Heroin avoidance may be achieved by MMT or ABT; however, the neural mechanism underlying these therapeutic methods differs.
Objective: Pneumonia is a lung infection and causes the inflammation of the small air sacs (Alveoli) in one or both lungs. Proper and faster diagnosis of pneumonia at an early stage is imperative for optimal patient care. Currently, chest X-ray is considered as the best imaging modality for diagnosing pneumonia. However, the interpretation of chest X-ray images is challenging. To this end, we aimed to use an automated convolutional neural network-based transfer-learning approach to detect pneumonia in paediatric chest radiographs. Methods: Herein, an automated convolutional neural network-based transfer-learning approach using four different pre-trained models (i.e. VGG19, DenseNet121, Xception, and ResNet50) was applied to detect pneumonia in children (1–5 years) chest X-ray images. The performance of different proposed models for testing data set was evaluated using five performances metrics, including accuracy, sensitivity/recall, Precision, area under curve, and F1 score. Results: All proposed models provide accuracy greater than 83.0% for binary classification. The pre-trained DenseNet121 model provides the highest classification performance of automated pneumonia classification with 86.8% accuracy, followed by Xception model with an accuracy of 86.0%. The sensitivity of the proposed models was greater than 91.0%. The Xception and DenseNet121 models achieve the highest classification performance with F1-score greater than 89.0%. The plotted area under curve of receiver operating characteristics of VGG19, Xception, ResNet50, and DenseNet121 models are 0.78, 0.81, 0.81, and 0.86, respectively. Conclusion: Our data showed that the proposed models achieve a high accuracy for binary classification. Transfer learning was used to accelerate training of the proposed models and resolve the problem associated with insufficient data. We hope that these proposed models can help radiologists for a quick diagnosis of pneumonia at radiology departments. Moreover, our proposed models may be useful to detect other chest-related diseases such as novel Coronavirus 2019. Advances in knowledge: Herein, we used transfer learning as a machine learning approach to accelerate training of the proposed models and resolve the problem associated with insufficient data. Our proposed models achieved accuracy greater than 83.0% for binary classification.
Purpose. To investigate the factors contributing to mortality in coronavirus disease 2019 (COVID-19) patients admitted in the intensive care unit (ICU) and design a model to predict the mortality rate. Method. We retrospectively evaluated the medical records and CT images of the ICU-admitted COVID-19 patients who had an on-admission chest CT scan. We analyzed the patients’ demographic, clinical, laboratory, and radiologic findings and compared them between survivors and nonsurvivors. Results. Among the 121 enrolled patients (mean age, 62.2 ± 14.0 years; male, 82 (67.8%)), 41 (33.9%) survived, and the rest succumbed to death. The most frequent radiologic findings were ground-glass opacity (GGO) (71.9%) with peripheral (38.8%) and bilateral (98.3%) involvement, with lower lobes (94.2%) predominancy. The most common additional findings were cardiomegaly (63.6%), parenchymal band (47.9%), and crazy-paving pattern (44.4%). Univariable analysis of radiologic findings showed that cardiomegaly p : 0.04 , pleural effusion p : 0.02 , and pericardial effusion p : 0.03 were significantly more prevalent in nonsurvivors. However, the extension of pulmonary involvement was not significantly different between the two subgroups (11.4 ± 4.1 in survivors vs. 11.9 ± 5.1 in nonsurvivors, p : 0.59 ). Among nonradiologic factors, advanced age p : 0.002 , lower O2 saturation p : 0.01 , diastolic blood pressure p : 0.02 , and hypertension p : 0.03 were more commonly found in nonsurvivors. There was no significant difference between survivors and nonsurvivors in terms of laboratory findings. Three following factors remained significant in the backward logistic regression model: O2 saturation (OR: 0.91 (95% CI: 0.84–0.97), p : 0.006 ), pericardial effusion (6.56 (0.17–59.3), p : 0.09 ), and hypertension (4.11 (1.39–12.2), p : 0.01 ). This model had 78.7% sensitivity, 61.1% specificity, 90.0% positive predictive value, and 75.5% accuracy in predicting in-ICU mortality. Conclusion. A combination of underlying diseases, vital signs, and radiologic factors might have prognostic value for mortality rate prediction in ICU-admitted COVID-19 patients.
