Data Science Platform Research Articles

Modified Metaheuristics Optimization for Cyberbullying Detection on Online Data Science Platform

Online harassment detection faces significant challenges due to its expansive reach and anonymity. Addressing this issue demands effective detection mechanisms capable of processing vast data streams and adapting to evolving online language. Leveraging advancements in artificial intelligence, we propose a novel approach grounded in natural language processing (NLP) and metaheuristic algorithms. Our methodology integrates term frequency-inverse document frequency (TF-IDF) encoding and the AdaBoost algorithm for classification. To tackle the NP-hard problem of hyperparameter selection, we introduce a modified crayfish optimization algorithm (COA), termed GI-COA. This paper represents a pioneering effort in utilizing metaheuristic algorithms for hyperparameter selection in harassment detection models. Through experimentation, we demonstrate the efficacy of our approach in fostering a safer online environment. The best performing optimize models demonstrate and accuracy exceeding 77%.

From Fetus to Eight: the CHILD Cohort Study.

The CHILD Cohort Study is an active multi-center longitudinal, prospective, population pregnancy cohort study following Canadian infants from fetal life until adulthood. We hypothesized that early life physical and psychosocial environments interact with biological factors (e.g. immunologic, genetic, physiologic, and metabolic) influencing burdensome non-communicable disease outcomes, including asthma and allergic disorders, growth and development, cardio-metabolic health, and neurodevelopmental outcomes that manifest during the life-course. Detailed clinical and physiologic phenotyping at strategic intervals was complemented by environmental sampling, actigraphy and global positioning system measures, biological sampling including gut, breastmilk and nasal microbiome, nutritional studies, genetics, and epigenetic profiling. Of 3,454 families recruited from 2008 to 2012, study retention was 96.0% at 1-year, 93.2% at 5-years and 90.7% at 8-years. Data collection during the SARS-2 COVID-19 pandemic was partially completed via virtual visits. A sub-cohort was implemented, capturing detailed information on the prevalence and predictors of SARS-CoV-2 infection and the health and psychosocial impact of the pandemic on Canadian families. The 13-year clinical assessment launched in 2022 will be completed in 2025. Ultimately, the CHILD Cohort Study provides a data science platform designed to enable a deep understanding of early life factors associated with the development of chronic non-communicable diseases and multimorbidity.

Forme Fruste Keratoconus Detection with OCT Corneal Topography Using Artificial Intelligence Algorithms.

Distinguishing early Keratoconus (KC) from normal corneas is challenging owing to their striking similarities. The aim of our study was to identify discriminating parameters to differentiate a normal cornea from a Form Fruste Keratoconus (FFKC) with the Swept-Source (SS) OCT-topography CASIA 2 (Tomey,Japan) using machine learning artificial intelligence algorithms. The study was monocentric, carried out in Bordeaux. This was a retrospective study, case control. Three groups were included: KC group (108 eyes), FFKC (88 eyes) and normal corneas (162 eyes).The data were analyzed and processed using the Dataiku data science platform. Machine learning models (Random Forest, Logistic Regression) were used to develop a multiclass classifier for automated early KC detection. The models were trained using a training database and tested using a test database. Then algorithms were compared to the Ectasia Screening Index (ESI), which is an OCT-topography inherent screening score for ectasia. The Logistic Regression (LR), and Random Forest (RF) detected FFKC with an AUC of 0,99, and 0,98 respectively. The sensitivities of LR (100%), RF (84%) were better than the ESI (28%) for the diagnosis of FFKC. However, ESI has a maximum specificity (100%) compared to the RL (100%) and 90% for RF. This study identified discriminating topographic parameters to be considered in refractive surgery screening on SS-OCT CASIA 2. We developed an algorithm capable of classifying normal eyes versus FFKC cases, with improved performance compared to the ESI score.

Data Science with Data Mining and Machine Learning A design science research approach

Access to large amounts of data, provided by continuous technological progress, becomes increasingly important for organisations. The access to data becomes even more useful when managers consult structured information in real time. This requires the adoption of data analysis techniques that, at a strategic level, allow the definition of business goals according to pre-defined objectives, and which help companies in their functional activities. With Data Science, Data Mining and Machine Learning platforms allow the development of decision support systems, capable of establishing patterns between data and making truly useful predictions for decision support. The methodology used in this work aims to combine Data Science methods, such as Data Mining and Machine Learning, in order to demonstrate the importance of transforming data into organised and oriented information to support decision making.

