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

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.

Fructose and metabolic diseases: too much to be good / 中华医学杂志(英文版)

Excessive consumption of fructose, the sweetest of all naturally occurring carbohydrates, has been linked to worldwide epidemics of metabolic diseases in humans, and it is considered an independent risk factor for cardiovascular diseases. We provide an overview about the features of fructose metabolism, as well as potential mechanisms by which excessive fructose intake is associated with the pathogenesis of metabolic diseases both in humans and rodents. To accomplish this aim, we focus on illuminating the cellular and molecular mechanisms of fructose metabolism as well as its signaling effects on metabolic and cardiovascular homeostasis in health and disease, highlighting the role of carbohydrate-responsive element-binding protein in regulating fructose metabolism.

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

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.

Application of miniSTR Loci and Its Detection System for Degraded Materials in Forensic Medicine / 法医学杂志

OBJECTIVES@#To establish multiplex system of 16 miniSTR loci, and explore its application value for the degraded materials in forensic medicine.@*METHODS@#The multiplex system of 16 miniSTR loci was established using a six-dye fluorescence labeling technology and its application value in forensic medicine was assessed.@*RESULTS@#A six-dye fluorescence labeling miniSTR amplification kit was developed, which enabled 15 autosomal STR loci, Amelogenin locus and DYS391 to be typed simultaneously. This method showed good specificity and could provide stable and accurate typing results with a sensitivity of 50 pg. This system also provided a good test result for the normal biological sample of actual cases.@*CONCLUSIONS@#The multiplex system of 16 miniSTR loci has application value for degraded and trace materials with the advantages of high sensitivity and database compatibility, which can be used for forensic casework.

Prospective-triggered high-pitch spiral versus sequential dual-source CT coronary angiography: comparison of image quality and radiation dose

Background: prospectively electrocardiography [ECG]-triggered high-pitch spiral coronary computed tomography angiography [CCTA] is a unique scan mode for dual -source CT [DSCT]. Our reports aim to compare image quality and radiation dose of CCTA using high-pitch spiral or sequential acquisition mode in patients with low and stable heart rates

Materials and Methods: patients with low and stable heart rates [HR] [HR >/= 70 beats per minute [bpm]; heart rate variability [HRV] < 10 bpm] were randomly assigned to high-pitch spiral mode [group A; n = 80] or sequential acquisition mode [group B; n = 80]. Image quality scores, image noise, effective radiation dose and influencing factors on image quality were assessed

Results: mean image quality scores were 1.51 +/- 0.32 and 1.70 +/- 0.38 for groups A and B [P < 0.05], respectively. Image noises of the two groups were 19.05 +/- 4.70 Hu and 27.21 +/- 8.88 Hu [P < 0.05]. Contrast media cost in-group A was lower than group B [P < 0.05]. No statistical difference was found in the rate of diagnostic patients between the two groups [P = 0.416]. The estimated radiation dose of group A was 26.0% reduced compared with group B [0.74 +/- 0.34 mSv vs. 1.00 +/- 0.48 mSv, P < 0.05]

Conclusion: in patients with regular and low heart rates, the prospectively high-pitch spiral acquisition mode can reduce radiation dose and contrast media cost while maintaining image quality compared with the prospectively sequential mode

Comparison of six irradiation techniques for delivering hypofractionated whole-breast radiotherapy with a simultaneous integrated boost after breast-conserving surgery

Background: to compare the following techniques for hypofractionated whole-breast irradiation [WBI] with simultaneous integrated boost [SIB] after breast-conserving surgery [BCS]: three-dimensional conformal radiation therapy plus electron boost [3DCRT-EB], intensity-modulated radiation therapy [IMRT] plus EB [IMRT-EB], field-in-field IMRT plus EB [FIF-IMRT-EB], FIF-IMRT plus IMRT boost [FIF-IMRT-IB], IMRT plus IMRT boost [IMRT-IB], and volumetric-modulated arc therapy [VMAT] plus VMAT boost [VMAT-VB]

Materials and Methods: twenty patients with left breast cancer were enrolled. The prescribed dose was 40.05 Gy in 15 fractions to the whole breast and an SIB to the tumor bed of 3.2 Gy/fraction [total, 48 Gy]. Target-volume coverage, dose-conformity index, homogeneity index [HI], doses to organs at risk [OAR], and costs were compared

Results: FIF-IMRT-EB performed the best, while FIF-IMRT-IB, IMRT-IB, and VMAT-VB performed the worst. The mean dose to the planning target volume for breast evaluation [PTV Eval-breast] was significantly lower for IMRT-EB and FIF-IMRT-EB than for the other plans. For both PTV Eval-breast and PTV Eval-boost, VMAT-VB had the lowest target-volume coverage for 95% of the prescription dose and the highest target-volume coverage for >105% of the prescription dose. Among the six plans, VMAT-VB had the best HI for PTV Eval-boost and the highest doses to all OAR, except the coronary artery. Plans with EBs had lower mean doses for the contralateral lung and contralateral breast than plans with IMRT boosts. FIF-IMRT-EB had a low cost; plans with IMRT boosts had the highest costs

Conclusion: FIF-IMRT-EB may be the most suitable irradiation technique for hypofractionated WBI with SIB after BCS

Clinical target volume [CTV] in postoperation radiotherapy of esophageal squamous cell carcinoma could benefit from the detection of telomere length in lymph node

Background: This study evaluated the relation between telomere length in lymph node [LN] and prognosis of esophageal squamous cell carcinoma [ESCC]

Materials and Methods: LNs collected from 50 patients were assessed by pathological examination and quantitative reverse transcription polymerase chain reaction [qRT-PCR], which was used for detecting telomere length. The relation between clinical factors and the number of lymph node metastasis [LNM] identified were analyzed by the x2 test. The comparison of the pattern of LNM identified by pathological examination and detection of telomere length was assessed by Wilcoxon signed-rank test. Overall survival was assessed using the Kaplan-Meier method, and Cox proportional hazard regression analysis was used to evaluate the relationship between survival and the number of LNM

Results: The best threshold values, which could define the positive metastasis by detecting the telomere length, were 1.50, using the critical value method of statistic. Length of tumor, depth of tumor invasion and differentiation of tumor correlated closely with LNM were identified by detecting telomere length. The rates of LNM identified by detecting telomere length were 34.4%, 22.4%, 22.9%, 5.0% in 108,107, 7, and 3 LN station, respectively. The number of LNM identified by detecting telomere length was more closely related to the prognosis of ESCC than that of pathological examination [HR: 1.23 VERSUS 1.04]

Conclusion: The change of telomere length in LN was closely related to the prognosis of ESCC. Delineation of clinical target volume [CTV] may benefit from the detection of telomere length in regional LN

Relationship between white spot syndrome virus [WSSV] loads and characterizations of water quality in Litopenaeus vannamei culture ponds during the tropical storm

An in-situ experiment was conducted to investigate the effect of tropical storm on the white spot syndrome virus [WSSV] loads in Litopenaeus vannamei rearing ponds. White spot syndrome virus loads, heterotrophic bacteria, Vibrio and water quality [including temperature, dissolved oxygen [DO], salinity, pH, NH4-N, and NO2-N] were continually monitored through one tropical storm. The WSSV loads decreased when tropical storm made landfall, and substantially increased when typhoon passed. The variation of WSSV loads was correlated with DO, temperature, heterotrophic bacteria count, and ammonia-N concentrations. These results suggested that maintaining high level DO and promoting heterotrophic bacteria growth in the shrimp ponds might prevent the diseases' outbreak after the landfall of tropical storm