Applied Artificial Intelligence in the Subsurface

In this paper we present an example of improved approaches for how to interact with data and leverage artificial intelligence for the subsurface. Currently, subsurface workflows typically rely on a lot of time-consuming manual input and analysis, but the promise of artificial intelligence is that, once properly trained, an AI can take care of the more routine tasks, leaving the domain expert free to work on more complex and creative parts of the job. Artificial intelligence work on subsurface datasets in recent years has typically taken the form of research and proof of concept type work, with a lot of one-off solutions showing up in the literature using new and innovative ideas (e.g. Hussein et al, 2021; Misra et al, 2019). Oftentimes this work requires a good degree of data science knowledge and programming skills on the part of the scientist, putting many of the approaches outlined in these and a multitude of other papers out of reach for many subsurface experts in the Oil and Gas industry. In order for Artificial Intelligence to become applied as part of regular workflows in the subsurface, the industry needs tools built to help subsurface experts access AI techniques in a more practical, targeted way. We present herein a practical guide to help in developing applied artificial Intelligence tools to roll out within your organization or to the industry more broadly.

Using an Online Disturbance Rejection and Anticipation System to Reduce Hyperglycemia in a Fully Closed-Loop Artificial Pancreas System

Introduction: Hyperglycemia following meals is a recurring challenge for people with type 1 diabetes, and even the most advanced available automated systems currently require manual input of carbohydrate amounts. To progress toward fully automated systems, we present a novel control system that can automatically deliver priming boluses and/or anticipate eating behaviors to improve postprandial full closed-loop control. Methods: A model predictive control (MPC) system was enhanced by an automated bolus system reacting to early glucose rise and/or a multistage MPC (MS-MPC) framework to anticipate historical patterns. Priming was achieved by detecting large glycemic disturbances, such as meals, and delivering a fraction of the patient’s total daily insulin (TDI) modulated by the disturbance’s likelihood (bolus priming system [BPS]). In the anticipatory module, glycemic disturbance profiles were generated from historical data using clustering to group days with similar behaviors; the probability of each cluster is then evaluated at every controller step and informs the MS-MPC framework to anticipate each profile. We tested four configurations: MPC, MPC + BPS, MS-MPC, and MS-MPC + BPS in simulation to contrast the effect of each controller module. Results: Postprandial time in range was highest for MS-MPC + BPS: 60.73 ± 25.39%, but improved with each module: MPC + BPS: 56.95±25.83 and MS-MPC: 54.83 ± 26.00%, compared with MPC: 51.79 ± 26.12%. Exposure to hypoglycemia was maintained for all controllers (time below 70 mg/dL <0.5%), and improvement came primarily from a reduction in postprandial time above range (MS-MPC + BPS: 39.10 ± 25.32%, MPC + BPS: 42.99 ± 25.81%, MS-MPC: 45.09 ± 25.96%, MPC: 48.18 ± 26.09%). Conclusions: The BPS and anticipatory disturbance profiles improved blood glucose control and were most efficient when combined.

Pemilihan Platform Tanda Tangan Digital Berdasarkan Faktor Keberlanjutan Selama Pandemi COVID-19 Menggunakan Metode AHP

Sistem tanda tangan digital telah banyak dimanfaatkan terutama untuk kegiatan persetujuan dokumen selama pandemi COVID-19 di Indonesia. Penelitian ini bertujuan untuk mengetahui faktor-faktor dari dimensi sustainability yang mempengaruhi keberlanjutan dalam menggunakan sistem tanda tangan digital berdasarkan metode pendekatan AHP. Penelitian ini menggunakan metode AHP karena metode ini mampu menyelesaikan persoalan dalam suatu kerangka berpikir yang terorganisir sehingga dapat mengambil keputusan secara efektif dan akurat terhadap suatu persoalan dalam penelitian. Hasil penelitian menunjukkan bahwa sub-kriteria continuous improvement dari dimensi ekonomi menjadi prioritas utama yang menjadi penunjang dalam keberlanjutan bisnis penyedia tanda tangan digital. Selanjutnya, melakukan analisis pemilihan platform tanda tangan digital antara PrivyID, DigiSign, dan Manual Input. Berdasarkan hasil perhitungan terhadap tiga alternatif menghasilkan platform DigiSign yang paling unggul dibandingkan dengan platform PrivyID dan Manual Input. Hasil tersebut sejalan dengan komitmen dari platform DigiSign yang memberikan kemudahan bagi pengguna agar dapat memeriksa dokumen yang tertunda dengan cepat, menandatangani dokumen dengan tingkat keamanan yang tinggi dan enkripsi berstandar Internasional, serta dapat melacak status dokumen dengan mudah. Sedangkan tanda tangan elektronik yang dilakukan dengan cara Manual Input memiliki kelemahan yang sangat kritikal, dimana tanda tangan elektronik tersebut tidak terenkripsi sehingga tidak mampu untuk melindungi dokumen dari pencurian data identitas atau entitas perusahaan oleh pihak yang tidak bertanggung jawab. Abstract The digital signature system has been widely used, especially for document approval activities during the COVID-19 pandemic in Indonesia. This study aims to determine the factors of the sustainability dimension that affect sustainability in using a digital signature system based on the AHP approach. This study uses the AHP method because this method is able to solve problems in an organized framework so that it can take a decisions effectively and accurately on a research problem. The results of this study indicate that the sub-criteria for continuous improvement from the economic dimension is the main priority that supports the sustainability of the digital signature provider business. Furthermore, analyze the selection of digital signature platforms between PrivyID, DigiSign, and Manual Input. Based on the results of the calculation of the three alternatives, the DigiSign platform is the most superior compared to the PrivyID platform and Manual Input. This results are in line with the commitment of the DigiSign platform which makes it easy for users to quickly check pending documents, sign the documents with a high level of security and International standard encryption, and easy to tracking the document status. Meanwhile, electronic signatures made by Manual Input have a very critical weakness, where the electronic signature is not encrypted so it is unable to protect documents from theft of identity data or corporate entities by irresponsible parties.

