssGSEA algorithms. Using GESA, we investigated its associations with various patterns of programmed cell death (PCD) and the metabolism process. The somatic mutation features of each CR-risk group were also probed using 'maftools' R package and cBioPortal database. The potential linkages between CR risk score and the efficacy of multiple therapeutic approaches were elucidated using tumor mutation burden, the expressions of immune checkpoints, the TIDE score, and the GDSC database. Finally, we ascertained the biofunctions of LIPT1 (Lipoyltransferase 1) in pancreatic cancer (PC) cells through immunohistochemistry, qPCR (quantitative polymerase chain reaction), colony formation, and Transwell assays. RESULTS: LIPT1, LIAS (lipoyl synthase), PDP1 (Pyruvate dehydrogenase phosphatase1), and GCSH (Glycine cleavage system H protein) constituted the CR risk signature. The CR risk signature possessed a high prognostic value and could improve the traditional prognostic model. Moreover, the CR risk score was indicative of the changes in infiltration levels of CD8+T cells and macrophages, whereas it was not associated with the enrichment of various PCD patterns and multiple metabolic processes. As for therapeutic correlation, CR risk score was a potential biomarker for predicting the efficacy of ICBs but failed in targeted drugs and chemotherapeutic agents. Through qPCR and immunohistochemistry detection in clinical samples, we confirmed that LIPT1 was significantly downregulated in pancreatic adenocarcinoma (PAAD) samples. Experiments in vitro revealed that silencing LIPT1 promoted the proliferation, migration, and invasion of PANC-1 and SW1990 cells. CONCLUSIONS: The novel CR risk signature contributed to the risk stratification of PAAD patients. Cuproptosis regulatory genes, well represented by LIPT1, provided new insights into PAAD treatment and assessment.

[Application value of laparoscopic double stapler firings and double stapling technique combined with rectal eversion and total extra-abdominal resection in the sphincter-preserving resection of low rectal cancer].

Objectives: To investigate the application value of laparoscopic double stapler firings and double stapling technique combined with rectal eversion and total extra-abdominal resection (LDER) in the anal preservation treatment of low rectal cancer. Methods: Inclusion criteria: (1) age was 18-70; (2) the distance of the lower tumor edge from the anal verge was 4-5 cm; (3) primary tumor with a diameter ≤3 cm; (4) preoperative staging of T1~2N1~2M0; (5) "difficult pelvis", defined as ischial tuberosity diameter<10 cm or body mass index>25 kg/m2; (6) patients with strong intention for sphincter preservation; (7) no preoperative treatment (e.g., chemotherapy, radiotherapy, molecular targeted therapy, or immunotherapy); (8) no lateral lymph node enlargement; (9) no previous anorectal surgery; (10) patients with good basic condition who could tolerate surgery. Exclusion criteria: (1) previously suffered from malignant tumors of the digestive tract or currently suffering from malignant tumors out of the digestive tract; (2) patients with preoperative anal dysfunction (Wexner score ≥ 10), or fecal incontinence. The specific surgical steps are as follows: the distal end of the rectum was dissected to the level of the interspace between internal and external sphincters of anal canal. Five centimeters proximal to the tumor, the mesorectum was ligated, and a liner stapler was used to transect the rectum. The distal rectum with the tumor were then everted and extracted through the anus. The rectum was transected 0.5-1.0 cm distal to the tumor with a linear stapler. Full thickness suture was used to reinforce the stump of the rectum, which was then brought back into the pelvic cavity. Finally, an end-to-end anastomosis between the colon and the rectum was performed. A retrospective descriptive study was performed of the clinical and pathological data of 12 patients with T1-T2 stage low rectal cancer treated with LDER at Henan Provincial People's Hospital from January 2020 to December 2022. Results: All 12 patients successfully completed LDER with sphincter preservation, without conversion to open surgery or changes in surgical approach. The median surgical time was 272 (155-320) minutes, with a median bleeding volume of 100 (50-200) mL. No protective stoma was performed, and all patients received R0 resection. The average hospital stay was 9 (7-15) days. There were no postoperative anastomotic leakage or perioperative deaths. All 12 patients received postoperative follow-up, with a median follow-up of 12 months (6-36 months) and a Wexner score of 8 (5-14) at 6 months postoperatively. There was no tumor recurrence or metastasis during the follow-up period. Conclusions: LDER is safe and effective for the treatment of low rectal cancer.

