Application and utility of boosting machine learning model based on laboratory test in the differential diagnosis of non-COVID-19 pneumonia and COVID-19.

BACKGROUND: Non-Coronavirus disease 2019 (COVID-19) pneumonia and COVID-19 have similar clinical features but last for different periods, and consequently, require different treatment protocols. Therefore, they must be differentially diagnosed. This study uses artificial intelligence (AI) to classify the two forms of pneumonia using mainly laboratory test data. METHODS: Various AI models are applied, including boosting models known for deftly solving classification problems. In addition, important features that affect the classification prediction performance are identified using the feature importance technique and SHapley Additive exPlanations method. Despite the data imbalance, the developed model exhibits robust performance. RESULTS: eXtreme gradient boosting, category boosting, and light gradient boosted machine yield an area under the receiver operating characteristic of 0.99 or more, accuracy of 0.96-0.97, and F1-score of 0.96-0.97. In addition, D-dimer, eosinophil, glucose, aspartate aminotransferase, and basophil, which are rather nonspecific laboratory test results, are demonstrated to be important features in differentiating the two disease groups. CONCLUSIONS: The boosting model, which excels in producing classification models using categorical data, excels in developing classification models using linear numerical data, such as laboratory tests. Finally, the proposed model can be applied in various fields to solve classification problems.

Deep learning approach for early prediction of COVID-19 mortality using chest X-ray and electronic health records.

BACKGROUND: An artificial-intelligence (AI) model for predicting the prognosis or mortality of coronavirus disease 2019 (COVID-19) patients will allow efficient allocation of limited medical resources. We developed an early mortality prediction ensemble model for COVID-19 using AI models with initial chest X-ray and electronic health record (EHR) data. RESULTS: We used convolutional neural network (CNN) models (Inception-ResNet-V2 and EfficientNet) for chest X-ray analysis and multilayer perceptron (MLP), Extreme Gradient Boosting (XGBoost), and random forest (RF) models for EHR data analysis. The Gradient-weighted Class Activation Mapping and Shapley Additive Explanations (SHAP) methods were used to determine the effects of these features on COVID-19. We developed an ensemble model (Area under the receiver operating characteristic curve of 0.8698) using a soft voting method with weight differences for CNN, XGBoost, MLP, and RF models. To resolve the data imbalance, we conducted F1-score optimization by adjusting the cutoff values to optimize the model performance (F1 score of 0.77). CONCLUSIONS: Our study is meaningful in that we developed an early mortality prediction model using only the initial chest X-ray and EHR data of COVID-19 patients. Early prediction of the clinical courses of patients is helpful for not only treatment but also bed management. Our results confirmed the performance improvement of the ensemble model achieved by combining AI models. Through the SHAP method, laboratory tests that indicate the factors affecting COVID-19 mortality were discovered, highlighting the importance of these tests in managing COVID-19 patients.

Comparison of humoral immunogenicity in solid organ transplant recipients after third-dose mRNA vaccine with homologous or heterologous schedules: An observational study.

BACKGROUND: Solid organ transplant recipients (SOTRs) are susceptible to severe coronavirus disease 2019 (COVID-19); however, immunogenicity studies of the Omicron variants per vaccination schedules are still lacking. We examined humoral immunogenicity following third-dose mRNA vaccine administration in Korean SOTRs who received primary COVID-19 vaccine series on homologous or heterologous schedules. METHODS: We recruited SOTRs at Severance Hospital from October 27, 2021, to March 31, 2022. Blood samples were collected between 14 days and 5 months after the second and third mRNA vaccine (BNT162b2 or mRNA-1273) doses. SARS-CoV-2 anti-spike IgG titer was analyzed. The neutralization inhibition rate was analyzed using the surrogate neutralization assay for the wild-type, Delta, and Omicron variants. RESULTS: No significant differences existed in the SARS-CoV-2 anti-spike IgG positivity rate between the homologous BNT162b2/BNT162b2/BNT162b2 (85%) and other heterologous groups (83% of ChAdOx1/ChAdOx1/BNT162b2, 90% of ChAdOx1/ChAdOx1/mRNA-1273, and 78% of ChAdOx1/BNT162b2/BNT162b2). No significant difference existed in the neutralization inhibition rates between the four groups for wild-type, Delta, and Omicron variants. Median neutralization inhibition rates against the Omicron variant (2-5%) were significantly lower than those against the wild-type (87-97%) and Delta (55-89%) variants (P < 0.001). CONCLUSIONS: Regardless of the schedule, the neutralization inhibition rate against the Omicron variant was poor; therefore, additional preventive measures are required in such high-risk populations.

