Technical performance analysis of different types of spirometers

Background The spirometer is an important element in lung function examinations, and its accuracy is directly related to the accuracy of the results of these examinations and to the diagnosis and treatment of diseases. Our aim was to conduct a performance analysis of the detection techniques of differential pressure and ultrasonic portable spirometers commonly used in China. Methods A standard flow/volume simulator was used to analyze the performance (accuracy, repeatability, linearity, impedance, and so on) of portable spirometers, 4 imported and 6 domestic, based on 13 curves generated by different air sources in the ISO 26782:2009 standard. A Bland–Altman diagram was used to evaluate the consistency between the values measured by the spirometers and the simulator. Results The pass rates for accuracy, repeatability, linearity, and impedance for the 10 different portable spirometers were 50%, 100%, 70%, and 70%, respectively. Only 30% (3/10) of the spirometers—2 domestic and 1 imported—met all standards of quality and performance evaluation, while the rest were partially up to standard. In the consistency evaluation, only 3 spirometers were within both the consistency standard range and the acceptability range. Conclusion The quality and performance of different types of portable spirometers commonly used in the clinic differ. The use of a standard flow/volume simulator is helpful for the standard evaluation of the technical performance of spirometers.

Smart Farming and Surveillance using a Multipurpose Robot and Machine Learning

Agriculture is the procedure of producing food, feed, fiber, and many other favored merchandise by cultivating favorable vegetation and raising farm animals. The exercise of agriculture is also referred to as farming. However, there are a few challenges in raising agricultural productiveness in line with the unit of land, reducing rural poverty through a socially inclusive strategy that contains each agriculture in addition to non-farm employment, ensuring that agricultural boom responds to food security wishes. Nowadays, the advancement in ingenious farming techniques is progressively enhancing the crop yield making it greater worthwhile and reduce irrigation wastages. The proposed system is to layout and increases an autonomous vehicle that can carry out various agricultural activities inclusive of digging, sowing seeds, pumping insecticides, cutting undesirable grass within the discipline, etc. The farming land is autonomously irrigated with sufficient water with the help of moisture sensors within the land. The autonomous robot's general operation and the irrigation gadget are monitored and maintained using Machine Learning (ML) algorithms. The overall operation and records are measured via the sensors are stored in the cloud for device gaining knowledge of (ML) algorithm and future references. Thereby, it increases the machine's overall performance accuracy and reduces the human power and saves the time required to cultivate the farmland

Temporal Weighted Averaging for Asynchronous Federated Intrusion Detection Systems

Federated learning (FL) is an emerging subdomain of machine learning (ML) in a distributed and heterogeneous setup. It provides efficient training architecture, sufficient data, and privacy-preserving communication for boosting the performance and feasibility of ML algorithms. In this environment, the resultant global model produced by averaging various trained client models is vital. During each round of FL, model parameters are transferred from each client device to the server while the server waits for all models before it can average them. In a realistic scenario, waiting for all clients to communicate their model parameters, where client models are trained on low-power Internet of Things (IoT) devices, can result in a deadlock. In this paper, a novel temporal model averaging algorithm is proposed for asynchronous federated learning (AFL). Our approach uses a dynamic expectation function that computes the number of client models expected in each round and a weighted averaging algorithm for continuous modification of the global model. This ensures that the federated architecture is not stuck in a deadlock all the while increasing the throughput of the server and clients. To implicate the importance of asynchronicity in cybersecurity, the proposed algorithm is tested using NSL-KDD intrusion detection system datasets. The performance accuracy of the global model is about 99.5% on the dataset, outperforming traditional FL models in anomaly detection. In terms of asynchronicity, we get an increased throughput of almost 10.17% for every 30 timesteps.

Predictive Group Learning Behavior Approach and Inquiry Learning For Higher Education

Group work can inspire students, encourage constructive learning, and improve essential critical thinking, communication, and decision-making in the present competitive world. The risk factors in group learning include students who prefer working alone and strongly despise dealing with things created by gathering in teams. Online learners often have problems locating lasting peace times for group therapy sessions are considered an essential factor. A predictive Group Learning Behavior Approach (PGLBA) has been proposed. Students who commute to college agree to group conferences and workgroup learning and Inquiry learning for higher education. The grounded Inquiry Learning Approach is invented to strengthen students’ enjoyment of active group learning, and the students find times for group meetings that are often mutually advantageous. The simulation analysis is performed based on performance, accuracy, and efficiency proves the proposed framework’s reliability. The experimental results show that the proposed PGLBA-IL model enhances the accuracy ratio of 81.2%, an efficiency ratio of the number of students 86.4%, and the overall performance analysis ratio of 85.1% compared to others existing approaches.

