A Modified XG Boost Classifier Model for Detection of Seizures and Non-Seizures

Diagnosis of Epilepsy is immensely important but challenging process, especially while using traditional manual seizure detection methods with the help of neurologists or brain experts’ guidance which are time consuming. Thus, an automated classification method is require to quickly detect seizures and non-seizures. Therefore, a machine learning algorithm based on a modified XGboost classifier model is employed to detect seizures quickly and improve classification accuracy. A focal loss function is employed with traditional XGboost classifier model to minimize mismatch of training and testing samples and enhance efficiency of the classification model. Here, CHB-MIT SCALP Electroencephalography (EEG) dataset is utilized to test the proposed classification model. Here, data gathered for all 24 patients from CHB-MIT Database is used to analyze the performance of proposed classification model. Here, 2-class-seizure experimental results of proposed classification model are compared against several state-of-art-seizure classification models. Here, cross validation experiments determine nature of 2-class-seizure as the prediction is seizure or non-seizure. The metrics results for average sensitivity and average specificity are nearly 100%. The proposed model achieves improvement in terms of average sensitivity against the best traditional method as 0.05% and for average specificity as 1%. The proposed modified XGBoost classifier model outperforms all the state-of-art-seizure detection techniques in terms of average sensitivity, average specificity.

Double Thresholding with Sine Entropy for Thermal Image Segmentation

Traditional thresholding methods are often used for image segmentation of real images. However, due to distinct characteristics of infrared thermal images, it is difficult to ensure an optimal image segmentation using the traditional thresholding algorithms, and therefore, sometimes this can lead to over-segmentation, missing object information, and/or spurious responses in the output. To overcome these issues, we propose a new thresholding technique that makes use of the sine entropy-based criterion. Moreover, we build a double thresholding technique that makes use of two thresholds to get the final image thresholding result. Besides, we introduce the sine entropy concept as a supplement of the Shannon entropy in creating threshold-dependent criterion derived from the grayscale histogram. We found that the sine entropy is more robust in interpreting the strength of the long-range correlation in the gray levels compared to the Shannon entropy. We have experimented with our method on several infrared thermal images collected from standard image databases to describe the performance. On comparing with the state-of-art methods, the qualitative results from the experiments show that the proposed method achieves the best performance with an average sensitivity of 0.98 and an average misclassification error of 0.01, and second-best performance with an average sensitivity of 0.99 and an average specificity of 0.93.

Another Perspective of the Miller Forensic Assessment of Symptoms Test – Part II: A Quantitative Review

Reference [1] presented pooled data for the specificity of the M-FAST cut-off, but ignored or excluded data based on poor justifications and used questionable analytic methods. The analyses here corrected the problems associated with [1]. No moderator substantively influenced sensitivity values. Therefore, sensitivity values were pooled across all studies (k = 25) to provide an overall estimate. Overall, the average sensitivity of the M-FAST cut-off was estimated to be 0.87, 95% CI [0.80, 0.91], and 80% of true sensitivity values were estimated to range from 0.63 to 0.96. Thus, there could be methodological scenarios when the M-FAST cut-off may not operate efficiently. Average specificity values for the M-FAST cut-off were moderated by one variable: the comparison group. On average, specificity values for clinical comparison (k = 15) groups (i.e., 0.80, 95% CI [0.73, 0.85]) were lower than specificity values for non-clinical comparison (k = 11) groups (i.e., 0.96, 95% CI [0.89, 0.99]). Unlike the CIs, the estimated distributions of true specificity values for the two subgroups overlapped, which suggests there could be scenarios when these subgroups share the same true specificity value. The M-FAST was designed to be a screener to detect potential feigning of psychiatric symptoms. An examinee is never to be designating as feigning or malingering psychiatric symptoms based on only a positive M-FAST result. As a screening instrument, the results here show that the M-FAST cut-off is operating adequately overall and negate the conclusions of [1].