Elevating the RRE Framework for Geospatial Analysis with Visual Programming Platforms: An Exploration with Geospatial Analytics Extension for KNIME

Reproducibility, replicability, and expandability (RRE) have emerged as fundamental concerns in the realm of scientific research and development. Wherein, devising effective solutions for RRE within geospatial analysis stands out as a particularly critical challenge that demands immediate attention. Although there has been an evolution from basic reproducibility of code and data to a more comprehensive cyberinfrastructure, this integrated solution is still grappling with issues of limited user accessibility, steep learning curves particularly in coding skills, and difficulties in achieving collaboration with other data science platforms This study proposes a framework that combines open-source GIS with visual programming platforms, grounded in principles of standardization and educationalization, to advance the RRE framework in geographic analysis. Using the Geospatial Analytics Extension for KNIME as an example, we demonstrate the platform’s adaptability and utility through case studies in a recent textbook with an in-depth illustration of spatial accessibility analysis, specifically via the Generalized Two-Step Floating Catchment Area (G2SFCA) method. Our findings shed light on the transformative potential of such an integrative strategy, offer fresh perspectives for enhancing the RRE in geospatial analysis and craft a well-structured, intuitive, and extensive GIS knowledge tree.

Psychiatryai.com: Real-Time AI Scoping Review (RAISR 4D) in Psychiatry and Mental Health with Live Real-World Evidence and CPD/CME for Psychiatrists

IntroductionPsychiatryai.com was launched in 2021 and initial findings were published at EPA 2023. The portal is an advanced computing science project in Applied Data Science and Evidence-Based Healthcare for my MSc studies at the University of Oxford (Kellogg College). Artificial Intelligence (AI) and Data Science (DS) technology are utilised to analyse live Real-World Evidence (RWE) in Psychiatry and Mental Health from PubMed to provide CPD/CME online. A two-year review of the site and its performance will be presented to EPA 2024.ObjectivesTo develop and study an experimental real-time AI and DS platform in Global Mental Health and Psychiatry, to provide the latest RWE from PubMed for online education and training, and to report findings to EPA 2024 for peer review in Budapest. AI and misinformation are newly identified risks in healthcare (AI Safety Summit 2023). The site also aims to raise awareness about “Aiatrogenesis” to address this problem, with RWE and CPD/CME utilising AI and DS technology for the categorisation and meta-analysis of evidence, rather than the production of possibly misleading or false Generative AI evidence (Monteith et al. BJP 2023; 1-3).MethodsAs reported to EPA 2023 in Paris, a free open-code WordPress site was launched on the 22nd of November 2021 (Psychiatryai.com). The portal has been further developed and now features over 90k pages comprising 7GB of data with Cloudflare security and speed. Live evidence is collected into an open database and research articles are categorised into evidence nodes with AI. The results are presented in a real-time Evidence Matrix and Blueprint, creating 15-minute CPD/CME reflection modules. Data analytics from Psychiatryai.com with Google Analytics (G4A) along with platform insights from two years of development and research will be presented to EPA 2024. The site is conceptualised and designed to be viewed in an interactive VR headset.ResultsLive Citations380000+PubMed Articles Analysed with AI92142CPD/CME23002 hoursAlgorithms/Topics in Psychiatry291Open Data7 GBImage:Image 2:Image 3:ConclusionsPsychiatryai.com has successfully developed a novel AI and DS platform that incorporates the latest research in mental health and psychiatry, providing real-world evidence (RWE) for psychiatrists and healthcare professionals worldwide, along with CPD/CME online. This enhances hypothesis testing in research by presenting a related Evidence Matrix and Blueprint (from the last 365 days) for each evidence node on the site (RAISR 4D). These matrices provide a real-time visual table (8 x 8 / 64) of global research related to the evidence node in the preceding year. The site is VR-ready and has a special focus on AI and Psychiatry, Disaster and Traumatology Sciences, and Youth Mental Health. This project is dedicated to the memory of Dr Denis O’Leary and Dr Navin Venkatraman.Disclosure of InterestNone Declared

Nursing workload: use of artificial intelligence to develop a classifier model.

to describe the development of a predictive nursing workload classifier model, using artificial intelligence. retrospective observational study, using secondary sources of electronic patient records, using machine learning. The convenience sample consisted of 43,871 assessments carried out by clinical nurses using the Perroca Patient Classification System, which served as the gold standard, and clinical data from the electronic medical records of 11,774 patients, which constituted the variables. In order to organize the data and carry out the analysis, the Dataiku® data science platform was used. Data analysis occurred in an exploratory, descriptive and predictive manner. The study was approved by the Ethics and Research Committee of the institution where the study was carried out. the use of artificial intelligence enabled the development of the nursing workload assessment classifier model, identifying the variables that most contributed to its prediction. The algorithm correctly classified 72% of the variables and the area under the Receiver Operating Characteristic curve was 82%. a predictive model was developed, demonstrating that it is possible to train algorithms with data from the patient's electronic medical record to predict the nursing workload and that artificial intelligence tools can be effective in automating this activity.

