Machine Learning-Based Prediction of COVID-19 Severity and Progression to Critical Illness Using CT Imaging and Clinical Data

Purkayastha Subhanik,Xiao Yanhe,Jiao Zhicheng,Thepumnoeysuk Rujapa,Halsey Kasey,Wu Jing,Tran Thi My Linh,Hsieh Ben,Choi Ji Whae,Wang Dongcui,Vallières Martin,Wang Robin,Collins Scott,Feng Xue,Feldman 대한영상의학회 2021 Korean Journal of Radiology Vol.22 No.7

Objective: To develop a machine learning (ML) pipeline based on radiomics to predict Coronavirus Disease 2019 (COVID-19) severity and the future deterioration to critical illness using CT and clinical variables. Materials and Methods: Clinical data were collected from 981 patients from a multi-institutional international cohort with real-time polymerase chain reaction-confirmed COVID-19. Radiomics features were extracted from chest CT of the patients. The data of the cohort were randomly divided into training, validation, and test sets using a 7:1:2 ratio. A ML pipeline consisting of a model to predict severity and time-to-event model to predict progression to critical illness were trained on radiomics features and clinical variables. The receiver operating characteristic area under the curve (ROC-AUC), concordance index (C-index), and time-dependent ROC-AUC were calculated to determine model performance, which was compared with consensus CT severity scores obtained by visual interpretation by radiologists. Results: Among 981 patients with confirmed COVID-19, 274 patients developed critical illness. Radiomics features and clinical variables resulted in the best performance for the prediction of disease severity with a highest test ROC-AUC of 0.76 compared with 0.70 (0.76 vs. 0.70, p = 0.023) for visual CT severity score and clinical variables. The progression prediction model achieved a test C-index of 0.868 when it was based on the combination of CT radiomics and clinical variables compared with 0.767 when based on CT radiomics features alone (p < 0.001), 0.847 when based on clinical variables alone (p = 0.110), and 0.860 when based on the combination of visual CT severity scores and clinical variables (p = 0.549). Furthermore, the model based on the combination of CT radiomics and clinical variables achieved time-dependent ROC-AUCs of 0.897, 0.933, and 0.927 for the prediction of progression risks at 3, 5 and 7 days, respectively. Conclusion: CT radiomics features combined with clinical variables were predictive of COVID-19 severity and progression to critical illness with fairly high accuracy.

In-hospital mortality in the emergency department: clinical and etiological differences between early and late deaths among patients awaiting admission

Gabriele Valli,Elisabetta Galati,Francesca De Marco,Chiara Bucci,Paolo Fratini,Elisa Cennamo,Carlo Ancona,Nicola Volpe,Maria Pia Ruggieri 대한응급의학회 2021 Clinical and Experimental Emergency Medicine Vol.8 No.4

Objective Given that there are no studies on diseases that occur by waiting for hospitalization, we aimed to evaluate the main causes of death in the emergency room (ER) and their relationship with overcrowding. Methods Patients who died in the ER in the past 2 years (pediatrics and trauma victims excluded) were divided into two groups: patients who died within 6 hours of arrival (emergency department [ED] group) and patients who died later (LD group). We compared the causes of death, total vital signs, diagnostic tests performed, and therapy between the groups. We assessed for possible correlation between the number of monthly deaths per group and four variables of overcrowding: number of patients treated per month, waiting time before medical visit (W-Time), mean intervention time (I-Time), and number of patients admitted to the ward per month (NPA). Results During the two years, 175 patients had died in our ER (52% in ED group and 48% in LD group). The total time spent in the ER was, respectively, 2.9±0.2 hours for ED group and 17.9± 1.5 hours for LD group. The more frequent cause of death was cardiovascular syndrome (30%) in ED group and sepsis (27%) and acute respiratory failure (27%) in LD group. Positive correlations between number of monthly deaths and W-Time (R2 0.51, P< 0.0001), and NPA (R2 0.37, P

[ANALYSIS] 고양이 코로나바이러스: 사이프러스에 신종 변이가 퍼지고 있다.

