AIoT in Agriculture: Safeguarding Crops from Pest and Disease Threats

A significant proportion of the world's agricultural production is lost to pests and diseases. To mitigate this problem, an AIoT system for the early detection of pest and disease risks in crops is proposed. It presents a system based on low-power and low-cost sensor nodes that collect environmental data and transmit it once a day to a server via a NB-IoT network. In addition, the sensor nodes use individual, retrainable and updatable machine learning algorithms to assess the risk level in the crop every 30 min. If a risk is detected, environmental data and the risk level are immediately sent. Additionally, the system enables two types of notification: email and flashing LED, providing online and offline risk notifications. As a result, the system was deployed in a real-world environment and the power consumption of the sensor nodes was characterized, validating their longevity and the correct functioning of the risk detection algorithms. This allows the farmer to know the status of their crop and to take early action to address these threats.

A systematic review of the application of machine-learning algorithms in multiple sclerosis

INTRODUCTION

The applications of artificial intelligence, and in particular automatic learning or "machine learning" (ML), constitute both a challenge and a great opportunity in numerous scientific, technical, and clinical disciplines. Specific applications in the study of multiple sclerosis (MS) have been no exception, and constitute an area of increasing interest in recent years.

OBJECTIVE

We present a systematic review of the application of ML algorithms in MS.

MATERIALS AND METHODS

We used the PubMed search engine, which allows free access to the MEDLINE medical database, to identify studies including the keywords "machine learning" and "multiple sclerosis." We excluded review articles, studies written in languages other than English or Spanish, and studies that were mainly technical and did not specifically apply to MS. The final selection included 76 articles, and 38 were rejected.

CONCLUSIONS

After the review process, we established 4 main applications of ML in MS: 1) classifying MS subtypes; 2) distinguishing patients with MS from healthy controls and individuals with other diseases; 3) predicting progression and response to therapeutic interventions; and 4) other applications. Results found to date have shown that ML algorithms may offer great support for health professionals both in clinical settings and in research into MS.

A systematic review of the application of machine-learning algorithms in multiple sclerosis

INTRODUCTION

The applications of artificial intelligence, and in particular automatic learning or "machine learning" (ML), constitute both a challenge and a great opportunity in numerous scientific, technical, and clinical disciplines. Specific applications in the study of multiple sclerosis (MS) have been no exception, and constitute an area of increasing interest in recent years.

OBJECTIVE

We present a systematic review of the application of ML algorithms in MS.

MATERIALS AND METHODS

We used the PubMed search engine, which allows free access to the MEDLINE medical database, to identify studies including the keywords "machine learning" and "multiple sclerosis." We excluded review articles, studies written in languages other than English or Spanish, and studies that were mainly technical and did not specifically apply to MS. The final selection included 76 articles, and 38 were rejected.

CONCLUSIONS

After the review process, we established 4 main applications of ML in MS: 1) classifying MS subtypes; 2) distinguishing patients with MS from healthy controls and individuals with other diseases; 3) predicting progression and response to therapeutic interventions; and 4) other applications. Results found to date have shown that ML algorithms may offer great support for health professionals both in clinical settings and in research into MS.

Reliability analysis of individual visual P1 and N1 maps indicates the heterogeneous topographies involved in early visual processing among human subjects

There is a lack of studies regarding the reliability of the event-related components (ERPs) of an electroencephalogram (EEG) used to assess cognitive processing in human subjects. To explore the reliability scores for the P1 and N1 components in two sessions (separated by an average of 116 days), twenty subjects performed a visual lateralized detection paradigm and EEG recording (58 channels) were employed. The session factor did not modulate the P1/N1 latencies. The visual field factor (left (LVF) or right (RVF)) was a determinant for the P1 and N1 topographical distributions as shown in previous studies. Moreover, topographical maps of the grand average showed a very strong correlation level between sessions (>0.9). Finally, individual maps demonstrated that the classic contralateral pattern for the P1 and N1 components was not always present in all subjects. In particular, compared to the N1 component, the P1 component exhibited a more complex set of individual topographical distributions, revealing that some steps are more heterogeneous among human subjects in early visual processing.

Individual test-retest reliability of evoked and induced alpha activity in human EEG data

Diverse psychological mechanisms have been associated with modulations of different EEG frequencies. To the extent of our knowledge, there are few studies of the test-retest reliability of these modulations in the human brain. To assess evoked and induced alpha reliabilities related to cognitive processing, EEG data from twenty subjects were recorded in 58 derivations in two different sessions separated by 49.5 ± 48.9 (mean ± standard deviation) days. A visual oddball was selected as the cognitive task, and three main parameters were analyzed for evoked and induced alpha modulations (latency, amplitude and topography). Latency and amplitude for evoked and induced modulations showed stable behavior between the two sessions. The correlation between sessions for alpha evoked and induced topographies in the grand average (group level) was r = 0.923, p<0.001; r = 0.962, p<0.001, respectively. The within-subject correlation values for evoked modulation ranged from 0.472 to 0.974 (mean: 0.766), whereas induced activity showed a different range, 0.193 to 0.892 (mean: 0.655). Individual analysis of the test-retest reliability showed a higher heterogeneity in the induced modulation, probably due to the heterogeneous phases found in the second case. However, despite this heterogeneity in phase values for induced activity relative to the onset of the stimuli, an excellent correlation score was obtained for group topography, with values that were better than those of the grand average evoked topography. As a main conclusion, induced alpha activity can be observed as a stable and reproducible response in the cognitive processing of the human brain.

Information Theoretic Approaches for Motor-Imagery BCI Systems: Review and Experimental Comparison

Brain computer interfaces (BCIs) have been attracting a great interest in recent years. The common spatial patterns (CSP) technique is a well-established approach to the spatial filtering of the electroencephalogram (EEG) data in BCI applications. Even though CSP was originally proposed from a heuristic viewpoint, it can be also built on very strong foundations using information theory. This paper reviews the relationship between CSP and several information-theoretic approaches, including the Kullback-Leibler divergence, the Beta divergence and the Alpha-Beta log-det (AB-LD)divergence. We also revise other approaches based on the idea of selecting those features that are maximally informative about the class labels. The performance of all the methods will be also compared via experiments.