Developing deep learning-based strategies to predict the risk of hepatocellular carcinoma among patients with nonalcoholic fatty liver disease from electronic health records.

OBJECTIVE: The accuracy of deep learning models for many disease prediction problems is affected by time-varying covariates, rare incidence, covariate imbalance and delayed diagnosis when using structured electronic health records data. The situation is further exasperated when predicting the risk of one disease on condition of another disease, such as the hepatocellular carcinoma risk among patients with nonalcoholic fatty liver disease due to slow, chronic progression, the scarce of data with both disease conditions and the sex bias of the diseases. The goal of this study is to investigate the extent to which the aforementioned issues influence deep learning performance, and then devised strategies to tackle these challenges. These strategies were applied to improve hepatocellular carcinoma risk prediction among patients with nonalcoholic fatty liver disease. METHODS: We evaluated two representative deep learning models in the task of predicting the occurrence of hepatocellular carcinoma in a cohort of patients with nonalcoholic fatty liver disease (n = 220,838) from a national EHR database. The disease prediction task was carefully formulated as a classification problem while taking censorship and the length of follow-up into consideration. RESULTS: We developed a novel backward masking scheme to deal with the issue of delayed diagnosis which is very common in EHR data analysis and evaluate how the length of longitudinal information after the index date affects disease prediction. We observed that modeling time-varying covariates improved the performance of the algorithms and transfer learning mitigated reduced performance caused by the lack of data. In addition, covariate imbalance, such as sex bias in data impaired performance. Deep learning models trained on one sex and evaluated in the other sex showed reduced performance, indicating the importance of assessing covariate imbalance while preparing data for model training. CONCLUSIONS: The strategies developed in this work can significantly improve the performance of hepatocellular carcinoma risk prediction among patients with nonalcoholic fatty liver disease. Furthermore, our novel strategies can be generalized to apply to other disease risk predictions using structured electronic health records, especially for disease risks on condition of another disease.

Ensemble pretrained language models to extract biomedical knowledge from literature.

OBJECTIVES: The rapid expansion of biomedical literature necessitates automated techniques to discern relationships between biomedical concepts from extensive free text. Such techniques facilitate the development of detailed knowledge bases and highlight research deficiencies. The LitCoin Natural Language Processing (NLP) challenge, organized by the National Center for Advancing Translational Science, aims to evaluate such potential and provides a manually annotated corpus for methodology development and benchmarking. MATERIALS AND METHODS: For the named entity recognition (NER) task, we utilized ensemble learning to merge predictions from three domain-specific models, namely BioBERT, PubMedBERT, and BioM-ELECTRA, devised a rule-driven detection method for cell line and taxonomy names and annotated 70 more abstracts as additional corpus. We further finetuned the T0pp model, with 11 billion parameters, to boost the performance on relation extraction and leveraged entites' location information (eg, title, background) to enhance novelty prediction performance in relation extraction (RE). RESULTS: Our pioneering NLP system designed for this challenge secured first place in Phase I-NER and second place in Phase II-relation extraction and novelty prediction, outpacing over 200 teams. We tested OpenAI ChatGPT 3.5 and ChatGPT 4 in a Zero-Shot setting using the same test set, revealing that our finetuned model considerably surpasses these broad-spectrum large language models. DISCUSSION AND CONCLUSION: Our outcomes depict a robust NLP system excelling in NER and RE across various biomedical entities, emphasizing that task-specific models remain superior to generic large ones. Such insights are valuable for endeavors like knowledge graph development and hypothesis formulation in biomedical research.

Systematic characterization of protein structural features of alternative splicing isoforms using AlphaFold 2.

