The Evaluation of Prognostic Value and Immune Characteristics of Ferroptosis-Related Genes in Lung Squamous Cell Carcinoma.

Background The purpose of our study was to construct a prognostic model based on ferroptosis-related gene signature to improve the prognosis prediction of lung squamous carcinoma (LUSC). Methods The mRNA expression profiles and clinical data of LUSC patients were downloaded. LUSC-related essential differentially expressed genes were integrated for further analysis. Prognostic gene signatures were identified through random forest regression and univariate Cox regression analyses for constructing a prognostic model. Finally, in a preliminary experiment, we used the reverse transcription-quantitative polymerase chain reaction assay to verify the relationship between the expression of three prognostic gene features and ferroptosis. Results Fifty-six ferroptosis-related essential genes were identified by using integrated analysis. Among these, three prognostic gene signatures (HELLS, POLR2H, and POLE2) were identified, which were positively affected by LUSC prognosis but negatively affected by immune cell infiltration. Significant overexpression of immune checkpoint genes occurred in the high-risk group. In preliminary experiments, we confirmed that the occurrence of ferroptosis can reduce three prognostic gene signature expression. Conclusions The three ferroptosis-related genes could predict the LUSC prognostic risk of antitumor immunity.

Risk factors and drug discovery for cognitive impairment in type 2 diabetes mellitus using artificial intelligence interpretation and graph neural networks.

Background: Among the 382 million diabetic patients worldwide, approximately 30% experience neuropathy, and one-fifth of these patients eventually develop diabetes cognitive impairment (CI). However, the mechanism underlying diabetes CI remains unknown, and early diagnostic methods or effective treatments are currently not available. Objective: This study aimed to explore the risk factors for CI in patients with type 2 diabetes mellitus (T2DM), screen potential therapeutic drugs for T2DM-CI, and provide evidence for preventing and treating T2DM-CI. Methods: This study focused on the T2DM population admitted to the First Affiliated Hospital of Hunan College of Traditional Chinese Medicine and the First Affiliated Hospital of Hunan University of Chinese Medicine. Sociodemographic data and clinical objective indicators of T2DM patients admitted from January 2018 to December 2022 were collected. Based on the Montreal Cognitive Assessment (MoCA) Scale scores, 719 patients were categorized into two groups, the T2DM-CI group with CI and the T2DM-N group with normal cognition. The survey content included demographic characteristics, laboratory serological indicators, complications, and medication information. Six machine learning algorithms were used to analyze the risk factors of T2DM-CI, and the Shapley method was used to enhance model interpretability. Furthermore, we developed a graph neural network (GNN) model to identify potential drugs associated with T2DM-CI. Results: Our results showed that the T2DM-CI risk prediction model based on Catboost exhibited superior performance with an area under the receiver operating characteristic curve (AUC) of 0.95 (specificity of 93.17% and sensitivity of 78.58%). Diabetes duration, age, education level, aspartate aminotransferase (AST), drinking, and intestinal flora were identified as risk factors for T2DM-CI. The top 10 potential drugs related to T2DM-CI, including Metformin, Liraglutide, and Lixisenatide, were selected by the GNN model. Some herbs, such as licorice and cuscutae semen, were also included. Finally, we discovered the mechanism of herbal medicine interventions in gut microbiota. Conclusion: The method based on Interpreting AI and GNN can identify the risk factors and potential drugs associated with T2DM-CI.

Optimizing bowel preparation for colonoscopy: A cross-sectional study of the Chinese population.