핫산 밸리(Hassan Vally) 유니콘웨일 2024 BBC science Vol.2024 No.3

Halo, Reversed Halo, or Both? Atypical Computed Tomography Manifestations of Coronavirus Disease (COVID-19) Pneumonia: The “Double Halo Sign”

Antonio Poerio,Matilde Sartoni,Giammichele Lazzari,Michele Valli,Miria Morsiani,Maurizio Zompatori 대한영상의학회 2020 Korean Journal of Radiology Vol.21 No.10

The epidemic of 2019 novel coronavirus, later named as coronavirus disease (COVID-19), began in Wuhan, China in December 2019 and has spread rapidly worldwide. Early diagnosis is crucial for the management of the patients with COVID-19, but the gold standard diagnostic test for this infection, the reverse transcriptase polymerase chain reaction, has a low sensitivity and an increased turnaround time. In this scenario, chest computed tomography (CT) could play a key role for an early diagnosis of COVID-19 pneumonia. Here, we have reported a confirmed case of COVID-19 with an atypical CT presentation showing a “double halo sign,” which we believe represents the pathological spectrum of this viral pneumonia.

Impact of feature proportion on matching performance of multi-biometric systems

Wencheng Yang,Song Wang,Guanglou Zheng,Craig Valli 한국통신학회 2019 ICT Express Vol.5 No.1

Biometrics as a tool for information security has been used in various applications. Feature-level fusion is widely used in the design of multi-biometric systems due to its advantages in increasing recognition accuracy and security. However, most existing multi-biometric systems that use feature-level fusion assign each biometric trait an equal proportion when combining features from multiple sources. For example, multi-biometric systems with two biometric traits commonly adopt a 50–50 feature proportion setting, which means that fused feature data contains half elements from each biometric modality. In this paper, we investigate the impact of feature proportion on the matching performance of multi-biometric systems. By using a fingerprint and face based multi-biometric system that applies feature-level fusion, we employ a random projection based transformation and a proportion weight factor. By adjusting this weight factor, we show that allocating unequal proportions to features from different biometric traits yields different matching performance. Our experimental results indicate that optimal performance, achieved with unequal feature proportions, could be better than the performance obtained with the commonly used 50–50 feature proportion. Therefore, the impact of feature proportion, which has been ignored by most existing work, should be taken into account and more study is required as to how to make feature proportion allocation benefit the performance of multi-biometric systems.

Characteristics of Zonda wind in South American Andes

Loredo-Souza, Acir M.,Wittwer, Adrian R.,Castro, Hugo G.,Vallis, Matthew B. Techno-Press 2017 Wind and Structures, An International Journal (WAS Vol.24 No.6

This paper discusses some features and conditions that characterize the Zonda wind, focusing particularly on the implications for wind engineering applications. This kind of wind, typical of mountainous regions, is far from being adequately characterized for computational simulations and proper modeling in experimental facilities such as boundary layer wind tunnels. The objective of this article is to report the research works that are being developed on this kind of wind, describing the main obtained results, and also to establish some general guidelines for the proper analysis of the Zonda in the wind engineering context. A classification for the Zonda wind is indicated and different cases of structural and environmental effects are described. Available meteorological data is analyzed from the wind engineering point of view to obtain the Zonda wind gust factors, as well as basic wind speeds relevant for structural design. Some considerations and possible directions for the Zonda wind-tunnel and computational modeling are provided. Gust factor values larger than those used for open terrain were obtained, nevertheless, the basic wind speed values obtained are similar to values presented by the Argentinian Wind Code for three-second gust, principally at Mendoza airport.

Characteristics of Zonda wind in South American Andes

Acir M. Loredo-Souza,Adrian R. Wittwer,Hugo G. Castro,Matthew B.Vallis 한국풍공학회 2017 Wind and Structures, An International Journal (WAS Vol.24 No.6

This paper discusses some features and conditions that characterize the Zonda wind, focusing particularly on the implications for wind engineering applications. This kind of wind, typical of mountainous regions, is far from being adequately characterized for computational simulations and proper modeling in experimental facilities such as boundary layer wind tunnels. The objective of this article is to report the research works that are being developed on this kind of wind, describing the main obtained results, and also to establish some general guidelines for the proper analysis of the Zonda in the wind engineering context. A classification for the Zonda wind is indicated and different cases of structural and environmental effects are described. Available meteorological data is analyzed from the wind engineering point of view to obtain the Zonda wind gust factors, as well as basic wind speeds relevant for structural design. Some considerations and possible directions for the Zonda wind-tunnel and computational modeling are provided. Gust factor values larger than those used for open terrain were obtained, nevertheless, the basic wind speed values obtained are similar to values presented by the Argentinian Wind Code for three-second gust, principally at Mendoza airport.