Alternative splicing is an important cellular process in eukaryotes, altering pre-mRNA to yield multiple protein isoforms from a single gene. However, our understanding of the impact of alternative splicing events on protein structures is currently constrained by a lack of sufficient protein structural data. To address this limitation, we employed AlphaFold 2, a cutting-edge protein structure prediction tool, to conduct a comprehensive analysis of alternative splicing for approximately 3,000 human genes, providing valuable insights into its impact on the protein structural. Our investigation employed state of the art high-performance computing infrastructure to systematically characterize structural features in alternatively spliced regions and identified changes in protein structure following alternative splicing events. Notably, we found that alternative splicing tends to alter the structure of residues primarily located in coils and beta-sheets. Our research highlighted a significant enrichment of loops and highly exposed residues within human alternatively spliced regions. Specifically, our examination of the Septin-9 protein revealed potential associations between loops and alternative splicing, providing insights into its evolutionary role. Furthermore, our analysis uncovered two missense mutations in the Tau protein that could influence alternative splicing, potentially contributing to the pathogenesis of Alzheimer's disease. In summary, our work, through a thorough statistical analysis of extensive protein structural data, sheds new light on the intricate relationship between alternative splicing, evolution, and human disease.

[Injury death spectrum in children aged 0-14 years and its urban-rural difference in Tianjin, 1999-2021].

Objective: To analyze the characteristics, change of injury death spectrum in children aged 0-14 years and its urban-rural difference in Tianjin. Methods: The incidence data of injury death in children aged 0-14 years in Tianjin from 1999 to 2021 were collected from the "Population Based Mortality Surveillance System in Tianjin". We calculated constituent ratio, crude and standardized mortality rates in different subgroups of the population and major injury subtypes, and compared the rural-urban differences. The Cochran-Armitage trend test was used to analyze temporal trends in cause-of-death component ratios. Joinpoint regression model was used to calculate average annual percent change (AAPC). Seasonal difference in injury mortality risk were expressed as mortality ratio and 95% confidence interval. Results: From 1999 to 2021, injury was the third cause of death in children aged 0-14 years in Tianjin. The percentage of children who died in health-care facilities in rural area was 31.08%, which was lower than 37.82% in urban area. There was a downward trend in the standardized mortality rate of injury in children (AAPC=-5.54%, P<0.001). The standardized mortality rates of drowning and road traffic injury declined in both urban area and rural area (P<0.001). The mortality rate of accidental poisoning decreased in rural area (AAPC=-8.09%, P<0.001), but showed no significant change trend in urban area (P>0.05). The standardized mortality rate of suicide showed no significant change trend in urban area, but there was an increasing trend in the standardized mortality rate of suicide in rural children aged 10-14 years (AAPC=4.58%). No significant change trend was observed in mortality rate of falls in urban and rural children (P>0.05). The injury mortality rate showed obvious seasonality in children in Tianjin. Overall injury death risk and risk for drowning-caused death were highest in summer in both urban area and rural area. The risk for road traffic injury-related death was highest in autumn in urban area and in summer in rural area. The risk for death caused by accidental poisoning was highest in winter in both urban area and rural area. Conclusions: In recent decades, the injury mortality rate in children aged 0-14 years in Tianjin showed a decreasing trend. There is still a significant difference in the injury mortality level between urban area and rural area, to which close attention needs to be paid in the future policy development.

The IL6/JAK/STAT3 signaling axis is a therapeutic vulnerability in SMARCB1-deficient bladder cancer.

SMARCB1 loss has long been observed in many solid tumors. However, there is a need to elucidate targetable pathways driving growth and metastasis in SMARCB1-deficient tumors. Here, we demonstrate that SMARCB1 deficiency, defined as genomic SMARCB1 copy number loss associated with reduced mRNA, drives disease progression in patients with bladder cancer by engaging STAT3. SMARCB1 loss increases the chromatin accessibility of the STAT3 locus in vitro. Orthotopically implanted SMARCB1 knockout (KO) cell lines exhibit increased tumor growth and metastasis. SMARCB1-deficient tumors show an increased IL6/JAK/STAT3 signaling axis in in vivo models and patients. Furthermore, a pSTAT3 selective inhibitor, TTI-101, reduces tumor growth in SMARCB1 KO orthotopic cell line-derived xenografts and a SMARCB1-deficient patient derived xenograft model. We have identified a gene signature generated from SMARCB1 KO tumors that predicts SMARCB1 deficiency in patients. Overall, these findings support the clinical evaluation of STAT3 inhibitors for the treatment of SMARCB1-deficient bladder cancer.

Correlation of lipocalin 2 and glycolipid metabolism and body composition in a large cohort of children with osteogenesis imperfecta.