Methodological Framework for the Development of an English-Lithuanian Cybersecurity Termbase

The aim of the paper is to present a methodological framework for the development of an English-Lithuanian bilingual termbase in the cybersecurity domain, which can be applied as a model for other language pairs and other specialised domains. It is argued that the presented methodological approach can ensure creation of high-quality bilingual termbases even with limited available resources. The paper touches upon the methods and problems of dataset (corpora) compilation, terminology annotation, automatic bilingual term extraction (BiTE) and alignment, knowledge-rich context extraction, and linguistic linked open data (LLOD) technologies. The paper presents theoretical considerations as well as the arguments on the effectiveness of the described methods. The theoretical analysis and a pilot study allow arguing that: 1) a combination of parallel and comparable corpora enable to considerably expand the amount and variety of data sources that can be used for terminology extraction; this methodology is especially important for less-resourced languages which often lack parallel data; 2) deep learning systems trained by using manually annotated data (gold standard corpora) allow effective automatization of extraction of terminological data and metadata, which enables to regularly update termbases with minimised manual input; 3) LLOD technologies enable to integrate the terminological data into the global linguistic data ecosystem and make it reusable, searchable and discoverable across the Web.

Automatic diagnosis of COVID-19 infection based on ontology reasoning

Background 2019-nCoV has been spreading around the world and becoming a global concern. To prevent further widespread of 2019-nCoV, confirmed and suspected cases of COVID-19 infection are suggested to be kept in quarantine. However, the diagnose of COVID-19 infection is quite time-consuming and labor-intensive. To alleviate the burden on the medical staff, we have done some research on the intelligent diagnosis of COVID-19. Methods In this paper, we constructed a COVID-19 Diagnosis Ontology (CDO) by utilizing Protégé, which includes the basic knowledge graph of COVID-19 as well as diagnostic rules translated from Chinese government documents. Besides, SWRL rules were added into the ontology to infer intimate relationships between people, thus facilitating the efficient diagnosis of the suspected cases of COVID-19 infection. We downloaded real-case data and extracted patients’ syndromes from the descriptive text, so as to verify the accuracy of this experiment. Results After importing those real instances into Protégé, we demonstrated that the COVID-19 Diagnosis Ontology showed good performances to diagnose cases of COVID-19 infection automatically. Conclusions In conclusion, the COVID-19 Diagnosis Ontology will not only significantly reduce the manual input in the diagnosis process of COVID-19, but also uncover hidden cases and help prevent the widespread of this epidemic.

Automatic, global registration in laparoscopic liver surgery

Purpose The initial registration of a 3D pre-operative CT model to a 2D laparoscopic video image in augmented reality systems for liver surgery needs to be fast, intuitive to perform and with minimal interruptions to the surgical intervention. Several recent methods have focussed on using easily recognisable landmarks across modalities. However, these methods still need manual annotation or manual alignment. We propose a novel, fully automatic pipeline for 3D–2D global registration in laparoscopic liver interventions. Methods Firstly, we train a fully convolutional network for the semantic detection of liver contours in laparoscopic images. Secondly, we propose a novel contour-based global registration algorithm to estimate the camera pose without any manual input during surgery. The contours used are the anterior ridge and the silhouette of the liver. Results We show excellent generalisation of the semantic contour detection on test data from 8 clinical cases. In quantitative experiments, the proposed contour-based registration can successfully estimate a global alignment with as little as 30% of the liver surface, a visibility ratio which is characteristic of laparoscopic interventions. Moreover, the proposed pipeline showed very promising results in clinical data from 5 laparoscopic interventions. Conclusions Our proposed automatic global registration could make augmented reality systems more intuitive and usable for surgeons and easier to translate to operating rooms. Yet, as the liver is deformed significantly during surgery, it will be very beneficial to incorporate deformation into our method for more accurate registration.