Background: The quality of bowel preparation is an important factor in the success of colonoscopy. However, the quality of bowel preparation is often affected by multiple factors. The main objective of this study was to explore the specific factors that affect the quality of bowel preparation. Methods: Patients were consecutively recruited from the gastroenterology department in Union Hospital, Tongji Medical College, Huazhong University of Science and Technology in Wuhan from May 2018 to December 2018. All patients were undergoing colonoscopy. Bowel preparation was evaluated by the Ottawa Bowel preparation Scale (OBPS) and all patients were categorized into 2 groups according to the OBPS. Multivariate analysis was conducted to identify the factors associated with bowel preparation quality. Results: A total of 910 patients were included in the analysis with an average age of 48.62 ± 13.57 years. Patient source (P < 0.001) and the preparation method (P = 0.029) were correlated with OBPS adequacy. In addition, after stratified by age, preparation method (P = 0.022) was a significant factor among patients under 50 years old; whereas waiting time (P = 0.005) was a significant factor among patients over 50 years old. Conclusion: Bowel preparation should be tailored based on the age of the patients to determine the most appropriate plan, including the most appropriate waiting time and the most appropriate purgative combination. Doctors should also focus more on the quality of bowel preparation in inpatients, who are more likely than outpatients to have an inadequate bowel preparation.

An Explainable AI Approach for the Rapid Diagnosis of COVID-19 Using Ensemble Learning Algorithms.

Background: Artificial intelligence-based disease prediction models have a greater potential to screen COVID-19 patients than conventional methods. However, their application has been restricted because of their underlying black-box nature. Objective: To addressed this issue, an explainable artificial intelligence (XAI) approach was developed to screen patients for COVID-19. Methods: A retrospective study consisting of 1,737 participants (759 COVID-19 patients and 978 controls) admitted to San Raphael Hospital (OSR) from February to May 2020 was used to construct a diagnosis model. Finally, 32 key blood test indices from 1,374 participants were used for screening patients for COVID-19. Four ensemble learning algorithms were used: random forest (RF), adaptive boosting (AdaBoost), gradient boosting decision tree (GBDT), and extreme gradient boosting (XGBoost). Feature importance from the perspective of the clinical domain and visualized interpretations were illustrated by using local interpretable model-agnostic explanations (LIME) plots. Results: The GBDT model [area under the curve (AUC): 86.4%; 95% confidence interval (CI) 0.821-0.907] outperformed the RF model (AUC: 85.7%; 95% CI 0.813-0.902), AdaBoost model (AUC: 85.4%; 95% CI 0.810-0.899), and XGBoost model (AUC: 84.9%; 95% CI 0.803-0.894) in distinguishing patients with COVID-19 from those without. The cumulative feature importance of lactate dehydrogenase, white blood cells, and eosinophil counts was 0.145, 0.130, and 0.128, respectively. Conclusions: Ensemble machining learning (ML) approaches, mainly GBDT and LIME plots, are efficient for screening patients with COVID-19 and might serve as a potential tool in the auxiliary diagnosis of COVID-19. Patients with higher WBC count, higher LDH level, or higher EOT count, were more likely to have COVID-19.

Graph neural network and multi-data heterogeneous networks for microbe-disease prediction.

The research on microbe association networks is greatly significant for understanding the pathogenic mechanism of microbes and promoting the application of microbes in precision medicine. In this paper, we studied the prediction of microbe-disease associations based on multi-data biological network and graph neural network algorithm. The HMDAD database provided a dataset that included 39 diseases, 292 microbes, and 450 known microbe-disease associations. We proposed a Microbe-Disease Heterogeneous Network according to the microbe similarity network, disease similarity network, and known microbe-disease associations. Furthermore, we integrated the network into the graph convolutional neural network algorithm and developed the GCNN4Micro-Dis model to predict microbe-disease associations. Finally, the performance of the GCNN4Micro-Dis model was evaluated via 5-fold cross-validation. We randomly divided all known microbe-disease association data into five groups. The results showed that the average AUC value and standard deviation were 0.8954 ± 0.0030. Our model had good predictive power and can help identify new microbe-disease associations. In addition, we compared GCNN4Micro-Dis with three advanced methods to predict microbe-disease associations, KATZHMDA, BiRWHMDA, and LRLSHMDA. The results showed that our method had better prediction performance than the other three methods. Furthermore, we selected breast cancer as a case study and found the top 12 microbes related to breast cancer from the intestinal flora of patients, which further verified the model's accuracy.