Development of Machine-Learning Model to Predict COVID-19 Mortality: Application of Ensemble Model and Regarding Feature Impacts.

This study was designed to develop machine-learning models to predict COVID-19 mortality and identify its key features based on clinical characteristics and laboratory tests. For this, deep-learning (DL) and machine-learning (ML) models were developed using receiver operating characteristic (ROC) area under the curve (AUC) and F1 score optimization of 87 parameters. Of the two, the DL model exhibited better performance (AUC 0.8721, accuracy 0.84, and F1 score 0.76). However, we also blended DL with ML, and the ensemble model performed the best (AUC 0.8811, accuracy 0.85, and F1 score 0.77). The DL model is generally unable to extract feature importance; however, we succeeded by using the Shapley Additive exPlanations method for each model. This study demonstrated both the applicability of DL and ML models for classifying COVID-19 mortality using hospital-structured data and that the ensemble model had the best predictive ability.

Matched Versus Mixed COVID-19 Vaccinations in Korean Solid Organ Transplant Recipients: An Observational Study.

BACKGROUND: Solid organ transplant recipients (SOTRs) are vulnerable to severe coronavirus disease 2019 (COVID-19) and exhibit poor antibody responses to COVID-19 vaccines. Herein, we compared the humoral immunogenicity of a mixed vaccine (ChAdOx1 nCoV-19 [ChAd]/BNT162b2 [BNT]) with that of conventional matched vaccines (mRNA, adenoviral vector [AdV-Vec]) in SOTRs. METHODS: Serum samples were collected at Severance Hospital (Seoul, Korea) between September and October 2021 (14 d-5 mo after COVID-19 vaccination; V2). The severe acute respiratory syndrome coronavirus 2 antispike IgG titer (BAU/mL; ELISA) and neutralization inhibition (percentage; neutralization assay) were compared between vaccination groups overall and stratified by V2 (poststudy vaccination visit) timing. RESULTS: Of the 464 participants, 143 (31%) received mRNA vaccines, 170 (37%) received AdV-Vec vaccines, and 151 (33%) received mixed vaccines (all ChAd/BNT). The geometric mean titer for the ChAd/BNT group was 3.2-fold higher than that of the AdV-Vec group (geometric mean ratio, 3.2; confidence interval, 1.9-5.4) but lower than that of the mRNA group (geometric mean ratio, 0.4; confidence interval, 0.2-0.7). Neutralization inhibition in the ChAd/BNT group was 32%, which was higher than that in the AdV-Vec group (21%; P < 0.001) but lower than that in the mRNA group (55%; P = 0.02). There was no difference in geometric mean titer by V2 timing (ChAd/BNT, 45 versus 31, days 14-60; mRNA, 28 versus 15, days 61-150). CONCLUSIONS: The ChAd/BNT group showed higher humoral immunogenicity than the AdV-Vec group, with similar immunogenicity to the mRNA vaccine. Nevertheless, immunogenicity following the primary vaccination series was poor in all vaccine groups, supporting the justification for booster vaccination in SOTRs.

Understanding the History, Theoretical Perspectives, and Influence of Mindfulness: Building Sustainability Through Mindfulness in Education

This study presents an understanding of the influence of mindfulness in the educational field at student, teacher, and principal levels. The review includes a definition of mindfulness, value of mindfulness, and a new perspective of education based on physical and mental health. The prevailing COVID-19 pandemic subjects school academics to unexpected situations. I present in this paper the history, practice, and influence of mindfulness in both Eastern and Western traditions as a proven stress reliever. I also evaluate its impact on the training that students, teachers, and principals undergo, and are thus likely to provide, throughout their schools.

The feasibility of organ transplantation during the COVID-19 outbreak: experiences from South Korea.