PURPOSE: Lipocalin 2 (LCN2) is a newly recognized bone-derived factor that is important in regulation of energy metabolism. We investigated the correlation of serum LCN2 levels and glycolipid metabolism, and body composition in a large cohort of patients with osteogenesis imperfecta (OI). METHODS: A total of 204 children with OI and 66 age- and gender-matched healthy children were included. Circulating levels of LCN2 and osteocalcin were measured by enzyme-linked immunosorbent assay. Serum levels of fasting blood glucose (FBG), triglyceride (TG), total cholesterol (TC), and low- and high-density lipoprotein cholesterol (LDL-C, HDL-C) were measured by automated chemical analyzers. The body composition was measured by dual-energy X-ray absorptiometry. Grip strength and timed-up-and-go (TUG) were tested to evaluate the muscle function. RESULTS: Serum LCN2 levels were 37.65 ± 23.48 ng/ml in OI children, which was significantly lower than those in healthy control (69.18 ± 35.43 ng/ml, P < 0.001). Body mass index (BMI) and serum FBG level were significantly higher and HDL-C levels were lower in OI children than healthy control (all P < 0.01). Grip strength was significantly lower (P < 0.05), and the TUG was significantly longer in OI patients than healthy control (P < 0.05). Serum LCN2 level was negatively correlated to BMI, FBG, HOMA-IR, HOMA-ß, total body, and trunk fat mass percentage, and positively correlated to total body and appendicular lean mass percentage (all P < 0.05). CONCLUSIONS: Insulin resistance, hyperglycemia, obesity, and muscle dysfunction are common in OI patients. As a novel osteogenic cytokine, LCN2 deficiency may be relevant to disorders of glucose and lipid metabolism, and dysfunction of muscle in OI patients.

Spatial distribution, movement, body damage, and feather condition of laying hens in a multi-tier system.

The welfare and health of laying hens in the multitier system raise concern in public. The flock distributions during feeding time at 51 and 89 wk were studied in a multitier system. Furthermore, the ultra-high frequency radio frequency identification (UHF RFID) equipment was used to identify the transition between tiers and time spent in each tier of 48 focal hens (12 hens from each tier-group of the multitier system) at 92 wk of age. The body weight, tibia size (length and width), body damage (comb and rear part), and feather condition (neck, breast, back, tail, cloaca, and wings) of focal hens from different tier-groups were further compared. The results showed that the spatial distribution in flocks changed from top to bottom with increasing age. The hens at 51 wk of age were mainly distributed in the 4th tier (19.6 ± 5.0% in 1st tier, 9.6 ± 1.1% in 2nd tier, 23.6 ± 2.9% in 3rd tier and 47.3 ± 2.6% in 4th tier), and hens at 89 wk of age were mainly distributed in the lower tiers (33.5 ± 1.5% in 1st tier, 31.9 ± 5.1% in 2nd tier, 15.7 ± 3.4% in 3rd tier and 16.6 ± 3.1% in 4th tier). The spatial distribution of hens at 89 wk of age was more even than that at 51 wk of age. At 92 wk of age, the proportion of time spent in original tier of 4 tier-groups was 91.0 ± 5.7%, 51.9 ± 5.7%, 59.0 ± 7.0% and 63.0 ± 6.7%, respectively. Focal hens preferred to stay in the original tier and spent significantly less time in other tiers (P < 0.05). There was no significant difference in body weight, body damage score, tibia width and partial feather scores (neck, breast, tail, and cloaca) of focal hens among 4 tier-groups (P > 0.05). However, focal hens from 1st tier had worse feather scores on wings and back, and shorter tibia length compared to other tiers suggesting that there were more lower ranking birds that located in lower tier to avoid competition, but had equal access to resource, which is good for their welfare and health. In summary, the overcrowding situation was improved near the end of the laying cycle in the multitier system, thereby mitigating the potential negative effects to the lower ranking hens and maintain a satisfactory level of welfare and health for laying hens near the end of the laying cycle.

Development of the control system for electron cyclotron resonance heating on J-TEXT tokamak.

