Artificial intelligence in the risk prediction models of cardiovascular disease and development of an independent validation screening tool: a systematic review.

BACKGROUND: A comprehensive overview of artificial intelligence (AI) for cardiovascular disease (CVD) prediction and a screening tool of AI models (AI-Ms) for independent external validation are lacking. This systematic review aims to identify, describe, and appraise AI-Ms of CVD prediction in the general and special populations and develop a new independent validation score (IVS) for AI-Ms replicability evaluation. METHODS: PubMed, Web of Science, Embase, and IEEE library were searched up to July 2021. Data extraction and analysis were performed for the populations, distribution, predictors, algorithms, etc. The risk of bias was evaluated with the prediction risk of bias assessment tool (PROBAST). Subsequently, we designed IVS for model replicability evaluation with five steps in five items, including transparency of algorithms, performance of models, feasibility of reproduction, risk of reproduction, and clinical implication, respectively. The review is registered in PROSPERO (No. CRD42021271789). RESULTS: In 20,887 screened references, 79 articles (82.5% in 2017-2021) were included, which contained 114 datasets (67 in Europe and North America, but 0 in Africa). We identified 486 AI-Ms, of which the majority were in development (n = 380), but none of them had undergone independent external validation. A total of 66 idiographic algorithms were found; however, 36.4% were used only once and only 39.4% over three times. A large number of different predictors (range 5-52,000, median 21) and large-span sample size (range 80-3,660,000, median 4466) were observed. All models were at high risk of bias according to PROBAST, primarily due to the incorrect use of statistical methods. IVS analysis confirmed only 10 models as "recommended"; however, 281 and 187 were "not recommended" and "warning," respectively. CONCLUSION: AI has led the digital revolution in the field of CVD prediction, but is still in the early stage of development as the defects of research design, report, and evaluation systems. The IVS we developed may contribute to independent external validation and the development of this field.

Pitfalls in Developing Machine Learning Models for Predicting Cardiovascular Diseases: Challenge and Solutions.

UNSTRUCTURED: In recent years, there has been an explosive development of artificial intelligence (AI), which has been widely applied in the healthcare field. As a typical AI technology, machine learning (ML) models have emerged as great potential in predicting cardiovascular diseases (CVDs) by leveraging large amounts of medical data for training and optimization, which are expected to play a crucial role in reducing the incidence and mortality rates of CVDs. Although the field has become a research hotspot, there are still many pitfalls that researchers need to pay close attention to. These pitfalls may affect the predictive performance, credibility, reliability, reproducibility of the studied models, ultimately reducing the value of the research and affecting the prospects for clinical application. Therefore, identifying and avoiding these pitfalls is a crucial task before implementing the research. However, there is currently a lack of comprehensive summary on this topic. This viewpoint aims to analyze the existing problems in terms of data quality, dataset characteristics, model design and statistical methods as well as clinic implication, and provide possible solutions to these problems, like gathering objective data, improving training, repeating measurements, increasing sample size, preventing overfitting using statistical methods, utilizing specific AI algorithms to address targeted issues, standardizing outcomes and evaluation criteria, as well as enhancing fairness and replicability, with the goal of offering reference and assistance to researchers, algorithm developers, policy makers, and clinical practitioners.

[Multi-index components of Scutellariae Barbatae Herba according to UHPLC-QTRAP-MS coupled with chemometrics].