[Analysis on acupoint selection rules for impotence treated with acupuncture and moxibustion in ancient based on data mining technology].

Based on the data mining technology, the rules of acupoint selection and prescription were analyzed for impotence treated with acupuncture and moxibustion in ancient recorded in Zhonghua Yidian. By taking "yangwei" and "yinwei" as the searching terms, the database of Zhonghua Yidian (the 5th edition) were retrieved and the relevant information of impotence, such as prescription provision, acupoints and use frequency were extracted. Using the software, e.g. Microsoft Excel and Weka 3.8.4, the rules of acupoint selection and prescription for impotence treated with acupuncture and moxibustion in ancient were analyzed. Fifty five provisions of acupoint prescriptions were in compliance with the requirements and screened. Of them, there were 17 compound prescriptions and the rest were the single-point prescriptions, with 24 acupoints involved. Regarding the use frequency of acupoints in treatment of impotence, the top 5 acupoints were Yingu (KI 10), Ququan (LR 8), Qichong (ST 30), Taichong (LR 3) and Rangu (KI 2). The cluster analysis found that Yingu (KI 10), Ququan (LR 8)-Qichong (ST 30), Taichong (LR 3)-Rangu (KI 2)-Xingjian (LR 2), and Mingmen (GV 4)-Zhongfeng (LR 4)-Yuji (LU 10)-Yanggu (SI 5) were formed the group prescriptions respectively. Multilayer correlation analysis discovered that the commonly used meridians were the liver meridian of foot-jueyin, the kidney meridian of foot-shaoyin, the stomach meridian of foot-yangming and the conception vessel. The acupoints selected were generally on the lower extremities, the abdomen and the upper extremities. Regarding the special points, the five-shu points and the convergent points were mostly involved. By the analysis on compound prescriptions, 3 patterns of acupoint combination were discovered, the prescription by taking Yingu (KI 10), Rangu (KI 2) and Zhongfeng (LR 4) as the key points, the one by taking Shenshu (BL 23) and Yanggu (SI 5) as the key points and the relevant fixed combination of 4 acupoints, including Taichong (LR 3), Xingjian (LR 2), Neiting (ST 44) and Xiangu (ST 43). By the analysis on the compound prescriptions, 3 common meridian combinations were found, including the combination with the kidney meridian predominated, the relevant fixed combination with the liver meridian and the stomach meridian and the one with small intestine meridian and the lung meridian.

[Acupoint selection pattern of chronic atrophic gastritis based on data mining methods of latent structure model and frequency item set].

OBJECTIVE: To explore the acupoint selection pattern of chronic atrophic gastritis and provide reference for clinical treatment of chronic atrophic gastritis. METHODS: The literature regarding acupuncture for chronic atrophic gastritis published before September 5th of 2016 was searched in the databases of CNKI, CBM, PubMed, etc. The information of symptoms and acupoint selection was extracted to establish medical database of chronic atrophic gastritis. The data mining methods of latent structure model and frequency item set were applied to analyze the acupoint selection pattern of chronic atrophic gastritis. RESULTS: A total of 42 papers were collected in preliminary screening, and 32 papers were included, involving 604 medical cases. The data mining indicated 215 symptoms were involved in medical cases, including 16 high-frequency symptoms (stomach pain, stomach distension and hiccup, etc.), and the latent structure model of chronic atrophic gastritis symptoms was established. Fifty-two acupoints were identified, and high-frequency acupoints included Zusanli (ST 36), Zhongwan (CV 12), Neiguan (PC 6) and Weishu (BL 21), etc. Five frequency item sets of symptom-acupoint were identified, including stomach pain+stomach distension+Zusanli (ST 36)+Zhongwan (CV 12), etc. Six frequency item sets of symptom-syndrome-acupoint were identified, including stomach distension+dry mouth+dry defecation+insufficiency of stomach yin+Sanyinjiao (SP 6). CONCLUSION: Acupuncture for chronic atrophic gastritis selected Zusanli (ST 36), Zhongwan (CV 12) and Neiguan (PC 6) as main acupoints, and selected other acupoints based on clinical symptoms. This could provide reference for clinical treatment of chronic atrophic gastritis.