Background: The coronavirus disease 2019 (COVID-19) has forced healthcare systems to reduce transplant activities in order to preserve resources and minimize the risk of nosocomial transmission. Although transplantation societies around the world have proposed interim recommendations, little is known about the safety of transplant surgery under pandemic conditions and how transplant medicine should move forward after the peak of the pandemic. Methods: We describe our experiences regarding the continuation of living and deceased donor transplantation under infection control measures during the COVID-19 outbreak in South Korea. We reviewed consecutive liver and kidney transplantations at Severance Hospital and analyzed national transplantation activities in South Korea. Results: Transplantation activities with living and deceased donors remained stable during the COVID-19 outbreak compared to the same period in 2019. We performed 94 transplantations (58 kidney, 35 liver, and 1 simultaneous liver-kidney) during the COVID-19 outbreak. Twenty-five patients underwent desensitization therapy prior to transplant (nine ABO-incompatible kidney, eight human leukocyte antigen-incompatible kidney, and eight ABO-incompatible liver). No transplant recipients in our center contracted COVID-19. In South Korea, national transplant activities with living and deceased donors remained stable in 2020 compared to 2019. Conclusions: Organ transplantation during pandemics appears to be feasible with appropriate infection prevention measures.

Methylene Blue-Based Nano and Microparticles: Fabrication and Applications in Photodynamic Therapy.

Methylene blue (MB) has been used in the textile industry since it was first extracted by the German chemist Heinrich Caro. Its pharmacological properties have also been applied toward the treatment of certain diseases such as methemoglobinemia, ifosfamide-induced encephalopathy, and thyroid conditions requiring surgery. Recently, the utilization of MB as a safe photosensitizer in photodynamic therapy (PDT) has received attention. Recent findings demonstrate that photoactivated MB exhibits not only anticancer activity but also antibacterial activity both in vitro and in vivo. However, due to the hydrophilic nature of MB, it is difficult to create MB-embedded nano- or microparticles capable of increasing the clinical efficacy of the PDT. This review aims to summarize fabrication techniques for MB-embedded nano and microparticles and to provide both in vitro and in vivo examples of MB-mediated PDT, thereby offering a future perspective on improving this promising clinical treatment modality. We also address examples of MB-mediated PDT in both cancer and infection treatments. Both in-vitro and in-vivo studies are summarized here to document recent trends in utilizing MB as an effective photosensitizer in PDT. Lastly, we discuss how developing efficient MB-carrying nano- and microparticle platforms would be able to increase the benefits of PDT.

The COVID-19 and Stock Return Volatility: Evidence from South Korea

This study examines the impact of the number of coronavirus cases on regime-switching in stock return volatility. This study documents the empirical evidence that the COVID-19 cases had an asymmetric effect on the regime of stock return volatility. When the stock return is in the low volatility regime, the probability of switching to the high volatility regime in the next trading day increases as the number of cumulative cases increases. In contrast, in the high volatility regime, the effect of cumulative cases on the transition probability is not statistically significant. This study also documents the evidence that the government measures against the pandemic contribute to promoting the high volatility regime of the KOSPI during the pandemic. Besides, this study projects future stock prices through the Monte Carlo simulation based on the estimated parameters and the predicted number of the COVID-19 new cases. Under a scenario where the number of new cases rapidly increases, stock price indices in Korea are expected to be in a downward trend over the next three months. On the other hand, under the moderate scenario and the best scenario, the stock indices are likely to continue to rise.

An Earth-mass Planet in a Time of Covid-19: KMT-2020-BLG-0414Lb (preprint)