The present study developed an ultra-fast liquid chromatography coupled with triple quadrupole-linear ion trap composite mass spectrometry(UHPLC-QTRAP-MS) to simultaneously determine the content of potential active components in Scutellariae Barbatae Herba and also to provide a reference approach for screening out the differential quality control components among different batches of Scutellariae Barbatae Herba. Chromatographic separations were conducted on a Thermo Acclaim~(TM) RSLC 120 C_(18) column(3.0 mm×100 mm, 2.2 µm) in a gradient program. The mobile phase consisted of 0.1% aqueous formic acid and acetonitrile, and the column temperature was maintained at 40 ℃. The flow rate was 0.4 mL·min~(-1) and the injection volume was 2 µL. The targeted compounds were monitored in the multiple reaction monitoring(MRM) mode. The acquired data were processed by hierarchical cluster analysis(HCA) and partial least square discriminant analysis(PLS-DA). Sixteen compounds all showed good linear relationship within the corresponding linear ranges and the R~2 values were all higher than 0.993 2. The RSDs of precision, repeatability, and stability were less than or equal to 3.7%. Mean recovery rates were in the range of 95.67% and 104.8% with RSDs≤3.2%. According to HCA and PLS-DA, all samples were clustered into four categories. Scutellarin, acteoside, scutellarein, and scutebarbatine X(VIP>1) were considered as differential chemical markers in the four categories. In conclusion, the developed method can be used for the simulta-neous determination of the multiple components and quality control of Scutellariae Barbatae Herba.

Association between cytokine profiles and lung injury in COVID-19 pneumonia.

BACKGROUND: Coronavirus disease 2019 (COVID-19) is a new respiratory and systemic disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection. The purpose of the present study was to investigate the association between cytokine profiles and lung injury in COVID-19 pneumonia. METHODS: This retrospective study was conducted in COVID-19 patients. Demographic characteristics, symptoms, signs, underlying diseases, and laboratory data were collected. The patients were divided into COVID-19 with pneumonia and without pneumonia. CT severity score and PaO2/FiO2 ratio were used to assess lung injury. RESULTS: 106 patients with 12 COVID-19 without pneumonia and 94 COVID-19 with pneumonia were included. Compared with COVID-19 without pneumonia, COVID-19 with pneumonia had significantly higher serum interleukin (IL)-2R, IL-6, and tumor necrosis factor (TNF)-α. Correlation analysis showed that CT severity score and PaO2/FiO2 were significantly correlated with age, presence of any coexisting disorder, lymphocyte count, procalcitonin, IL-2R, and IL-6. In multivariate analysis, log IL6 was the only independent explanatory variables for CT severity score (ß = 0.397, p < 0.001) and PaO2/FiO2 (ß = - 0.434, p = 0.003). CONCLUSIONS: Elevation of circulating cytokines was significantly associated with presence of pneumonia in COVID-19 and the severity of lung injury in COVID-19 pneumonia. Circulating IL-6 independently predicted the severity of lung injury in COVID-19 pneumonia.

Metabolomics approach for predicting response to neoadjuvant chemotherapy for colorectal cancer.

INTRODUCTION: Colorectal cancer (CRC) is a clinically heterogeneous disease, which necessitates a variety of treatments and leads to different outcomes. Only some CRC patients will benefit from neoadjuvant chemotherapy (NACT). OBJECTIVES: An accurate prediction of response to NACT in CRC patients would greatly facilitate optimal personalized management, which could improve their long-term survival and clinical outcomes. METHODS: In this study, plasma metabolite profiling was performed to identify potential biomarker candidates that can predict response to NACT for CRC. Metabolic profiles of plasma from non-response (n = 30) and response (n = 27) patients to NACT were studied using UHPLC-quadruple time-of-flight)/mass spectrometry analyses and statistical analysis methods. RESULTS: The concentrations of nine metabolites were significantly different when comparing response to NACT. The area under the receiver operating characteristic curve value of the potential biomarkers was up to 0.83 discriminating the non-response and response group to NACT, superior to the clinical parameters (carcinoembryonic antigen and carbohydrate antigen 199). CONCLUSION: These results show promise for larger studies that could result in more personalized treatment protocols for CRC patients.

Optimization of green deep eutectic solvent (DES) extraction of Chenopodium quinoa Willd. husks saponins by response surface methodology and their antioxidant activities.

