A Bayesian Network Concept for Pain Assessment (Preprint)

UNSTRUCTURED Pain is a subjective phenomenon caused/perceived centrally and modified by physical, physiological, or social influences. Currently, the most commonly used approaches for pain measurement rely on self-reporting of pain level on a discrete rating scale. This provides a subjective and only semi-quantitative indicator of pain. This paper presents an approach that combines self-reported pain with pain-related biomarkers to be obtained from biosensors (in development) and possibly other sources of evidence to provide more dependable estimates of experienced pain, a clinical decision support system. We illustrate the approach using a Bayes network, but also describe other artificial intelligence (AI) methods that provide other ways to combine evidence. We also propose an optimization approach for tuning the AI method parameters (opaque to clinicians) so as to best approximate the kinds of outputs most useful to medical practitioners. We present some data from a sample of 379 patients that illustrate several evidence patterns we may expect in real healthcare situations. The majority (79.7%) of our patients show consistent evidence suggesting this biomarker approach may be reasonable. We also found five patterns of inconsistent evidence. These suggest a direction for further exploration. Finally, we sketch out an approach for collecting medical experts’ guidance as to the way the combined evidence might be presented so as to provide the most useful guidance (also needed for any optimization approach). We recognize that one possible outcome may be that all this approach may be able to provide is a quantified measure of the extent to which the evidence is consistent or not, leaving the final decision to the clinicians (where it must reside). Pointers to additional sources of evidence might also be possible in some situations.

Analysis of the safety of unmanned vessels based on the structure of the Bayes network-based risk model

Беспилотное торговое судно (БЭС) уже не кажется выдумкой фантастов - оно уже практически часть нашей реальности. Технические разработки ведутся повсеместно – остается лишь доказать безопасность концепции БЭС. Аспект безопасности безэкипажного судна является активно исследуемой проблемой – но она до сих пор не решена на сегодняшний день. Причиной является отсутствие статистической информации о реальных условиях эксплуатации и конструкции беспилотных судов, которые еще находятся в стадии разработок. В попытке преодолеть этот пробел необходимо провести анализ рисков, связанных с эксплуатацией беспилотных судов, где все соответствующие опасности и последствия должны быть систематически и количественно оценены. В данной работе представлены результаты первого этапа такого анализа, а именно анализа опасностей, связанных с эксплуатацией беспилотных судов. Перечень опасностей охватывает различные аспекты беспилотного судоходства, возникающие как на этапе проектирования, так и на этапе эксплуатации судна. Впоследствии эти опасности и связанные с ними последствия организуются случайным образом, что требует разработки структуры модели риска. The unmanned merchant ship (UES) no longer seems to be a fiction of science fiction-it is already practically part of our reality. Technical developments are being conducted everywhere – it remains only to prove the safety of the BES concept. The safety aspect of an unmanned vessel is an actively researched problem – but it is still not solved to date. The reason is the lack of statistical information about the actual operating conditions and design of unmanned vessels, which are still under development. In an attempt to bridge this gap, a risk analysis of the operation of unmanned vessels should be conducted, where all relevant hazards and consequences should be systematically and quantified. This paper presents the results of the first stage of such analysis, namely, the analysis of the dangers associated with the operation of unmanned vessels. The list of hazards covers various aspects of unmanned navigation that arise both at the design stage and during the operation of the vessel. Subsequently, these hazards and their associated consequences are organized randomly, which requires the development of a risk model structure.

Identification of significant climatic risk factors and machine learning models in dengue outbreak prediction

Background Dengue fever is a widespread viral disease and one of the world’s major pandemic vector-borne infections, causing serious hazard to humanity. The World Health Organisation (WHO) reported that the incidence of dengue fever has increased dramatically across the world in recent decades. WHO currently estimates an annual incidence of 50–100 million dengue infections worldwide. To date, no tested vaccine or treatment is available to stop or prevent dengue fever. Thus, the importance of predicting dengue outbreaks is significant. The current issue that should be addressed in dengue outbreak prediction is accuracy. A limited number of studies have conducted an in-depth analysis of climate factors in dengue outbreak prediction. Methods The most important climatic factors that contribute to dengue outbreaks were identified in the current work. Correlation analyses were performed in order to determine these factors and these factors were used as input parameters for machine learning models. Top five machine learning classification models (Bayes network (BN) models, support vector machine (SVM), RBF tree, decision table and naive Bayes) were chosen based on past research. The models were then tested and evaluated on the basis of 4-year data (January 2010 to December 2013) collected in Malaysia. Results This research has two major contributions. A new risk factor, called the TempeRain factor (TRF), was identified and used as an input parameter for the model of dengue outbreak prediction. Moreover, TRF was applied to demonstrate its strong impact on dengue outbreaks. Experimental results showed that the Bayes Network model with the new meteorological risk factor identified in this study increased accuracy to 92.35% for predicting dengue outbreaks. Conclusions This research explored the factors used in dengue outbreak prediction systems. The major contribution of this study is identifying new significant factors that contribute to dengue outbreak prediction. From the evaluation result, we obtained a significant improvement in the accuracy of a machine learning model for dengue outbreak prediction.