A Cost-Effective and Non-Invasive pfeRNA-Based Test Differentiates Benign and Suspicious Pulmonary Nodules from Malignant Ones

The ability to differentiate between benign, suspicious, and malignant pulmonary nodules is imperative for definitive intervention in patients with early stage lung cancers. Here, we report that plasma protein functional effector sncRNAs (pfeRNAs) serve as non-invasive biomarkers for determining both the existence and the nature of pulmonary nodules in a three-stage study that included the healthy group, patients with benign pulmonary nodules, patients with suspicious nodules, and patients with malignant nodules. Following the standards required for a clinical laboratory improvement amendments (CLIA)-compliant laboratory-developed test (LDT), we identified a pfeRNA classifier containing 8 pfeRNAs in 108 biospecimens from 60 patients by sncRNA deep sequencing, deduced prediction rules using a separate training cohort of 198 plasma specimens, and then applied the prediction rules to another 230 plasma specimens in an independent validation cohort. The pfeRNA classifier could (1) differentiate patients with or without pulmonary nodules with an average sensitivity and specificity of 96.2% and 97.35% and (2) differentiate malignant versus benign pulmonary nodules with an average sensitivity and specificity of 77.1% and 74.25%. Our biomarkers are cost-effective, non-invasive, sensitive, and specific, and the qPCR-based method provides the possibility for automatic testing of robotic applications.

Application of the Non-Enzymatic Glucose Sensor Combined with Microfluidic System and Calibration Readout Circuit

In this research, we proposed a potentiometric sensor based on copper doped zinc oxide (CZO) films to detect glucose. Silver nanowires were used to improve the sensor’s average sensitivity, and we used the low power consumption instrumentation amplifier (UGFPCIA) designed by our research group to measure the sensing characteristics of the sensor. It was proved that the sensor performs better when using this system. In order to observe the stability of the sensor, we also studied the influence of two kinds of non-ideal effects on the sensor, such as the drift effect and the hysteresis effect. For this reason, we chose to combine the calibration readout circuit with the voltage-time (V-T) measurement system to optimize the measurement environment and successfully reduced the instability of the sensor. The drift rate was reduced by about 51.1%, and the hysteresis rate was reduced by 13% and 28% at different measurement cycles. In addition, the characteristics of the sensor under dynamic conditions were also investigated, and it was found that the sensor has an average sensitivity of 13.71 mV/mM and the linearity of 0.998 at a flow rate of 5.6 μL/min.

Rapid identification of pathogens associated with ventilator-associated pneumonia by Nanopore sequencing

Background Aetiology detection is crucial in the diagnosis and treatment of ventilator-associated pneumonia (VAP). However, the detection method needs improvement. In this study, we used Nanopore sequencing to build a quick detection protocol and compared the efficiency of different methods for detecting 7 VAP pathogens. Methods The endotracheal aspirate (ETA) of 83 patients with suspected VAP from Peking University Third Hospital (PUTH) was collected, saponins were used to deplete host genomes, and PCR- or non-PCR-amplified library construction methods were used and compared. Sequence was performed with MinION equipment and local data analysis methods were used for sequencing and data analysis. Results Saponin depletion effectively removed 11 of 12 human genomes, while most pathogenic bacterial genome results showed no significant difference except for S. pneumoniae. Moreover, the average sequence time decreased from 19.6 h to 3.62 h. The non-PCR amplification method and PCR amplification method for library build has a similar average sensitivity (85.8% vs. 86.35%), but the non-PCR amplification method has a better average specificity (100% VS 91.15%), and required less time. The whole method takes 5–6 h from ETA extraction to pathogen classification. After analysing the 7 pathogens enrolled in our study, the average sensitivity of metagenomic sequencing was approximately 2.4 times higher than that of clinical culture (89.15% vs. 37.77%), and the average specificity was 98.8%. Conclusions Using saponins to remove the human genome and a non-PCR amplification method to build libraries can be used for the identification of pathogens in the ETA of VAP patients within 6 h by MinION, which provides a new approach for the rapid identification of pathogens in clinical departments.