Deep Semi-Supervised Algorithm for Learning Cluster- Oriented Representations of Medical Images Using Partially Observable DICOM Tags and Images

The task of automatically extracting large homogeneous datasets of medical images based on detailed criteria and/or semantic similarity can be challenging because the acquisition and storage of medical images in clinical practice is not fully standardised and can be prone to errors, which are often made unintentionally by medical professionals during manual input. In this paper, we propose an algorithm for learning cluster-oriented representations of medical images by fusing images with partially observable DICOM tags. Pairwise relations are modelled by thresholding the Gower distance measure which is calculated using eight DICOM tags. We trained the models using 30,000 images, and we tested them using a disjoint test set consisting of 8000 images, gathered retrospectively from the PACS repository of the Clinical Hospital Centre Rijeka in 2017. We compare our method against the standard and deep unsupervised clustering algorithms, as well as the popular semi-supervised algorithms combined with the most commonly used feature descriptors. Our model achieves an NMI score of 0.584 with respect to the anatomic region, and an NMI score of 0.793 with respect to the modality. The results suggest that DICOM data can be used to generate pairwise constraints that can help improve medical images clustering, even when using only a small number of constraints.

Cyber Threat Hunting

Security and Information Event Management (SIEM) systems require significant manual input; SIEM tools with machine learning minimizes this effort but are reactive and only effective if known attack patterns are captured by the configured rules and queries. Cyber threat hunting, a proactive method of detecting cyber threats without necessarily knowing the rules or pre-defined knowledge of threats, still requires significant manual effort and is largely missing the required machine intelligence to deploy autonomous analysis. This paper proposes a novel and interactive cognitive and predictive threat-hunting prototype tool to minimize manual configuration tasks by using machine intelligence and autonomous analytical capabilities. This tool adds proactive threat-hunting capabilities by extracting unique network communication behaviors from multiple endpoints autonomously while also providing an interactive UI with minimal configuration requirements and various cognitive visualization techniques to help cyber experts quickly spot events of cyber significance from high-dimensional data.

Application of Digital Well Construction Planning Tool During Well Conceptualization Phase

Digital well construction tools are becoming more widely considered today for well design planning, enabling automated engineering and simultaneous team collaboration under a single solution. This paper shows the results of using a digital well construction planning solution during a project’s conceptual planning stage. This method shortens the time needed to estimate the well times and risk profile for a drilling campaign by applying smart engines to quickly and accurately perform critical offset analysis for defined well types that is required for project sanction. With this solution, the Offset Well Analysis (OWA) process is done automatically based on the location of the planned well, trajectory and well architecture. Various information and reports (both subsurface and surface data) from neighboring wells is stored in cloud solutions, enabling ease of access and data reliability for both large or smaller scale data storage. The software selects the most relevant offset wells, displays the risk analysis and generates the stick chart. For a conceptual design, the risk levels can be manually set higher due to potential unknowns in surface and subsurface risks which can later be refined. Quick validation of the well design allows the engineer to design a conceptual drilling campaign quickly and more efficiently. The solution minimizes the time to perform probabilistic time and risk estimations. It reduces the risk of biased decision making due to manual input and design. This allows for better-informed decisions on project feasibility, alignment of stakeholders, increased design reliability as well as reducing the amount of time and resources invested in OWA. The work presented here is aimed at sharing the experience of applying a digital well construction planning solution specifically on the conceptual project stage and discuss the value it adds to the well design process.

An electric power system object model creating experience for researching the operation of digital means of relay protection and automation

The article is devoted to the operation logic modeling of relay protection and automation terminals in order to their verification, adjustment and further exploitation. The problem of adjusting protection terminals mutual interaction is unlikely to appear in real conditions due to wide variety of them. The authors propose a solution to this problem by creating a verified model based on a digital twin of an electric power network section created in the MatLab software package. This model helps to study the functioning of the researched protection settings in nominal, repair, emergency and post-emergency equipment operation modes. A model of the selected substation was created displaying all the properties that are significant for research of the original one. In addition, the requirements analysis for the main and backup protection operation settings of the three-winding transformers was carried out. The main unit is a differential transformer relay protection and the backup one is maximal current protection in amount of three units for every transformer winding circuit: higher, middle and lower transformer voltage branch. The model makes it possible to analyze the relay protection operation selectivity by checking the current settings which could be imported from XML documents unloaded from existing terminals and to evaluate the correctness of new calculated ones with the possibility of their manual input. As a result of the researched object modeling, a three-stage operation analysis of the differential and maximal current protections was carried out. It has shown relay protection selective operation both in the case of nominal and abnormal modes, including the event of the main transformer protection malfunction. This technique can be extended to the other electric power network.