A 105 GHz/500 kW/1 s electron cyclotron resonance heating (ECRH) system has been developed on J-TEXT tokamak since 2017. The core component of the ECRH system is a gyrotron manufactured by Gyrotron Complexes Ltd. (GYCOM Ltd.), which generates microwaves of a certain frequency and power. To guarantee safe and stable operation, it is necessary to design a specialized control system. The control system is expected to perform time sequence trigger, protection, signal monitoring, communication, and data acquisition. The hardware is built with real-time processors and data acquisition modules from National Instruments. The control program is realized by LabVIEW. Test results indicate that the control system can commit stable and safe operation of the gyrotron, which guarantees the integrated commissioning tests of the whole ECRH system and ECRH related physics experiments. Under the operation of this control system, the gyrotron can generate microwaves as expected, and the ECRH system is well protected when a fault takes place.

[Clinical analysis of 10 cases of multi-center tumor necrosis factor receptor-associated periodic syndrome].

Objective: To summarize the clinical characteristics of tumour necrosis factor receptor-associated periodic syndrome (TRAPS) in children. Methods: The clinical manifestations, laboratory tests, genetic testing and follow-up of 10 children with TRAPS from May 2011 to May 2021 in 6 hospitals in China were retrospectively analyzed. Results: Among the 10 patients with TRAPS, including 8 boys and 2 girls. The age of onset was 2 (1, 5) years, the age of diagnosis was (8±4) years, and the time from onset to diagnosis was 3 (1, 7) years. A total of 7 types of TNFRSF1A gene variants were detected, including 5 paternal variations, 1 maternal variation and 4 de novo variations. Six children had a family history of related diseases. Clinical manifestations included recurrent fever in 10 cases, rash in 4 cases, abdominal pain in 6 cases, joint involvement in 6 cases, periorbital edema in 1 case, and myalgia in 4 cases. Two patients had hematological system involvement. The erythrocyte sedimentation rate and C-reactive protein were significantly increased in 10 cases. All patients were negative for autoantibodies. In the course of treatment, 5 cases were treated with glucocorticoids, 7 cases with immunosuppressants, and 7 cases with biological agents. Conclusions: TRAPS is clinically characterized by recurrent fever accompanied by joint, gastrointestinal, skin, and muscle involvement. Inflammatory markers are elevated, and autoantibodies are mostly negative. Treatment mainly involves glucocorticoids, immunosuppressants, and biological agents.

[Analysis on the current situation and influencing factors of residents' satisfaction with the built environment of China's Hygienic City Initiative].

Objective: To understand the current situation and the main influencing factors of residents' satisfaction with the built environment of China's Hygienic City Initiative. Methods: From the list of China's hygienic cities (excluding county-level cities), 61 cities were randomly selected in equal proportion and the eligible respondents were randomly selected by using the "Questionnaire Star" network platform to carry out the online questionnaire survey. A self-made satisfaction evaluation scale was used to investigate the satisfaction of the included respondents with the urban built environment and search for relevant data on the city level. The two-level multi-factor mixed effect model was constructed to analyze the influencing factors of residents' satisfaction with the built environment of China's Hygienic City Initiative. Results: The age range of 2 465 respondents was mainly between 18 and 40 years old (79.9%), with males being the main group (45.8%). The total score of residents' satisfaction with the built environment of China's hygienic cities was (69.14±13.24) points. Based on four standardized dimensions of sense of gain, the result showed that the satisfaction of urban governance had the highest score (65.08 points), followed by urban environmental sanitation (63.68 points), urban lifestyle (59.97 points) and urban basic function (59.02 points). The analysis results of the two-level multi-factor mixed effect model showed that compared with residents with an annual average concentration of inhalable fine particles in the environment>48 micrograms/cubic meter, residents with an average concentration between 38 and 48 micrograms/cubic meter [ß (95%CI): 1.65 (0.08, 3.21)] and≤37 micrograms/cubic meter or less [ß (95%CI): 1.98 (0.53, 3.43)] had higher satisfaction. Compared with residents whose proportion of the secondary industry to GDP was≤40.9%, residents in cities with a larger proportion had a lower satisfaction level [residents with a proportion of 40.9%-48.03%, ß (95%CI):-2.21 (-3.93, -0.49); residents with a proportion greater than 48.03%, ß (95%CI):-2.58 (-4.58, -0.59)]. Compared with residents with a junior high school or lower education level, residents with a higher education level had a lower satisfaction level [ß (95%CI):-2.37 (-4.57, -0.17)]. Residents of universities and above [ß (95%CI):-3.82 (-6.05, -1.60)], regularly participate in physical exercise [ß (95%CI): 5.78 (4.71, 6.84)] and self-rated good health status [ß (95%CI): 6.39 (5.33, 7.45)] had a higher satisfaction level. Conclusion: The satisfaction of residents with the built environment of China's hygienic cities is still acceptable. Satisfaction is related to individual characteristics such as residents' cultural level, type of residence, frequent participation in physical exercise, and self-rated good health status, as well as urban-level factors such as green coverage rate in built-up areas, annual average concentration of inhalable fine particles, and the proportion of GDP in the secondary industry.