Quinoa saponins have outstanding activity, and there are an increasing number of extraction methods, but there are few research programs on green preparation technology. The extraction conditions of quinoa saponins with deep eutectic solvents (DESs) were optimized by single-factor experiments combined with response surface methodology. The antioxidant capacity of saponins extracted by DESs and traditional methods was evaluated by the DPPH clearance rate, iron ion chelation rate and potassium ferricyanide reducing power. The results show that the optimal DES is choline chloride: 1,2-propylene glycol (1 : 1), and its water content is 40%. The optimal extraction conditions were as follows: the solid-to-solvent ratio was 0.05 g mL-1, the extraction time was 89 min, and the extraction temperature was 75 °C. Under these conditions, the extraction of quinoa saponins by DES was more effective than the traditional extraction methods. The saponins extracted by DES and traditional methods were analyzed by UPLC-MS, and five main saponins were identified. Quantitative analysis by HPLC-UV showed that Q1 (m/z = 971) and Q2 (m/z = 809) had higher contents of saponins. In vitro antioxidant experiments showed that all DES saponin extracts showed good antioxidant capacity. This study provides new insight into the development and utilization of quinoa saponins.

Risk factors for mortality of coronavirus disease-2019 (COVID-19) patients in two centers of Hubei province, China: A retrospective analysis.

PURPOSE: Since the outbreak in late December 2019 in Wuhan, China, coronavirus disease-2019 (COVID-19) has become a global pandemic. We analyzed and compared the clinical, laboratory, and radiological characteristics between survivors and non-survivors and identify risk factors for mortality. METHODS: Clinical and laboratory variables, radiological features, treatment approach, and complications were retrospectively collected in two centers of Hubei province, China. Cox regression analysis was conducted to identify the risk factors for mortality. RESULTS: A total of 432 patients were enrolled, and the median patient age was 54 years. The overall mortality rate was 5.09% (22/432). As compared with the survivor group (n = 410), those in the non-survivor group (n = 22) were older, and they had a higher frequency of comorbidities and were more prone to suffer from dyspnea. Several abnormal laboratory variables indicated that acute cardiac injury, hepatic damage, and acute renal insufficiency were detected in the non-survivor group. Non-surviving patients also had a high computed tomography (CT) score and higher rate of consolidation. The most common complication causing death was acute respiratory distress syndrome (ARDS) (18/22, 81.8%). Multivariate Cox regression analysis revealed that hemoglobin (Hb) <90 g/L (hazard ratio, 10.776; 95% confidence interval, 3.075-37.766; p<0.0001), creatine kinase (CK-MB) >8 U/L (9.155; 2.424-34.584; p = 0.001), lactate dehydrogenase (LDH) >245 U/L (5.963; 2.029-17.529; p = 0.001), procalcitonin (PCT) >0.5 ng/ml (7.080; 1.671-29.992; p = 0.008), and CT score >10 (39.503; 12.430-125.539; p<0.0001) were independent risk factors for the mortality of COVID-19. CONCLUSIONS: Low Hb, high LDH, PCT, and CT score on admission were the predictors for mortality and could assist clinicians in early identification of poor prognosis among COVID-19 patients.

Prognostic value of neutrophil-to-lymphocyte ratio, lactate dehydrogenase, D-dimer, and computed tomography score in patients with coronavirus disease 2019.

BACKGROUND: This study aimed to explore the significance of neutrophil-to-lymphocyte ratio (NLR), lactate dehydrogenase (LDH), D-dimer, and CT score in evaluating the severity and prognosis of coronavirus disease 2019 (COVID-19). METHODS: Patients with laboratory-confirmed COVID-19 were retrospectively enrolled. The baseline data, laboratory findings, chest computed tomography (CT) results evaluated by CT score on admission, and clinical outcomes were collected and compared. Logistic regression was used to assess the independent relationship between the baseline level of the four indicators (NLR, LDH, D-dimer, and CT score) and the severity of COVID-19. RESULTS: Among the 432 patients, 125 (28.94%) and 307 (71.06%) were placed in the severe and non-severe groups, respectively. As per the multivariate logistic regression, high levels of NLR and LDH were independent predictors of severe COVID-19 (OR=2.163; 95% CI=1.162-4.026; p=0.015 for NLR>3.82; OR=2.298; 95% CI=1.327-3.979; p=0.003 for LDH>246 U/L). Combined NLR>3.82 and LDH>246 U/L increased the sensitivity of diagnosis in patients with severe disease (NLR>3.82 [50.40%] vs. combined diagnosis [72.80%]; p=0.0007; LDH>246 [59.2%] vs. combined diagnosis [72.80%]; p<0.0001). CONCLUSIONS: High levels of serum NLR and LDH have potential value in the early identification of patients with severe COVID-19. Moreover, the combination of LDH and NLR can improve the sensitivity of diagnosis.