Purpose:To train a convolutional neural network (CNN) model from scratch to automatically detect tuberculosis (TB) from chest X-ray (CXR) images and compare its performance with transfer learning based technique of different pre-trained CNNs. Material and methods:We used two publicly available datasets of postero-anterior chest radiographs, which are from Montgomery County, Maryland, and Shenzhen, China. A CNN (ConvNet) from scratch was trained to automatically detect TB on chest radiographs. Also, a CNN-based transfer learning approach using five different pre-trained models, including Inception_v3, Xception, ResNet50, VGG19, and VGG16 was utilized for classifying TB and normal cases from CXR images. The performance of models for testing datasets was evaluated using five performances metrics, including accuracy, sensitivity/recall, precision, area under curve (AUC), and F1-score. Results:All proposed models provided an acceptable accuracy for two-class classification. Our proposed CNN architecture (i.e., ConvNet) achieved 88.0% precision, 87.0% sensitivity, 87.0% F1-score, 87.0% accuracy, and AUC of 87.0%, which was slightly less than the pre-trained models. Among all models, Exception, ResNet50, and VGG16 provided the highest classification performance of automated TB classification with precision, sensitivity, F1-score, and AUC of 91.0%, and 90.0% accuracy. Conclusions:Our study presents a transfer learning approach with deep CNNs to automatically classify TB and normal cases from the chest radiographs. The classification accuracy, precision, sensitivity, and F1-score for the detection of TB were found to be more than 87.0% for all models used in the study. Exception, ResNet50, and VGG16 models outperformed other deep CNN models for the datasets with image augmentation methods.
Breast cancer is the second most common cause of cancer-related mortality in women worldwide, with a lifetime risk of approximately 12% (1). Tumor size is one of the main prognostic factors in breast cancer and is reported to correlate with lymph node involvement, tumor grade, and overall survival rate (2). Tumor size is also a factor assessed to determine treatment plans: breast conservation, mastectomy, or neoadjuvant chemotherapy (3). Accordingly, precise estimation of tumor size is of utmost importance for planning a therapeutic strategy, and the main imaging modalities are mammogram (MGM), ultrasound (US), and magnetic resonance imaging (MRI). Each of these modalities has certain strengths and weaknesses in breast tumor evaluation. For instance, MGM is superior in identifying malignant calcifications; however, the obscurity of the margins and magnification variability limits the accuracy of measurements by this method (4). The sensitivity of MGM to detect malignant lesions in younger patients with dense breast tissue is also reported to be poor (5, 6). As for US, its ability to measure tumors in multiple planes is a great strength that enables a skilled operator to make measurements of its largest dimension (7). However, one main limitation of US is that it is highly operator dependent (8). MRI also offers the merit of multiplanar imaging along with a higher accuracy in assessing multicentric and multifocal lesions (9, 10); however, MRI has been reported to overestimate tumor size (9, 11), and the extent of background parenchymal enhancement (BPE) affects its accuracy (12).In this regard, studies have assessed the accuracy of tumor size estimation by MGM and US (11,[13][14][15], and compared their measurements with those by MRI (10,16,17). In comparing US and MGM, some studies have reported that US has a higher accuracy than MGM (7, 11,
The Syntaxin Binding Protein 1 (STXBP1) plays an important role in regulating neurotransmitter release and synaptic vesicle fusion through binding to syntaxin-1A (STX1A) and changing its conformation. In this study, we identified a de novo mutation (c.C1162T: p.R388X) in exon 14 of the STXBP1 gene causing an epileptic encephalopathy, early infantile, non-epileptic movement, and unclassified infantile spasms disorders in a 5-year-old boy by whole-exome sequencing. The segregation of this genetic variant was examined in the patient as well as in his parents. We found the R388X in heterozygous state in the proband but not in his parents. This genetic change could be due to de nova mutation or germlinemosaicism. © 2019 Tehran University of Medical Sciences. All rights reserved. Acta MedIran 2019;57(8):518-521.
Uterine Arteriovenous malformations (AVM) are vascular disorders characterized by complex high‐flow tangles of abnormal vessels connecting arteries and veins with bypassing capillaries. Recently, the terminology applied to describe uterine AVMs has been modified. Most AVMs are acquired. The term enhanced myometrial vascularity (EMV) is used to describe any condition in which any uterine pathology may lead to increased myometrial vascularity regardless of the absence or presence of residual tissue of gestation.
Background:: Polycystic kidney disease (PKD) is an autosomal recessive disorder resulting from mutations in the PKHD1 gene on chromosome 6 (6p12), a large gene spanning 470 kb of genomic DNA. Objective: The aim of the present study was to report newly identified mutations in the PKHD1 gene in two Iranian families with PKD. Materials and Methods: Genetic alterations of a 3-month-old boy and a 27-year-old girl with PKD were evaluated using whole-exome sequencing. The PCR direct sequencing was performed to analyse the co-segregation of the variants with the disease in the family. Finally, the molecular function of the identified novel mutations was evaluated by in silico study. Results: In the 3 month-old boy, a novel homozygous frameshift mutation was detected in the PKHD1 gene, which can cause PKD. Moreover, we identified three novel heterozygous missense mutations in ATIC, VPS13B, and TP53RK genes. In the 27-year-old woman, with two recurrent abortions history and two infant mortalities at early weeks due to metabolic and/or renal disease, we detected a novel missense mutation on PKHD1 gene and a novel mutation in ETFDH gene. Conclusion: In general, we have identified two novel mutations in the PKHD1 gene. These molecular findings can help accurately correlate genotype and phenotype in families with such disease in order to reduce patient births through preoperative genetic diagnosis or better management of disorders.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.