Autonomous Driving of a Rover Based on Traffic Signals and Signs

The traditional driving system has several disadvantages such as human error, driver fatigue and the inability to handle complex situations. These limitations make traditional driving unsafe and unreliable, leading to accidents and traffic congestion. The necessity for Autonomous Driving of a Rover based on Traffic Signals & Signs is to address these issues by automating the driving process and making it safer and more efficient. A dataset with traffic signs will be used to train a deep-learning model for classifying signs. A transfer learning technique will be used to deploy the trained model on the rover, considering hardware limitations. A camera on the rover captures images and sends them to the model for classification, enabling autonomous navigation based on traffic signs. The required software for the project includes Anaconda, a popular data science platform, and MaixPy, which is a version of MicroPython specifically designed for the Kendryte K210 chipset. The hardware required for the system includes the Zumo Shield for Arduino, which serves as the interface between the rover and the computer vision software, the Maixduino board, which is used to process the image data, and batteries to power the system. The system is designed to detect traffic signs and signals in real-time and respond accordingly, enabling the rover to navigate through traffic safely and efficiently.

Developing a data science platform for pipeline applications Tennessee Eastman

Developing a data science platform for pipeline applications Tennessee Eastman

Emerging technologies for assessing ecosystem services: A synthesis of opportunities and challenges

Rapid technological development opens up new opportunities for assessing ecosystem services (ES), which may help to overcome current knowledge gaps and limitations in data availability. At the same time, emerging technologies, such as mobile devices, social media platforms, and artificial intelligence, give rise to a series of challenges and limitations. This study provides a comprehensive overview of the broad range of technologies that are increasingly used for collecting, analyzing, and visualizing data on ES, including Earth observation, mobile technologies, data science, modeling/simulation, immersive technologies, visualization, and web-based tools. To identify current opportunities and challenges, we systematically reviewed the literature on ES of the last 10 years (2012–2022). We first describe the state-of-the-art of emerging technologies, synthesizing their applicability, opportunities, and limitations. Then, we discuss open issues, future research needs, and potential for further applications in ES research. Our findings indicate great opportunities to increase knowledge on ES thanks to low costs, high data availability, and high flexibility of the technologies. We also find a strong potential to support decision-making, learning and communication. However, limitations and challenges related to the accuracy of the variables and models, accessibility to data, technologies, and information as well as ethical concerns need to be addressed by the research community to assure an inclusive and meaningful use of the technologies. Our findings also suggest that further insights into ES may be achieved through a better integration of different technologies in the future, e.g., Earth observation, data science, and web-based platforms, a stronger transdisciplinary collaboration to advance knowledge on ES, and by broadening the perspective to technological developments in other related fields of research.

DEVELOPING NEW SUGGESTIONS FOR THE CONTENTS OF A DIGITAL PLATFORM USING RECOMMENDATION SYSTEMS ALGORITHMS

In recent years, machine learning applications are being used in almost all areas of lives. The main benefits of using machine learning in marketing can be exemplified as follows; content creation, marketing budget optimization and recommendation systems. Recommendation systems are very important and useful when it comes to retaining the current customer. With the help of recommendation systems, companies can retain their customers by recommending their own products, services and contents. In this study, text mining, forecasting processes were carried out using the Netflix contents dataset shared by the data science platform called Kaggle. TfidVectorizer function was used to deal with text data while creating recommendation systems. Two different recommendation systems functions were created in this study. While first recommendation system function performs only based on title feature of the Netflix contents dataset, the second recommendation system function performs with title, director, cast, listed_in and description features. Thanks to the results of the analysis, it is possible to evaluate the new productions on Netflix on the basis of the features of Netflix contents dataset included in the study. The proposed recommendation system functions provide greater prediction accuracy than conventional systems in data mining. Espicially the recommendation system function that has been developed secondly with the name “get_recommendation_new” uses all features in Netflix contents dataset to recommend new contents to the users.