An explainable Artificial Intelligence approach to study MCI to AD conversion via HD-EEG processing

An explainable Artificial Intelligence (xAI) approach is proposed to longitudinally monitor subjects affected by Mild Cognitive Impairment (MCI) by using high-density electroencephalography (HD-EEG). To this end, a group of MCI patients was enrolled at IRCCS Centro Neurolesi Bonino Pulejo of Messina (Italy) within a follow-up protocol that included two evaluations steps: T0 (first evaluation) and T1 (three months later). At T1, four MCI patients resulted converted to Alzheimer’s Disease (AD) and were included in the analysis as the goal of this work was to use xAI to detect individual changes in EEGs possibly related to the degeneration from MCI to AD. The proposed methodology consists in mapping segments of HD-EEG into channel-frequency maps by means of the power spectral density. Such maps are used as input to a Convolutional Neural Network (CNN), trained to label the maps as “T0” (MCI state) or “T1” (AD state). Experimental results reported high intra-subject classification performance (accuracy rate up to 98.97% (95% confidence interval: 98.68–99.26)). Subsequently, the explainability of the proposed CNN is explored via a Grad-CAM approach. The procedure allowed to detect which EEG-channels (i.e., head region) and range of frequencies (i.e., sub-bands) resulted more active in the progression to AD. The xAI analysis showed that the main information is included in the delta sub-band and that, limited to the analyzed dataset, the highest relevant areas are: the left-temporal and central-frontal lobe for Sb01, the parietal lobe for Sb02, the left-frontal lobe for Sb03 and the left-frontotemporal region for Sb04.

A Machine Learning Approach to Predicting Diabetes Complications

Diabetes mellitus (DM) is a chronic disease that is considered to be life-threatening. It can affect any part of the body over time, resulting in serious complications such as nephropathy, neuropathy, and retinopathy. In this work, several supervised classification algorithms were applied for building different models to predict and classify eight diabetes complications. The complications include metabolic syndrome, dyslipidemia, neuropathy, nephropathy, diabetic foot, hypertension, obesity, and retinopathy. For this study, a dataset collected by the Rashid Center for Diabetes and Research (RCDR) located in Ajman, UAE, was utilized. The dataset consists of 884 records with 79 features. Some essential preprocessing steps were applied to handle the missing values and unbalanced data problems. Furthermore, feature selection was performed to select the top five and ten features for each complication. The final number of records used to train and build the binary classifiers for each complication was as follows: 428—metabolic syndrome, 836—dyslipidemia, 223—neuropathy, 233—nephropathy, 240—diabetic foot, 586—hypertension, 498—obesity, 228—retinopathy. Repeated stratified k-fold cross-validation (with k = 10 and a total of 10 repetitions) was employed for a better estimation of the performance. Accuracy and F1-score were used to evaluate the models’ performance reaching a maximum of 97.8% and 97.7% for accuracy and F1-scores, respectively. Moreover, by comparing the performance achieved using different attributes’ sets, it was found that by using a selected number of features, we can still build adequate classifiers.

Pupillary dynamics of optimal effort

ABSTRACTWhile a substantial body of work has shown that cognitive effort is aversive and costly, a separate line of research on intrinsic motivation suggests that people spontaneously seek challenging tasks. According to one prominent account of intrinsic motivation, the Learning Progress Motivation theory, the preference for difficult tasks reflects the dynamic range that these tasks yield for minimization of performance accuracy prediction errors (Oudeyer, Kaplan & Hafner, 2007). Here we test this hypothesis, by asking whether greater engagement with intermediately difficult tasks, indexed by subjective ratings and objective pupil measurements, is a function of trial-wise changes in performance prediction error. In a novel paradigm, we determined each individual’s capacity for task performance and used difficulty levels that are too low, intermediately challenging or high for that individual. We demonstrated that intermediately challenging tasks resulted in greater liking and engagement scores compared with easy tasks. Task-evoked and baseline pupil size tracked objective task difficulty, where challenging tasks were associated with smaller baseline and greater phasic pupil responses than easy tasks. Most importantly, pupil responses were predicted by trial-to-trial changes in expected accuracy, performance prediction error magnitude and changes in prediction errors (learning progress), whereas smaller baseline pupil responses also predicted greater subjective engagement scores. Together, these results suggest that what is underlying the link between task engagement and intermediate tasks might be the dynamic range that these tasks yield for minimization of performance accuracy prediction errors.

