The key component in deep learning research is the availability of training data sets. With a limited number of publicly available COVID-19 chest X-ray images, the generalization and robustness of deep learning models to detect COVID-19 cases developed based on these images are questionable. We aimed to use thousands of readily available chest radiograph images with clinical findings associated with COVID-19 as a training data set, mutually exclusive from the images with confirmed COVID-19 cases, which will be used as the testing data set. We used a deep learning model based on the ResNet-101 convolutional neural network architecture, which was pretrained to recognize objects from a million of images and then retrained to detect abnormality in chest X-ray images. The performance of the model in terms of area under the receiver operating curve, sensitivity, specificity, and accuracy was 0.82, 77.3%, 71.8%, and 71.9%, respectively. The strength of this study lies in the use of labels that have a strong clinical association with COVID-19 cases and the use of mutually exclusive publicly available data for training, validation, and testing.

Reasonable use of agricultural machinery has an extraordinary potential for poverty alleviation by increasing land and labor productivity in Thailand, Vietnam, and even in Bangladesh. This study was conducted under a program entitled “Agriculture Mechanization, Agro-Processing, Value addition and Export Market Development in Thailand and Vietnam from 1–14 November, 20I9” from the Ministry of Agriculture, Bangladesh. In all three distinct nations, farming activities represent a significant area of activity and remains the biggest wellspring of agricultural business. About 10.5% of Thailand’s, 21.5% of Vietnam’s, and 14.23% of Bangladesh’s GDP come from agriculture. For sustainable development, it is essential to modernize agriculture through the mechanization of its operations, which is therefore inevitable in the studied countries. Thailand’s government started mechanization in 1891 with the import of steam-powered tractor and rotary hoes. Since then the country has witnessed several milestones in the course of mechanization development. The focal plain agro-ecological zone of the state is the maximum and almost fully modernized area. As of now, there are two methods of practicing farming apparatus use: as a proprietor and/or through custom renting provision which coincides with Vietnam and Bangladesh. Historically, mechanization patterns in Vietnam can been described by tillage machinery with associated implement equipment use preceding 1975. This was non-linear, followed by a decreasing trend during the 80s prior to recovery during the 90s, with significant disparities in implementation status across the areas. In 2018, the number of tillage implements and harvesters was boosted about 1.6 and 25.6 times, respectively compared with 2006. The percentage of machinery use in soil tillage operation is 80% of the whole territory of cultivable land in Vietnam, compared to about 90% in Bangladesh and 100% in Thailand. Mechanization in Bangladesh started before independence with the importation of 2-wheel tractors and irrigation pumps in the last part of the 1960s as part of ‘Green Revolution’ activities. To continue this momentum, the Bangladesh Government permitted the continuation of agricultural machinery importation after later autonomy. Machinery use in different agricultural activities has increased in recent years in the areas of irrigation, land preparation, intercultural operation, and threshing. Though its degree of advancement is by and large still quite low contrasted with other South Asian nations, it is noticeable that the most recent two decades, the pace of mechanization has increased rapidly with the increase of mechanical power use in farm activities. The use of farm machinery in rice cultivation has been the most amazing when contrasted with different crops in these three nations. A clear comparison has been given in the paper, which aims to help researchers and policymakers take necessary measures.

No abstract

Recently, the concept of weakly supervised learning has gained popularity in the high-energy physics community due to its ability to learn even with a noisy and impure dataset. This method is valuable in the quest to discover the elusive beyond Standard Model (BSM) particle. Nevertheless, the weakly supervised learning method still requires a learning sample that describes the features of the BSM particle truthfully to the classification model. Even with the various theoretical framework such as supersymmetry and the quantum black hole, creating a BSM sample is not a trivial task since the exact feature of the particle is unknown. Due to these difficulties, we propose an alternative classifier type called the one-class classification (OCC). OCC algorithms require only background or noise samples in its training dataset, which is already abundant in the high-energy physics community. The algorithm will flag any sample that does not fit the background feature as an abnormality. In this paper, we introduce two new algorithms called EHRA and C-EHRA, which use machine learning regression and clustering to detect anomalies in samples. We tested the algorithms’ capability to create distinct anomalous patterns in the presence of BSM samples and also compare their classification output metrics to the Isolation Forest (ISF), a well-known anomaly detection algorithm. Five Monte Carlo supersymmetry datasets with the signal to noise ratio equal to 1, 0.1, 0.01, 0.001, and 0.0001 were used to test EHRA, C-EHRA and ISF algorithm. In our study, we found that the EHRA with an artificial neural network regression has the highest ROC-AUC score at 0.7882 for the balanced dataset, while the C-EHRA has the highest precision-sensitivity score for the majority of the imbalanced datasets. These findings highlight the potential use of the EHRA, C-EHRA, and other OCC algorithms in the quest to discover BSM particles.

Background. Early diagnosis and interceptive treatment of the maxillary canine impaction is crucial as it reduces treatment complexity and decreases complications and adverse outcomes. Aim and Objectives. To determine the mean maxillary canine position among 9-10-year-old children and predict the risk of impaction of the maxillary canines. Methodology. Panoramic radiographs (PANs) of 289 healthy children aged between 9 and 10 years were observed where the average position of maxillary canines was related to the lateral incisor, sector locations, and angulations to the bicondylar line were traced. The average position was obtained by using descriptive statistics. One sample Wilcoxon signed-rank test is done to predict the risk of canine impaction by comparing the data obtained to the average position from prior studies. Results. A total of 289 PANs (126 males and 163 females) were utilized for the analysis. The findings showed that the average position of the maxillary canines in our population was statistically different from the average position of nonimpacted canines in previous studies. However, on average, more than 85% of canines in our population were still located within the safe range of satisfactory position, with females showing slight predominance outside of the acceptable range. The mean scores of the angles between the right canine and lateral incisor were significantly higher among females than males ( p = 0.001 ). Similarly, females had a significantly higher mean angle of the left canine than males ( p < 0.001 ). In regard to the angles between the bicondylar line and permanent maxillary canine, the mean scores were not significantly different ( p > 0.05 ) on both the left and right side. Conclusion. There is a low risk of impaction of maxillary canines in the Malaysian population. However, more retrospective studies using more radiographic and clinical indicators need to be done to confirm the risk of impaction further.