Multi-Class Deep Learning Model for Detecting Pediatric Distal Forearm Fractures Based on the AO/OTA Classification.

Common pediatric distal forearm fractures necessitate precise detection. To support prompt treatment planning by clinicians, our study aimed to create a multi-class convolutional neural network (CNN) model for pediatric distal forearm fractures, guided by the AO Foundation/Orthopaedic Trauma Association (AO/ATO) classification system for pediatric fractures. The GRAZPEDWRI-DX dataset (2008-2018) of wrist X-ray images was used. We labeled images into four fracture classes (FRM, FUM, FRE, and FUE with F, fracture; R, radius; U, ulna; M, metaphysis; and E, epiphysis) based on the pediatric AO/ATO classification. We performed multi-class classification by training a YOLOv4-based CNN object detection model with 7006 images from 1809 patients (80% for training and 20% for validation). An 88-image test set from 34 patients was used to evaluate the model performance, which was then compared to the diagnosis performances of two readers-an orthopedist and a radiologist. The overall mean average precision levels on the validation set in four classes of the model were 0.97, 0.92, 0.95, and 0.94, respectively. On the test set, the model's performance included sensitivities of 0.86, 0.71, 0.88, and 0.89; specificities of 0.88, 0.94, 0.97, and 0.98; and area under the curve (AUC) values of 0.87, 0.83, 0.93, and 0.94, respectively. The best performance among the three readers belonged to the radiologist, with a mean AUC of 0.922, followed by our model (0.892) and the orthopedist (0.830). Therefore, using the AO/OTA concept, our multi-class fracture detection model excelled in identifying pediatric distal forearm fractures.

Identification of potential anti-hyperglycemic compounds in Cordyceps militaris ethyl acetate extract: in vitro and in silico studies.

Cordyceps militaris has been long known for valuable health benefits by folk experience and was recently reported with diabetes-tackling evidences, thus deserving extending efforts on screening for component-activity relationship. In this study, experiments were carried out to find the evidence, justification, and input for computations on the potential against diabetes-related protein structures: PDB-4W93, PDB-3W37, and PDB-4A3A. Liquid chromatography identified 14 bioactive compounds in the ethyl acetate extract (1-14) and quantified the contents of cordycepin (0.11%) and adenosine (0.01%). Bioassays revealed the overall potential of the extract against α-amylase (IC50 = 6.443 ± 0.364 mg.mL-1) and α-glucosidase (IC50 = 2.580 ± 0.194 mg.mL-1). A combination of different computational platforms was used to select the most promising candidates for applications as anti-diabetic bio-inhibitors, i.e. 1 (ground state: -888.49715 a.u.; dipole moment 3.779 Debye; DS¯ -12.3 kcal.mol-1; polarizability 34.7 Å3; logP - 1.30), 10 (ground state: -688.52406 a.u.; dipole moment 5.487 Debye; DS¯ -12.6 kcal.mol-1; polarizability 24.9 Å3; logP - 3.39), and 12 (ground state: -1460.07276 a.u.; dipole moment 3.976 Debye; DS¯ -12.5 kcal.mol-1; polarizability 52.4 Å3; logP - 4.39). The results encourage further experimental tests on cordycepin (1), mannitol (10), and adenosylribose (12) to validate their in-practice diabetes-related activities, thus conducive to hypoglycemic applications.Communicated by Ramaswamy H. Sarma.

Automatic Detection of Meniscus Tears Using Backbone Convolutional Neural Networks on Knee MRI.

BACKGROUND: Timely diagnosis of meniscus injuries is key for preventing knee joint dysfunction and improving patient outcomes because it decreases morbidity and facilitates treatment planning. PURPOSE: To train and evaluate a deep learning model for automated detection of meniscus tears on knee magnetic resonance imaging (MRI). STUDY TYPE: Bicentric retrospective study. SUBJECTS: In total, 584 knee MRI studies, divided among training (n = 234), testing (n = 200), and external validation (n = 150) data sets, were used in this study. The public data set MRNet was used as a second external validation data set to evaluate the performance of the model. SEQUENCE: A 3 T, coronal, and sagittal images from T1-weighted proton density (PD) fast spin-echo (FSE) with fat saturation and T2-weighted FSE with fat saturation sequences. ASSESSMENT: The detection system for meniscus tear was based on the improved YOLOv4 model with Darknet-53 as the backbone. The performance of the model was also compared with that of three radiologists of varying levels of experience. The determination of the presence of a meniscus tear from surgery reports was used as the ground truth for the images. STATISTICAL TESTS: Sensitivity, specificity, prevalence, positive predictive value, negative predictive value, accuracy, and receiver operating characteristic curve were used to evaluate the performance of the detection model. Two-way analysis of variance, Wilcoxon signed-rank test, and Tukey's multiple tests were used to evaluate differences in performance between the model and radiologists. RESULTS: The overall accuracies for detecting meniscus tears using our model on the internal testing, internal validation, and external validation data sets were 95.4%, 95.8%, and 78.8%, respectively. One radiologist had significantly lower performance than our model in detecting meniscal tears (accuracy: 0.9025 ± 0.093 vs. 0.9580 ± 0.025). DATA CONCLUSION: The proposed model had high sensitivity, specificity, and accuracy for detecting meniscus tears on knee MRIs. EVIDENCE LEVEL: 3 TECHNICAL EFFICACY: Stage 2.