Comparative analysis of AI-based models for short-term photovoltaic power forecasting in energy cooperatives

Energy communities can support the energy transition, by engaging citizens through collective energy actions and generate positive economic, social and environmental outcomes. Renewable Energy Sources (RES) are gaining increasing share in the electricity mix as the economy decarbonises, with Photovoltaic (PV) plants to becoming more efficient and affordable. By incorporating Artificial Intelligence (AI) techniques, innovative applications can be developed to provide added value to energy communities. In this context, the scope of this paper is to compare Machine Learning (ML) and Deep Learning (DL) algorithms for the prediction of short-term production in a solar plant under an energy cooperative operation. Three different cases are considered, based on the data used as inputs for forecasting purposes. Lagged inputs are used to assess the historical data needed, and the algorithms’ accuracy is tested for the next hour’s PV production forecast. The comparative analysis between the proposed algorithms demonstrates the most accurate algorithm in each case, depending on the available data. For the highest performing algorithm, its performance accuracy in further forecasting horizons (3 hours, 6 hours and 24 hours) is also tested.

Prediction Models Founded on Soil Characteristics for the Estimated Uptake of Nine Metals by Okra Plant, Abelmoschus esculentus (L.) Moench., Cultivated in Agricultural Soils Modified with Varying Sewage Sludge Concentrations

Prediction models were developed to estimate the extent to which the metals Cd, Co, Cr, Cu, Fe, Mn, Ni, Pb, and Zn were taken up by the fruits, the leaves, the stems, and the roots of the okra plant, Abelmoschus esculentus (L.) Moench., grown under greenhouse conditions in soil modified with a spectrum of sewage sludge concentrations: 0, 10, 20, 30, 40, and 50 g/kg. All the metals under investigation, apart from Cd, were more concentrated in the A. esculentus roots than in any other organ. Overall, the sum of the metal concentration (mg/kg) within the varying plant tissues can be ranked in the following order: roots (13,795.5) > leaves (1252.7) > fruits (489.3) > stems (469.6). For five of the metals (i.e., Cd, Co, Fe, Mn, and Pb), the BCF was <1; for the remaining four metals, the BCF was >1, (i.e., Cr, 1.074; Cu, 1.347; Ni, 1.576; and Zn, 1.031). The metal BCFs were negatively correlated with the pH of the soil and positively correlated with soil OM content. The above-ground tissues exhibited a TF < 1 for all metals, apart from Cd with respect to the leaves (2.003) and the fruits (2.489), and with the exception of Mn in relation to the leaves (1.149). Further positive associations were demonstrated for the concentrations of all the metals in each examined plant tissue and the corresponding soil metal concentration. The tissue uptakes of the nine metals were negatively correlated with soil pH, but positively associated with the OM content in the soil. The generated models showed high performance accuracy; students’ t-tests indicated that any differences between the measured and forecasted concentrations of the nine metals within the four tissue types of A. esculentus failed to reach significance. It can, therefore, be surmised that the prediction models described in the current research form a feasible method with which to determine the safety and risk to human health when cultivating the tested species in soils modified with sewage sludge.

Music-Assisted Training for Dart Throwing Novices: Post-Training Effects on Heart Rate and Performance Accuracy

While previous research has suggested that lowering athletes’ heart rates can enhance sports performance, it is unknown whether slow-paced music might induce a lower heart rate and thereby improve some types of motor performance. In this study, we investigated the effects of different types of music during dart-throw training on both heart rate and dart-throwing performance in 45 ( M age = 19.7, SD = 0.31 years) novice dart throwers who were randomly assigned to either a Slow Music Group (SMG), a Fast Music Group (FMG), or a Control Group (CG). All participants completed three dart-throwing blocks - Pre-Test, Practice, and Post-Test. During the Practice block, participants practiced dart-throwing with either slow-paced, fast-paced or no music according to their assigned group. We recorded the participants’ heart rates and total dart-throwing accuracy scores during Pre-Test and Post-Test. Music-assisted dart-throw training with slow-paced music was effective in significantly inhibiting a performance-related increase in heart rate and was associated with the greatest dart throwing improvement after training.