Background: Effectively and efficiently diagnosing patients who have COVID-19 with the accurate clinical type of the disease is essential to achieve optimal outcomes for the patients as well as to reduce the risk of overloading the health care system. Currently, severe and nonsevere COVID-19 types are differentiated by only a few features, which do not comprehensively characterize the complicated pathological, physiological, and immunological responses to SARS-CoV-2 infection in the different disease types. In addition, these type-defining features may not be readily testable at the time of diagnosis. Objective: In this study, we aimed to use a machine learning approach to understand COVID-19 more comprehensively, accurately differentiate severe and nonsevere COVID-19 clinical types based on multiple medical features, and provide reliable predictions of the clinical type of the disease. Methods: For this study, we recruited 214 confirmed patients with nonsevere COVID-19 and 148 patients with severe COVID-19. The clinical characteristics (26 features) and laboratory test results (26 features) upon admission were acquired as two input modalities. Exploratory analyses demonstrated that these features differed substantially between two clinical types. Machine learning random forest models based on all the features in each modality as well as on the top 5 features in each modality combined were developed and validated to differentiate COVID-19 clinical types. Results: Using clinical and laboratory results independently as input, the random forest models achieved >90% and >95% predictive accuracy, respectively. The importance scores of the input features were further evaluated, and the top 5 features from each modality were identified (age, hypertension, cardiovascular disease, gender, and diabetes for the clinical features modality, and dimerized plasmin fragment D, high sensitivity troponin I, absolute neutrophil count, interleukin 6, and lactate dehydrogenase for the laboratory testing modality, in descending order). Using these top 10 multimodal features as the only input instead of all 52 features combined, the random forest model was able to achieve 97% predictive accuracy. Conclusions: Our findings shed light on how the human body reacts to SARS-CoV-2 infection as a unit and provide insights on effectively evaluating the disease severity of patients with COVID-19 based on more common medical features when gold standard features are not available. We suggest that clinical information can be used as an initial screening tool for self-evaluation and triage, while laboratory test results should be applied when accuracy is the priority.

Background: Effectively identifying patients with COVID-19 using nonpolymerase chain reaction biomedical data is critical for achieving optimal clinical outcomes. Currently, there is a lack of comprehensive understanding in various biomedical features and appropriate analytical approaches for enabling the early detection and effective diagnosis of patients with COVID-19. Objective: We aimed to combine low-dimensional clinical and lab testing data, as well as high-dimensional computed tomography (CT) imaging data, to accurately differentiate between healthy individuals, patients with COVID-19, and patients with non-COVID viral pneumonia, especially at the early stage of infection. Methods: In this study, we recruited 214 patients with nonsevere COVID-19, 148 patients with severe COVID-19, 198 noninfected healthy participants, and 129 patients with non-COVID viral pneumonia. The participants’ clinical information (ie, 23 features), lab testing results (ie, 10 features), and CT scans upon admission were acquired and used as 3 input feature modalities. To enable the late fusion of multimodal features, we constructed a deep learning model to extract a 10-feature high-level representation of CT scans. We then developed 3 machine learning models (ie, k-nearest neighbor, random forest, and support vector machine models) based on the combined 43 features from all 3 modalities to differentiate between the following 4 classes: nonsevere, severe, healthy, and viral pneumonia. Results: Multimodal features provided substantial performance gain from the use of any single feature modality. All 3 machine learning models had high overall prediction accuracy (95.4%-97.7%) and high class-specific prediction accuracy (90.6%-99.9%). Conclusions: Compared to the existing binary classification benchmarks that are often focused on single-feature modality, this study’s hybrid deep learning-machine learning framework provided a novel and effective breakthrough for clinical applications. Our findings, which come from a relatively large sample size, and analytical workflow will supplement and assist with clinical decision support for current COVID-19 diagnostic methods and other clinical applications with high-dimensional multimodal biomedical features.

Background: Stroke is the second leading cause of death and the third leading cause of disability worldwide, with ischemic stroke (IS) being the most prevalent. A substantial number of irreversible brain cell death occur in the short term, leading to impairment or death in IS. Limiting the loss of brain cells is the primary therapy target and a significant clinical issue for IS therapy. Our study aims to establish the gender specificity pattern from immune cell infiltration and four kinds of cell-death perspectives to improve IS diagnosis and therapy.Methods: Combining and standardizing two IS datasets (GSE16561 and GSE22255) from the GEO database, we used the CIBERSORT algorithm to investigate and compare the immune cell infiltration in different groups and genders. Then, ferroptosis-related differently expressed genes (FRDEGs), pyroptosis-related DEGs (PRDEGs), anoikis-related DEGs (ARDEGs), and cuproptosis-related DEGs (CRDEGs) between the IS patient group and the healthy control group were identified in men and women, respectively. Machine learning (ML) was finally used to generate the disease prediction model for cell death-related DEGs (CDRDEGs) and to screen biomarkers related to cell death involved in IS.Results: Significant changes were observed in 4 types of immune cells in male IS patients and 10 types in female IS patients compared with healthy controls. In total, 10 FRDEGs, 11 PRDEGs, 3 ARDEGs, and 1 CRDEG were present in male IS patients, while 6 FRDEGs, 16 PRDEGs, 4 ARDEGs, and 1 CRDEG existed in female IS patients. ML techniques indicated that the best diagnostic model for both male and female patients was the support vector machine (SVM) for CDRDEG genes. SVM’s feature importance analysis demonstrated that SLC2A3, MMP9, C5AR1, ACSL1, and NLRP3 were the top five feature-important CDRDEGs in male IS patients. Meanwhile, the PDK4, SCL40A1, FAR1, CD163, and CD96 displayed their overwhelming influence on female IS patients.Conclusion: These findings contribute to a better knowledge of immune cell infiltration and their corresponding molecular mechanisms of cell death and offer distinct clinically relevant biological targets for IS patients of different genders.

Recently, ultra-small platinum nanoparticles (USPtNs) have been found that can kill cancer cells by leaching Pt ions into acidic organelles, such as cell endosomes or lysosomes. Unfortunately, tumor-specific accumulation is difficult to achieve with such platinum nanodrugs of less than 5 nm due to their short half-life in vivo and broad range of toxicity to normal tissues. Programmable multi-drug release for combinational chemotherapy by hierarchical nanostructures provides a promising solution for cancer-targeted therapy. Herein, we demonstrated a pH/redox dual stimuli-responsive clustered nanoparticle as a vehicle for simultaneously delivering USPtNs and gemcitabine (GEM) to treat non-small-cell lung cancer. The clustered nanoparticle (denoted as GP-NA) was composed of disulfide-bond-containing GEM-grafted copolymers (PEG-b-P(LL-g-GEM)), pH-sensitive polypeptides (OAPI), and USPtNs. Such a hybrid nanosystem completes multiple tasks inside cancer cells, which include the generation of cytotoxic Pt ions in response to lysosomal acidic environments and the subsequent release of GEM in cytoplasmic reduction environments. Compared with non-acid-sensitive nanoparticles or free drugs, GP-NA exhibited cumulative and enhanced anti-tumor efficacy in vivo, which may be attributed to the simultaneous inhibition of ribonucleotide reductase and DNA replication in nuclei by the GEM and Pt ions. Together, our work provides a promising strategy in the co-delivery of USPtNs and GEM for precision cancer chemotherapy.