Prediction of students’ academic performance using ANN with mini-batch gradient descent and Levenberg-Marquardt optimization algorithms

Online learning indirectly increases stress, thereby reducing social interaction among students and leading to physical and mental fatigue, which in turn reduced students’ academic performance. Therefore, the prediction of academic performance is required sooner to identify at-risk students with declining performance. In this paper, we use artificial neural networks (ANN) to predict this performance. ANNs with two optimization algorithms, mini-batch gradient descent and Levenberg-Marquardt, are implemented on students’ learning activity data in course X, which is recorded on LMS UI. Data contains 232 students and consists of two periods: the first month and second month of study. Before ANNs are implemented, both normalization and usage of ADASYN are conducted. The results of ANN implementation using two optimization algorithms within 10 trials each are compared based on the average accuracy, sensitivity, and specificity values. We then determine the best period to predict unsuccessful students correctly. The results show that both algorithms give better predictions over two months instead of one. ANN with mini-batch gradient descent has an average sensitivity of 78%; the corresponding values for ANN with Levenberg-Marquardt are 75%. Therefore, ANN with mini-batch gradient descent as its optimization algorithm is more suitable for predicting students that have potential to fail.

Profiles of antimicrobial resistance of enterobacteria isolated at livestock enterprises of the Ural region

A study of the antibiotic susceptibility of bacteria of the genus Enterobacter, selected at regional dairy enterprises, was carried out. The purpose of this work was to assess the phenotypic resistance profiles of Enterobacter spp. in the loci of fermenal microbiocenoses related to milk production. Research methodology and methods. In the course of the work carried out, milk, mammary gland secretions, and udder washes from cows at dairy cattle breeding enterprises located in different districts of the Ural region were examined. The phenotypic resistance of Enterobacter spp. Isolates was analyzed to 10 antibacterial drugs: ciprofloxacin, enrofloxacin, ofloxacin, meropenem, doxycycline, chloramphenicol, ceftriaxone, amoxicillin, ampicillin, rifampicin. Results. Average sensitivity values of Enterobacter spp. for all surveyed enterprises were at the level of 2.0–3.3 conventional units (at maximum = 4) to target antibiotics, and at the level of 2.0–2.1 conventional units to non-target antibiotics. The highest bactericidal efficacy was found in fluoroquinolones, the lowest in doxycycline and chloramphenicol. For individual enterprises, the average resistance profile included good sensitivity to 3–4 antibiotics, reduced to 4–5 and resistance to 1–2 antibiotics. The main conclusion is that in eight surveyed enterprises, the usual pattern was the resistance of isolates or their low sensitivity to several antibiotics of different classes, which indicates an unfavorable situation with AMR. Scientific novelty. The results obtained in the course of the work performed made it possible to assess the current and actual levels of resistance of Enterobacter spp. Isolates inhabiting those loci of fermenal microbiocenoses that are directly related to milk production.

Multi-Class Classification of Lung Diseases Using CNN Models

In this study, we propose a multi-class classification method by learning lung disease images with Convolutional Neural Network (CNN). As the image data for learning, the U.S. National Institutes of Health (NIH) dataset divided into Normal, Pneumonia, and Pneumothorax and the Cheonan Soonchunhyang University Hospital dataset including Tuberculosis were used. To improve performance, preprocessing was performed with Center Crop while maintaining the aspect ratio of 1:1. As a Noisy Student of EfficientNet B7, fine-tuning learning was performed using the weights learned from ImageNet, and the features of each layer were maximally utilized using the Multi GAP structure. As a result of the experiment, Benchmarks measured with the NIH dataset showed the highest performance among the tested models with an accuracy of 85.32%, and the four-class predictions measured with data from Soonchunhyang University Hospital in Cheonan had an average accuracy of 96.1%, an average sensitivity of 92.2%, an average specificity of 97.4%, and an average inference time of 0.2 s.