Bayesian Networks for software testing

A software design methodology is proposed that involves development of approximate models based on Bayesian Networks capturing probabilistic representations of expected behavior, which are further used in developing and running tests that can dynamically diagnose bugs and attacks during production. While automation of Software design is still a very remote goal, it can already benefit from AI tools and ideas. One of the main challenges with automating software design methods, for any product with modest complexity, is the mere intractability of enumerating all scenarios of the product usage when also taking into account user intentions. This leads to an intractability of generating exact specifications and exhaustive tests. We show how approximate models of the design can exploit AI techniques to represent the system and to derive meaningful tests, warning when the environment is not behaving as designed, detecting bugs and attacks. The representation can use Bayesian Networks that are rather simple, enabling usage by novice practitioners. We validate the methodologies with on two different applications: a device driver for Wi-Fi Direct, and a website, MindBlog.com. In the Wi-Fi Direct use case, we successfully built a test ensuring the connection is fair and contrasted it experimentally to earlier work where we created a robust Bayes network based on expert knowledge. In the MindBlog.com use case, we show that the procedure is flexible and can detect when the developers found a bug and were attempting to debug their application yielding anomalous behavior.

Analysis of Complex Non-Linear Environment Exploration in Speech Recognition by Hybrid Learning Technique

Recently, the application of voice-controlled interfaces plays a major role in many real-time environments such as a car, smart home and mobile phones. In signal processing, the accuracy of speech recognition remains a thought-provoking challenge. The filter designs assist speech recognition systems in terms of improving accuracy by parameter tuning. This task is some degree of form filter’s narrowed specifications which lead to complex nonlinear problems in speech recognition. This research aims to provide analysis on complex nonlinear environment and exploration with recent techniques in the combination of statistical-based design and Support Vector Machine (SVM) based learning techniques. Dynamic Bayes network is a dominant technique related to speech processing characterizing stack co-occurrences. This method is derived from mathematical and statistical formalism. It is also used to predict the word sequences along with the posterior probability method with the help of phonetic word unit recognition. This research involves the complexities of signal processing that it is possible to combine sentences with various types of noises at different signal-to-noise ratios (SNR) along with the measure of comparison between the two techniques.

Analysis of Complex Non-Linear Environment Exploration in Speech Recognition by Hybrid Learning Technique

Recently, the application of voice-controlled interfaces plays a major role in many real-time environments such as a car, smart home and mobile phones. In signal processing, the accuracy of speech recognition remains a thought-provoking challenge. The filter designs assist speech recognition systems in terms of improving accuracy by parameter tuning. This task is some degree of form filter’s narrowed specifications which lead to complex nonlinear problems in speech recognition. This research aims to provide analysis on complex nonlinear environment and exploration with recent techniques in the combination of statistical-based design and Support Vector Machine (SVM) based learning techniques. Dynamic Bayes network is a dominant technique related to speech processing characterizing stack co-occurrences. This method is derived from mathematical and statistical formalism. It is also used to predict the word sequences along with the posterior probability method with the help of phonetic word unit recognition. This research involves the complexities of signal processing that it is possible to combine sentences with various types of noises at different signal-to-noise ratios (SNR) along with the measure of comparison between the two techniques.

Wildfire Risk Assessment of Transmission-Line Corridors Based on Naïve Bayes Network and Remote Sensing Data

Considering the complexity of the physical model of wildfire occurrence, this paper develops a method to evaluate the wildfire risk of transmission-line corridors based on Naïve Bayes Network (NBN). First, the data of 14 wildfire-related factors including anthropogenic, physiographic, and meteorologic factors, were collected and analyzed. Then, the relief algorithm is used to rank the importance of factors according to their impacts on wildfire occurrence. After eliminating the least important factors in turn, an optimal wildfire risk assessment model for transmission-line corridors was constructed based on the NBN. Finally, this model was carried out and visualized in Guangxi province in southern China. Then a cost function was proposed to further verify the applicability of the wildfire risk distribution map. The fire events monitored by satellites during the first season in 2020 shows that 81.8% of fires fall in high- and very-high-risk regions.