A Novel RFXANK Mutation in a Chinese Child With MHC II Deficiency: Case Report and Literature Review.

Major histocompatibility complex (MHC) II deficiency is a rare primary immunodeficiency disorder that is characterized by the deficiency of MHC class II molecules. The disease is caused by transcription factor mutations including class II transactivator (CIITA), regulatory factor X-5 (RFX5), RFX-associated protein (RFXAP), and RFXAP-containing ankyrin repeat (RFXANK), respectively. Mutations in the RFXANK gene account for >70% of all known patients worldwide. Herein, we reported a 10-month-old boy with MHC II deficiency caused by a novel mutation in the RFXANK gene (c.337 + 1G>C). The boy was admitted to the hospital due to pneumonia and diarrhea at 4 months of age. Genetic analysis revealed a novel homozygous mutation in the RFXANK gene, which derived from the c.337 + 1G>C heterozygous mutations in the RFXANK gene of his parents. The boy died 3 months after diagnosis. More than 200 cases have been reported, and a review of the literature revealed different mutation rates of 4 transcription factors in different countries or regions. This is the first case report of MHC II deficiency from East Asia. We also describe all gene mutations that cause MHC II deficiency and the epidemiology of MHC II deficiency with gene mutations in this paper.

Clinicopathological features and prognosis of patients with gastric neuroendocrine tumors: A population-based study.

BACKGROUND: Despite its rarity, studies have shown the incidence of gastric neuroendocrine tumors (G-NETs) is increasing. This study investigated the risk factors affecting the survival of G-NETs patients and their prognosis over time. METHOD: A retrospective analysis of 506 G-NETs patients who underwent surgery for nonmetastatic disease from the Surveillance, Epidemiology and End Result database from 1988 to 2011 was conducted. Multivariate Cox regression analyses identified the prognostic factors affecting overall survival (OS) and disease-specific survival (DSS). Three-year conditional survival (COS3 and CDS3) estimates at "x" year after treatment were calculated as follows: COS3 = OS(x + 3)/OS(x) and CDS3 = DSS(x + 3)/DSS(x). RESULTS: The 1-, 3-, and 5-year OS rates of all patients after surgery were 90.2%, 77.3%, and 68.8%, respectively. The 1-, 3-, and 5-year DSS rates after surgery were 93.9%, 84.5%, and 80.9%, respectively. In the multivariate analysis, age, tumor grade, and T stage were independent prognostic factors of OS and DSS (all P < 0.05). With 1-, 3-, and 5-year survivorship, the COS3 improved by +5.2 (82.2%), +7.2 (84.4%), and +8.5 (85.5%), respectively, and the CDS3 improved by +4.4 (89.4%), +9.1 (94.1%), and +12.5 (97.5%), respectively. Notably, the CDS3 improved dramatically among patients with advanced stage disease (eg, N0 stage: 93.0%-98.9%, Δ5.9% vs N1 stage: 52.0%-95.7%, Δ43.7%). CONCLUSION: For G-NETs patients, age, tumor grade, T stage, and N stage were the clinicopathological factors significantly associated with prognosis. There were excellent outcomes for most G-NETs patients, with a CDS3 of greater than 90% across all independent prognostic factors after 5 years of survival.