STUDY OF THE RELATIONSHIP OF CITIZENS’ APPEALS TO THE OPEN PORTAL OF THE GOVERNMENT OF MOSCOW WITH THE SOCIO-ECONOMIC INDICATORS OF THE ADMINISTRATIVE DISTRICT

The article discusses the possibilities of studying the state of the social sphere according to the repository of the open data portal of the Government of Moscow for administrative districts and urban districts using business intelligence platforms and intelligent technologies of Data Science and machine learning platforms. The possibilities of using machine learning technologies for business intelligence platforms to identify hidden patterns in order to make informed management decisions are presented.
 Purpose – is to familiarize with business intelligence platforms; the use of intelligent BI-platform technologies for monitoring socio-economic indicators.
 Method or methodology of the work: the clustering method and statistical research methods were used in the article.
 Results: to confirm the hypothesis of a positive relationship between the number of families receiving subsidies in the city of Moscow, in various sections, and the number of appeals of citizens to the open portal of the Government of Moscow
 Practical implications: the objective information obtained as a result of monitoring can be used to develop a strategy and make management decisions on the development of districts and improving the quality of life in Moscow

Developing an Artificial Intelligence-Based Representation of a Virtual Patient Model for Real-Time Diagnosis of Acute Respiratory Distress Syndrome.

Acute Respiratory Distress Syndrome (ARDS) is a condition that endangers the lives of many Intensive Care Unit patients through gradual reduction of lung function. Due to its heterogeneity, this condition has been difficult to diagnose and treat, although it has been the subject of continuous research, leading to the development of several tools for modeling disease progression on the one hand, and guidelines for diagnosis on the other, mainly the "Berlin Definition". This paper describes the development of a deep learning-based surrogate model of one such tool for modeling ARDS onset in a virtual patient: the Nottingham Physiology Simulator. The model-development process takes advantage of current machine learning and data-analysis techniques, as well as efficient hyperparameter-tuning methods, within a high-performance computing-enabled data science platform. The lightweight models developed through this process present comparable accuracy to the original simulator (per-parameter R2 > 0.90). The experimental process described herein serves as a proof of concept for the rapid development and dissemination of specialised diagnosis support systems based on pre-existing generalised mechanistic models, making use of supercomputing infrastructure for the development and testing processes and supported by open-source software for streamlined implementation in clinical routines.

Impact of COVID-19 pandemic on the treatment of prostate cancer in the Brazilian public health system.

e17106 Background: COVID-19 pandemic reached Brazil in the beginning of 2020 causing a profound impact in economic and social activities, up to the end of 2021. In Brazil, there is universal access to the public health system (BPHS-SUS); however, during the pandemic public resources were directed mainly to the management of COVID-19 infected patients. We evaluated the impact of the 2020/2021 pandemic on the treatment of prostate cancer in the BPHS-SUS. Methods: we compared data regarding the number of radical prostatectomies, radiotherapy treatments, and hormonal and chemotherapy treatments administered in the time frame from March 1st, 2020 (when lockdown measures were initially adopted in Brazil) to February 29th, 2022, with that of the two most recent pre-pandemic years (March 1st, 2018, to February 28th, 2020). A large proprietary healthcare data science platform offered by TECHTRIALSÒ was surveyed. Results: in the two pandemic years, 15,196 radical prostatectomies were performed in the BPHS, with a cost of U$ 14,271,297; there was a 23.3% reduction in the number of surgeries compared with the two pre-pandemic years (19,818 radical prostatectomies, with a cost of U$ 18,056,470). In the pandemic, 34,982 men were treated with radiation therapy, with a cost of U$ 39,616,736; this meant a 18.7% reduction in the number of patients treated with radiation therapy as compared with the two pre-pandemic years (43,013 men, with a cost of U$ 44,148,190). Regarding hormonal treatments, numbers were similar: 116,490 men were treated during the pandemic (cost of U$ 74,834,615); in the pre-pandemic timeframe, 115,949 patients received hormonal therapy (cost of U$ 75,179,674). As to chemotherapy for hormone-resistant disease, in the pandemic 14,946 men were treated (cost of U$ 23,402,947) compared with 12,695 men treated in the two pre-pandemic years (cost of U$ 19,239,099). Conclusions: there was a significant reduction in the number of radical prostatectomies and of radiation therapy treatments during the pandemic years in the BPHS-SUS. Abrupt reductions of potentially curative therapies will probably affect stage migration, costs to the public system and patients’ survival in the ensuing years. The number of systemic therapies (first- and second-line hormonal therapies, and chemotherapies) were slightly increased, probably reflecting the BPHS-SUS policy of maintaining systemic treatments active during the pandemic.

Prostate cancer public health system database from Brazil: Racial disparities and heterogeneity of treatment interventions to understand evidence-based practice patterns in a large real-world database.