Phylogeny of Dengue virus type 2 isolated in the Central Highlands, Vietnam / Filogenia del virus del Dengue tipo2 identificado en el Altiplano Central de Vietnam

AbstractDengue fever is perhaps the most important viral re-emergent disease especially in tropical and sub-tropical countries, affecting about 50 million people around the world every year. In the Central Highlands regions of Vietnam, dengue fever still remains as a major public health issue. Although four viral serotypes have been currently identified, dengue virus type 2 (DENV-2) was involved in the most important outbreaks during 2010-2012, especially, 2010 when the fatality rate highly increased. Detection of genotype of DENV2 provided information on origin, distribution and genotype of the virus. In this study, DEN-2 isolated from dengue patients during the 2010-2012 epidemics was amplified and sequenced with E gene. The consensus sequences were aligned with reference E gene sequences of globally available Genbank. Phylogenetic analysis was performed using Neighbor-joining and Kimura 2-parameter model to construct phylogenetic tree. A total of 15 isolates (seven from 2010; one from 2011 and seven from 2012) were obtained from human serum samples. Phylogenetic analysis revealed that Asian genotype 1 is currently circulating locally in Central Highlands region. Isolates of this genotype were closely related to viruses from Thailand, Laos, and Cambodia. It indicated that these epidemics maybe imported into the Central Highlands region from South-East Asia neighbor countries. The study results would help in planning for prevention and control of dengue virus in Vietnam. Continuous monitoring of DENV genotypes is necessary to confirm the current findings and detect possible genotype shifts within the dengue viruses in the future.

ResumenPosiblemente, la fiebre del dengue es la enfermedad viral recurrente más importante en los países tropicales y subtropicales que afecta cerca de 50 millones de personas cada año en todo el mundo. En las regiones del Altiplano Central de Vietman, la fiebre del dengue aun se considera como una gran preocupación de salud pública. Aunque los cuatro serotipos virales han sido identificados, el virus del dengue tipo 2 (DENV-2) estuvo involucrado en el brote más importante durante el 2010-2012, especialmente en el 2010 cuando los índices de mortalidad aumentaron considerablemente. El descubrimiento del genotipo DENV-2 proporcionó información del origen, distribución y genotipo del virus. En este estudio, el serotipo DENV-2 identificado de pacientes con dengue durante las epidemias de 2010-2012 se amplificaron y secuenciaron con E gene. Las secuencias consenso se alinearon con secuencias de referencia mundiales de E gene disponibles en GenBank. El análisis filogenético se llevó a cabo utilizando el modelo Neighbor-joining y Kimura 2-parámetros para construir el árbol filogenético. Un total de 15 cepas (siete de 2010, una de 2011 y 7 de 2012) se obtuvieron de las muestras de suero humano. El análisis filogenético reveló que el genotipo asiático 1 circula localmente en la región del Altiplano Central. Las cepas de este genotipo estan muy relacionadas con los virus de Tailandia, Laos y Camboya. El análisis también indicó que estas epidemias pudieron migrar a la región del Altiplano Central desde los países vecinos del sureste asiático. Los resultados de este estudio pueden ayudar en la planificación de la prevención y el control del virus del dengue en Vietnam. Un monitoreo contante de los genotipos DENV es necesario para confirmar los hallazgos recientes y detectar los posibles cambios del genotipo de los virus del dengue en el futuro.

Risk factors for human infection with avian influenza A H5N1, Vietnam, 2004.

To evaluate risk factors for human infection with influenza A subtype H5N1, we performed a matched case-control study in Vietnam. We enrolled 28 case-patients who had laboratory-confirmed H5N1 infection during 2004 and 106 age-, sex-, and location-matched control-respondents. Data were analyzed by matched-pair analysis and multivariate conditional logistic regression. Factors that were independently associated with H5N1 infection were preparing sick or dead poultry for consumption < or =7 days before illness onset (matched odds ratio [OR] 8.99, 95% confidence interval [CI] 0.98-81.99, p = 0.05), having sick or dead poultry in the household < or =7 days before illness onset (matched OR 4.94, 95% CI 1.21-20.20, p = 0.03), and lack of an indoor water source (matched OR 6.46, 95% CI 1.20-34.81, p = 0.03). Factors not significantly associated with infection were raising healthy poultry, preparing healthy poultry for consumption, and exposure to persons with an acute respiratory illness.