Developing deep learning-based strategies to predict the risk of hepatocellular carcinoma among patients with nonalcoholic fatty liver disease from electronic health records.

Background: Deep learning models showed great success and potential when applied to many biomedical problems. However, the accuracy of deep learning models for many disease prediction problems is affected by time-varying covariates, rare incidence, and covariate imbalance when using structured electronic health records data. The situation is further exasperated when predicting the risk of one disease on condition of another disease, such as the hepatocellular carcinoma risk among patients with nonalcoholic fatty liver disease due to slow, chronic progression, the scarce of data with both disease conditions and the sex bias of the diseases. Objective: The goal of this study is to investigate the extent to which time-varying covariates, rare incidence, and covariate imbalance influence deep learning performance, and then devised strategies to tackle these challenges. These strategies were applied to improve hepatocellular carcinoma risk prediction among patients with nonalcoholic fatty liver disease. Methods: We evaluated two representative deep learning models in the task of predicting the occurrence of hepatocellular carcinoma in a cohort of patients with nonalcoholic fatty liver disease (n = 220,838) from a national EHR database. The disease prediction task was carefully formulated as a classification problem while taking censorship and the length of follow-up into consideration. Results: We developed a novel backward masking scheme to evaluate how the length of longitudinal information after the index date affects disease prediction. We observed that modeling time-varying covariates improved the performance of the algorithms and transfer learning mitigated reduced performance caused by the lack of data. In addition, covariate imbalance, such as sex bias in data impaired performance. Deep learning models trained on one sex and evaluated in the other sex showed reduced performance, indicating the importance of assessing covariate imbalance while preparing data for model training. Conclusions: Devising proper strategies to address challenges from time-varying covariates, lack of data, and covariate imbalance can be key to counteracting data bias and accurately predicting disease occurrence using deep learning models. The novel strategies developed in this work can significantly improve the performance of hepatocellular carcinoma risk prediction among patients with nonalcoholic fatty liver disease. Furthermore, our novel strategies can be generalized to apply to other disease risk predictions using structured electronic health records, especially for disease risks on condition of another disease.

Defining metabolic flexibility in hair follicle stem cell induced squamous cell carcinoma.

Among the numerous changes associated with the transformation to cancer, cellular metabolism is one of the first discovered and most prominent[1, 2]. However, despite the knowledge that nearly every cancer is associated with the strong upregulation of various metabolic pathways, there has yet to be much clinical progress on the treatment of cancer by targeting a single metabolic enzyme directly[3-6]. We previously showed that inhibition of glycolysis through lactate dehydrogenase (LDHA) deletion in cancer cells of origin had no effect on the initiation or progression of cutaneous squamous cell carcinoma[7], suggesting that these cancers are metabolically flexible enough to produce the necessary metabolites required for sustained growth in the absence of glycolysis. Here we focused on glutaminolysis, another metabolic pathway frequently implicated as important for tumorigenesis in correlative studies. We genetically blocked glutaminolysis through glutaminase (GLS) deletion in cancer cells of origin, and found that this had little effect on tumorigenesis, similar to what we previously showed for blocking glycolysis. Tumors with genetic deletion of glutaminolysis instead upregulated lactate consumption and utilization for the TCA cycle, providing further evidence of metabolic flexibility. We also found that the metabolic flexibility observed upon inhibition of glycolysis or glutaminolysis is due to post-transcriptional changes in the levels of plasma membrane lactate and glutamine transporters. To define the limits of metabolic flexibility in cancer initiating hair follicle stem cells, we genetically blocked both glycolysis and glutaminolysis simultaneously and found that frank carcinoma was not compatible with abrogation of both of these carbon utilization pathways. These data point towards metabolic flexibility mediated by regulation of nutrient consumption, and suggest that treatment of cancer through metabolic manipulation will require multiple interventions on distinct pathways.