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

In recent years, there has been explosive development in artificial intelligence (AI), which has been widely applied in the health care field. As a typical AI technology, machine learning models have emerged with great potential in predicting cardiovascular diseases 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 cardiovascular diseases. Although the field has become a research hot spot, there are still many pitfalls that researchers need to pay close attention to. These pitfalls may affect the predictive performance, credibility, reliability, and 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 a comprehensive summary on this topic. This viewpoint aims to analyze the existing problems in terms of data quality, data set characteristics, model design, and statistical methods, as well as clinical implications, and provide possible solutions to these problems, such as gathering objective data, improving training, repeating measurements, increasing sample size, preventing overfitting using statistical methods, using specific AI algorithms to address targeted issues, standardizing outcomes and evaluation criteria, and enhancing fairness and replicability, with the goal of offering reference and assistance to researchers, algorithm developers, policy makers, and clinical practitioners.

Glycolytic PFKFB3 and Glycogenic UGP2 Axis Regulates Perfusion Recovery in Experimental Hind Limb Ischemia.

BACKGROUND: Despite being in an oxygen-rich environment, endothelial cells (ECs) use anaerobic glycolysis (Warburg effect) as the primary metabolic pathway for cellular energy needs. PFKFB (6-phosphofructo-2-kinase/fructose-2,6-biphosphatase)-3 regulates a critical enzymatic checkpoint in glycolysis and has been shown to induce angiogenesis. This study builds on our efforts to determine the metabolic regulation of ischemic angiogenesis and perfusion recovery in the ischemic muscle. METHODS: Hypoxia serum starvation (HSS) was used as an in vitro peripheral artery disease (PAD) model, and hind limb ischemia by femoral artery ligation and resection was used as a preclinical PAD model. RESULTS: Despite increasing PFKFB3-dependent glycolysis, HSS significantly decreased the angiogenic capacity of ischemic ECs. Interestingly, inhibiting PFKFB3 significantly induced the angiogenic capacity of HSS-ECs. Since ischemia induced a significant in PFKFB3 levels in hind limb ischemia muscle versus nonischemic, we wanted to determine whether glucose bioavailability (rather than PFKFB3 expression) in the ischemic muscle is a limiting factor behind impaired angiogenesis. However, treating the ischemic muscle with intramuscular delivery of D-glucose or L-glucose (osmolar control) showed no significant differences in the perfusion recovery, indicating that glucose bioavailability is not a limiting factor to induce ischemic angiogenesis in experimental PAD. Unexpectedly, we found that shRNA-mediated PFKFB3 inhibition in the ischemic muscle resulted in an increased perfusion recovery and higher vascular density compared with control shRNA (consistent with the increased angiogenic capacity of PFKFB3 silenced HSS-ECs). Based on these data, we hypothesized that inhibiting HSS-induced PFKFB3 expression/levels in ischemic ECs activates alternative metabolic pathways that revascularize the ischemic muscle in experimental PAD. A comprehensive glucose metabolic gene qPCR arrays in PFKFB3 silenced HSS-ECs, and PFKFB3-knock-down ischemic muscle versus respective controls identified UGP2 (uridine diphosphate-glucose pyrophosphorylase 2), a regulator of protein glycosylation and glycogen synthesis, is induced upon PFKFB3 inhibition in vitro and in vivo. Antibody-mediated inhibition of UGP2 in the ischemic muscle significantly impaired perfusion recovery versus IgG control. Mechanistically, supplementing uridine diphosphate-glucose, a metabolite of UGP2 activity, significantly induced HSS-EC angiogenic capacity in vitro and enhanced perfusion recovery in vivo by increasing protein glycosylation (but not glycogen synthesis). CONCLUSIONS: Our data present that inhibition of maladaptive PFKFB3-driven glycolysis in HSS-ECs is necessary to promote the UGP2-uridine diphosphate-glucose axis that enhances ischemic angiogenesis and perfusion recovery in experimental PAD.

The impact of spiritual care on the psychological health and quality of life of adults with heart failure: a systematic review of randomized trials.

Background: Heart failure (HF) brings not only physical pain but also psychological distress. This systematic review investigated the influence of spiritual care on the psychological well-being and quality of life in adults with HF. Methods: We conducted a systematic literature review following PRISMA guidelines, searching seven electronic databases for relevant randomized controlled studies without language or temporal restrictions. The studies were assessed for quality using the Cochrane Bias Risk tool. Results: A total of 13 studies (882 participants) were reviewed, investigating interventions such as religion, meditation, mental health, cognitive interventions, and spiritual support. Key factors influencing the effectiveness of spiritual care implementation included integration into routine care, respect for diversity, patient engagement, intervention quality, and alignment with patient beliefs. The majority of the studies indicated that spiritual care has a potentially beneficial impact on the mental health and quality of life of patients with HF. Conclusion: The findings provide valuable insights for healthcare professionals, highlighting the importance of adopting a spiritual care approach to healthcare for this population.