[Maximal Frequent Association Patterns between Symptoms of Coronary Heart Disease Patients with Blood Stasis Syndrome and Medication Constitutions].

A total of 121 medical cases concerning treating coronary heart disease (CHD) pa- tients with blood stasis syndrome (BSS) by Chinese Medicine were collected to establish a database for CHD patients with BSS. By using data mining authors tried to explore inherent laws of its symptoms and medication of Chinese Medicine, and to probe maximal frequent association patterns between its symp- toms and medication constitutions. Of the 121 medical cases, chest pain, chest stiffness, and headache were their common symptoms. Compatibilities of blood-activating drugs, stasis-resolving drugs, and qi- promoting drugs were most commonly used. The association between symptoms and compatibilities con- stituted a most often seen maximal frequent association pattern, which reflected an idea of treating both principal and subordinate symptoms.

Correlations between Phlegm Syndrome of Chinese Medicine and Coronary Angiography: A Systematic Review and Meta-Analysis.

Phlegm is one of the most common patterns of coronary artery disease (CAD) in Chinese medicine. Our research was aimed at investigating the association between phlegm syndrome of CAD and coronary angiography (CAG) by meta-analysis. According to inclusion criteria, a total of 30 studies involving 5,055 CAD patients were included. The meta-analysis showed that phlegm syndrome patients were prone to multivessel disease (28 studies, OR = 1.53, 95% CI, 1.24 to 1.88, P < 0.01) and higher Gensini score (2 studies, OR = 5.90, 95% CI, 1.86 to 9.94, P = 0.004), but not obviously relevant to severe stenosis (≥75%) of coronary arteries (13 studies, OR = 1.20, 95% CI, 0.63 to 2.27, P = 0.57). We concluded that the coronary arteries lesions of CAD patients with phlegm syndrome were more severe than those with nonphlegm syndromes. Phlegm syndrome should, therefore, be regarded as a dangerous pattern of CAD with worse prognosis.

[Acupoints selection rules analysis of ancient acupuncture for urinary incontinence based on data mining technology].

Based on ancient literature of acupuncture in Canon of Chinese Medicine (4th edition), the articles regarding acupuncture for urinary incontinence were retrieved and collected to establish a database. By Weka data mining software, the multi-level association rules analysis method was applied to analyze the acupoints selection characteristics and rules of ancient acupuncture for treatment of urinary incontinence. Totally 356 articles of acupuncture for urinary incontinence were collected, involving 41 acupoints with a total frequency of 364. As a result, (1) the acupoints in the yin-meridian of hand and foot were highly valued, as the frequency of acupoints in yin-meridians was 2.6 times than that in yang-meridians, and the frequency of acupoints selected was the most in the liver meridian of foot-jueyin; (2) the acupoints in bladder meridian of foot-taiyang were also highly valued, and among three yang-meridians of foot, the frequency of acupoints in the bladder meridian of foot-taiyang was 54, accounting for 65.85% (54/82); (3) more acupoints selected were located in the lower limbs and abdomen; (4) specific acupoints in above meridians were mostly selected, presenting 73.2% (30/41) to the ratio of number and 79.4% (289/364) to the frequency, respectively; (5) Zhongji (CV 3), the front-mu point of bladder meridian, was seldom selected in the ancient acupuncture literature, which was different from modern literature reports. The results show that urinary incontinence belongs to external genitalia diseases, which should be treated from yin, indicating more yin-meridians be used and special acupoints be focused on. It is essential to focus inheritance and innovation in TCM clinical treatment, and applying data mining technology to ancient literature of acupuncture could provide classic theory basis for TCM clinical treatment.