We report the discovery of KMT-2020-BLG-0414Lb, with a planet-to-host mass ratio $q_2 = 0.9$--$1.2 \times 10^{-5} = 3$--$4~q_{\oplus}$ at $1\sigma$, which is the lowest mass-ratio microlensing planet to date. Together with two other recent discoveries ($4 \lesssim q/q_\oplus \lesssim 6$), it fills out the previous empty sector at the bottom of the triangular $(\log s, \log q)$ diagram, where $s$ is the planet-host separation in units of the angular Einstein radius $\theta_{\rm E}$. Hence, these discoveries call into question the existence, or at least the strength, of the break in the mass-ratio function that was previously suggested to account for the paucity of very low-$q$ planets. Due to the extreme magnification of the event, $A_{\rm max}\sim 1450$ for the underlying single-lens event, its light curve revealed a second companion with $q_3 \sim 0.05$ and $|\log s_3| \sim 1$, i.e., a factor $\sim 10$ closer to or farther from the host in projection. The measurements of the microlens parallax $\pi_{\rm E}$ and the angular Einstein radius $\theta_{\rm E}$ allow estimates of the host, planet, and second companion masses, $(M_1, M_2, M_3) \sim (0.3M_{\odot}, 1.0M_{\oplus}, 17M_{J})$, the planet and second companion projected separations, $(a_{\perp,2}, a_{\perp,3}) \sim (1.5, 0.15~{\rm or}~15)$~au, and system distance $D_{\rm L} \sim 1$ kpc. The lens could account for most or all of the blended light ($I \sim 19.3$) and so can be studied immediately with high-resolution photometric and spectroscopic observations that can further clarify the nature of the system. The planet was found as part of a new program of high-cadence follow-up observations of high-magnification events. The detection of this planet, despite the considerable difficulties imposed by Covid-19 (two KMT sites and OGLE were shut down), illustrates the potential utility of this program.

CIMI: Classify and Itemize Medical Image System for PFT Big Data Based on Deep Learning

The value of pulmonary function test (PFT) data is increasing due to the advent of the Coronavirus Infectious Disease 19 (COVID-19) and increased respiratory disease. However, these PFT data cannot be directly used in clinical studies, because PFT results are stored in raw image files. In this study, the classification and itemization medical image (CIMI) system generates valuable data from raw PFT images by automatically classifying various PFT results, extracting texts, and storing them in the PFT database and Excel files. The deep-learning-based optical character recognition (OCR) technology was mainly used in CIMI to classify and itemize PFT images in St. Mary’s Hospital. CIMI classified seven types and itemized 913,059 texts from 14,720 PFT image sheets, which cannot be done by humans. The number, type, and location of texts that can be extracted by PFT type are all different, but CIMI solves this issue by classifying the PFT image sheets by type, allowing researchers to analyze the data. To demonstrate the superiority of CIMI, the validation results of CIMI were compared to the results of the other four algorithms. A total of 70 randomly selected sheets (ten sheets from each type) and 33,550 texts were used for the validation. The accuracy of CIMI was 95%, which was the highest accuracy among the other four algorithms.

Early Clinical and CT Manifestations of Coronavirus Disease 2019 (COVID-19) Pneumonia.

OBJECTIVE. The purpose of this study was to investigate early clinical and CT manifestations of coronavirus disease (COVID-19) pneumonia. MATERIALS AND METHODS. Patients with COVID-19 pneumonia confirmed by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) nucleic acid test (reverse transcription-polymerase chain reaction) were enrolled in this retrospective study. The clinical manifestations, laboratory results, and CT findings were evaluated. RESULTS. One hundred eight patients (38 men, 70 women; age range, 21-90 years) were included in the study. The clinical manifestations were fever in 94 of 108 (87%) patients, dry cough in 65 (60%), and fatigue in 42 (39%). The laboratory results were normal WBC count in 97 (90%) patients and normal or reduced lymphocyte count in 65 (60%). High-sensitivity C-reactive protein level was elevated in 107 (99%) patients. The distribution of involved lobes was one lobe in 38 (35%) patients, two or three lobes in 24 (22%), and four or five lobes in 46 (43%). The major involvement was peripheral (97 patients [90%]), and the common lesion shape was patchy (93 patients [86%]). Sixty-five (60%) patients had ground-glass opacity (GGO), and 44 (41%) had GGO with consolidation. The size of lesions varied from smaller than 1 cm (10 patients [9%]) to larger than 3 cm (56 patients [52%]). Vascular thickening (86 patients [80%]), crazy paving pattern (43 patients [40%]), air bronchogram sign (52 patients [48%]), and halo sign (69 [64%]) were also observed in this study. CONCLUSION. The early clinical and laboratory findings of COVID-19 pneumonia are low to midgrade fever, dry cough, and fatigue with normal WBC count, reduced lymphocyte count, and elevated high-sensitivity C-reactive protein level. The early CT findings are patchy GGO with or without consolidation involving multiple lobes, mainly in the peripheral zone, accompanied by halo sign, vascular thickening, crazy paving pattern, or air bronchogram sign.