Effect of miR-98/IL-6/STAT3 on Autophagy and Apoptosis of Cardiac Stem Cells Under Hypoxic Conditions In vitro.

BACKGROUND: The heavy burden of cardiovascular diseases demands innovative therapeutic strategies dealing with cardiomyocyte loss. Cardiac Stem Cells (CSCs) are renewable cells in the myocardium with differentiation and endocrine functions. However, their functions are significantly inhibited in conditions of severe hypoxia or inflammation. The mechanism of hypoxia affecting CSCs is not clear. Interleukin-6 (IL-6) appears active in both hypoxic and inflammatory microenvironments. The aim of this study was to explore whether IL-6 is related to CSC apoptosis and autophagy under severe hypoxia. METHODS: In this study, rat CSCs were extracted by alternate digestion. The interaction of miR-98 and IL-6 mRNA was detected by the dual luciferase method, and qPCR was applied to confirm the effect of miR-98 on IL-6 expression. The effect of IL-6 on CSC apoptosis was measured by flow cytometry and the effect of IL-6 on CSC autophagy by transmission electron microscopy. The western blot method was applied to detect the effect of IL-6 on the expressions of proteins related to apoptosis and autophagy. ANOVA and Dunnett T3's test were employed in the statistical analysis. When p < 0.05, the difference was significant. RESULTS: Under severe hypoxia conditions, IL-6 increased CSC apoptosis and decreased p-STAT3 expression significantly. CSC apoptosis increased significantly after inhibition of the STAT3 signaling pathway under severe hypoxia. IL-6 could also significantly inhibit CSCs' autophagy and block their autophagy flow under severe hypoxic conditions. Meanwhile, it was confirmed that miR-98 had a binding site on IL-6 mRNA and miR-98 significantly inhibited IL-6 mRNA expression in CSCs under severe hypoxic conditions. CONCLUSION: miR-98/IL-6/STAT3 has been found to be involved in the regulation of CSCs' apoptosis and autophagy under severe hypoxic conditions and there might be a mutual linkage between CSCs' apoptosis and their autophagy.

A nomogram for predicting choledochal cyst with perforation.

BACKGROUND: Choledochal cyst with perforation (CC with perforation) rarely occurs, early diagnosis and timely treatment plan are crucial for the treatment of CC with perforation. This study aims to forecast the occurrence of CC with perforation. METHODS: All 1111 patients were conducted, who underwent surgery for choledochal cyst at our hospital from January 2011 to October 2022. We conducted univariate and multivariate logistic regression analysis to screen for independent predictive factors for predicting CC with perforation, upon which established a nomogram. The predictive performance of the nomogram was evaluated using receiver operating characteristic (ROC) curves, calibration plots, and decision curve analysis (DCA) curves. RESULTS: The age of children with choledochal cyst perforation is mainly concentrated between 1 and 3 years old. Logistic regression analysis indicates that age, alanine aminotransferase, glutamyl transpeptidase, C-reactive protein, vomiting, jaundice, abdominal distension, and diarrhea are associated with predicting the occurrence of choledochal cyst perforation. ROC curves, calibration plots, and DCA curve analysis curves demonstrate that the nomogram has great discriminative ability and calibration, as well as significant clinical utility. CONCLUSION: The age of CC with perforation is mainly concentrated between 1 and 3 years old. A nomogram for predicting the perforation of choledochal cyst was established.

TM-Score predicts immunotherapy efficacy and improves the performance of the machine learning prognostic model in gastric cancer.

Immunotherapy is becoming increasingly important, but the overall response rate is relatively low in the treatment of gastric cancer (GC). The application of tumor mutational burden (TMB) in predicting immunotherapy efficacy in GC patients is limited and controversial, emphasizing the importance of optimizing TMB-based patient selection. By combining TMB and major histocompatibility complex (MHC) related hub genes, we established a novel TM-Score. This score showed superior performance for immunotherapeutic selection (AUC = 0.808) compared to TMB, MSI status, and EBV status. Additionally, it predicted the prognosis of GC patients. Subsequently, a machine learning model adjusted by the TM-Score further improved the accuracy of survival prediction (AUC > 0.8). Meanwhile, we found that GC patients with low TM-Score had a higher mutation frequency, higher expression of HLA genes and immune checkpoint genes, and higher infiltration of CD8+ T cells, CD4+ helper T cells, and M1 macrophages. This suggests that TM-Score is significantly associated with tumor immunogenicity and tumor immune environment. Notably, based on the RNA-seq and scRNA-seq, it was found that AKAP5, a key component gene of TM-Score, is involved in anti-tumor immunity by promoting the infiltration of CD4+ T cells, NK cells, and myeloid cells. Additionally, siAKAP5 significantly reduced MHC-II mRNA expression in the GC cell line. In addition, our immunohistochemistry assays confirmed a positive correlation between AKAP5 and human leukocyte antigen (HLA) expression. Furthermore, AKAP5 levels were higher in patients with longer survival and those who responded to immunotherapy in GC, indicating its potential value in predicting prognosis and immunotherapy outcomes. In conclusion, TM-Score, as an optimization of TMB, is a more precise biomarker for predicting the immunotherapy efficacy of the GC population. Additionally, AKAP5 shows promise as a therapeutic target for GC.

A density map-based method for counting wheat ears.

Introduction: Field wheat ear counting is an important step in wheat yield estimation, and how to solve the problem of rapid and effective wheat ear counting in a field environment to ensure the stability of food supply and provide more reliable data support for agricultural management and policy making is a key concern in the current agricultural field. Methods: There are still some bottlenecks and challenges in solving the dense wheat counting problem with the currently available methods. To address these issues, we propose a new method based on the YOLACT framework that aims to improve the accuracy and efficiency of dense wheat counting. Replacing the pooling layer in the CBAM module with a GeM pooling layer, and then introducing the density map into the FPN, these improvements together make our method better able to cope with the challenges in dense scenarios. Results: Experiments show our model improves wheat ear counting performance in complex backgrounds. The improved attention mechanism reduces the RMSE from 1.75 to 1.57. Based on the improved CBAM, the R2 increases from 0.9615 to 0.9798 through pixel-level density estimation, the density map mechanism accurately discerns overlapping count targets, which can provide more granular information. Discussion: The findings demonstrate the practical potential of our framework for intelligent agriculture applications.

An Aerial Image Detection Algorithm Based on Improved YOLOv5.

To enhance aerial image detection in complex environments characterized by multiple small targets and mutual occlusion, we propose an aerial target detection algorithm based on an improved version of YOLOv5 in this paper. Firstly, we employ an improved Mosaic algorithm to address redundant boundaries arising from varying image scales and to augment the training sample size, thereby enhancing detection accuracy. Secondly, we integrate the constructed hybrid attention module into the backbone network to enhance the model's capability in extracting pertinent feature information. Subsequently, we incorporate feature fusion layer 7 and P2 fusion into the neck network, leading to a notable enhancement in the model's capability to detect small targets. Finally, we replace the original PAN + FPN network structure with the optimized BiFPN (Bidirectional Feature Pyramid Network) to enable the model to preserve deeper semantic information, thereby enhancing detection capabilities for dense objects. Experimental results indicate a substantial improvement in both the detection accuracy and speed of the enhanced algorithm compared to its original version. It is noteworthy that the enhanced algorithm exhibits a markedly improved detection performance for aerial images, particularly under real-time conditions.

Fault-scale crustal structure across the Dunhua-Mishan fault (Tanlu northern segment) constrained from teleseismic P-wave receiver functions.

The Dunhua-Mishan fault, located in the northern segment of the Tanlu fault zone, experienced multiple tectonic processes associated with the effects of the Pacific Plate subduction and the Indo-Asia collision. The high-resolution fault-scale structure is critical for understanding the fault evolution and potential fault damage. However, the well-defined deep structure of the Dunhua-Mishan fault is still unclear due to the lack of the dense seismic array. In this study, we construct a high-resolution P-wave receiver function imaging based on linear dense seismic array across the fault. Our results reveal the strong Moho depth variation across the Dunhua-Mishan fault zone. The slightly higher Vp/Vs ratio values within the fault zone indicate the presence of a small amount of mafic crust composition. Interestingly, the significant double positive Ps converted phases are observed within the fault zone, which may represent double Moho discontinuities. The double Moho structure may be related to multiple significant tectonic activities in the Tanlu northern segment. These newly observed structures provide new seismic constraints on the formation and evolution of the Tanlu fault zone and probably reflect that the lithospheric structure of the Dunhua-Mishan fault has been modified by a series of tectonic processes.

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.