5025 Background: The Brazilian Public National Health System (SUS) provides an opportunity to understand epidemiologic information, treatments, discrepancies in care, and costs of men with prostate cancer (PCa) using a large real-world dataset. This analysis focuses on racial disparities and heterogeneity of specific interventions to understand evidence-based practice patterns in Brazil. Additionally, the dataset can serve as a model for other low-middle income countries (LMIC). Methods: Patients included in the database were those with a diagnosis of PCa receiving care in the SUS from 01/01/2008 to 08/31/2022. Data were analyzed using a proprietary Healthcare Data Science platform offered by TECHTRIALS and included information regarding stage, demographics, cancer-related information, treatment-related information, and resources utilized. Results: 657,609 PCa pts were diagnosed and treated at SUS during the 14.5-year study period of whom 382,263 (58%) were brown/black (non-white) and 262,820 (42%) white. Median age of the population was 72.6 years, and did not vary with race. 26.16% of the pts were diagnosed with stage IV and this was not impacted by race or localization, including state or type of city (capital or not). Rates of radical prostatectomy and radiation utilization were similar between race categories. Additionally, in patients with stage IV disease, rates of systemic therapy utilization including chemotherapy with docetaxel (7.7%), with a tendency to increase use in the last years, and androgen receptor signaling inhibitors (ARSI) with (0.6%) or enzalutamide (0.08%), were low and similar between race categories. Rates of first-generation anti-androgens and estrogens were higher (56%) The total costs of prostate cancer directed care was R$4.31 billions (US$820 millions) and race did not impact total cost of care and cost by procedure or therapy type options. Conclusions: This is one of the largest studies using public health-system databases for Prostate cancer with Real World Data ever published. Race had no impact in presenting stage, treatment received and even in the costs involved. Underutilization of chemotherapy (docetaxel) in a scenario of limiting resources and without the use of new hormonal therapies is a matter of concern, together with an overuse of agents without proven survival advantage. Our data highlights the need for increased access to life prolonging agents for patients with metastatic PCa in the SUS.

Building a collaborative cloud platform to accelerate heart, lung, blood, and sleep research.

Research increasingly relies on interrogating large-scale data resources. The NIH National Heart, Lung, and Blood Institute developed the NHLBI BioData CatalystⓇ (BDC), a community-driven ecosystem where researchers, including bench and clinical scientists, statisticians, and algorithm developers, find, access, share, store, and compute on large-scale datasets. This ecosystem provides secure, cloud-based workspaces, user authentication and authorization, search, tools and workflows, applications, and new innovative features to address community needs, including exploratory data analysis, genomic and imaging tools, tools for reproducibility, and improved interoperability with other NIH data science platforms. BDC offers straightforward access to large-scale datasets and computational resources that support precision medicine for heart, lung, blood, and sleep conditions, leveraging separately developed and managed platforms to maximize flexibility based on researcher needs, expertise, and backgrounds. Through the NHLBI BioData Catalyst Fellows Program, BDC facilitates scientific discoveries and technological advances. BDC also facilitated accelerated research on the coronavirus disease-2019 (COVID-19) pandemic.

Medical School Admissions: Focusing on Producing a Physician Workforce That Addresses the Needs of the United States.

The aging population, burnout, and earlier retirement of physicians along with the static number of training positions are likely to worsen the current physician shortage. There is an urgent need to transform the process for selecting medical students. In this Invited Commentary, the authors suggest that to build the physician workforce that the United States needs for the future, academic medicine should focus on building capacity in 3 overarching areas. First, medical schools need to develop a more diverse pool of capable applicants that better matches the demographic characteristics of health care trainees with those of the population, and they need to nurture applicants with diverse career aspirations. Second, medical schools should recalibrate their student selection process, aligning criteria for admission with competencies expected of medical school graduates, whether they choose to become practicing clinicians, physician-scientists, members of the public health workforce, or policy makers. Selection criteria that overweight the results of standardized test scores should be replaced by assessments that value and predict academic capacity, adaptive learning skills, curiosity, compassion, empathy, emotional maturity, and superior communication skills. Finally, to improve the equity and effectiveness of the selection processes, medical schools should leverage innovations in data science and generative artificial intelligence platforms. The ability of ChatGPT to pass the United States Medical Licensing Examination (USMLE) demonstrates the decreasing importance of memorization in medicine in favor of critical thinking and problem-solving skills. The 2022 change in the USMLE Step 1 to pass/fail plus the exodus of several prominent medical schools from the U.S. News and World Report rankings have exposed limitations of the current selection processes. Newer approaches that use precision education systems to leverage data and technology can help address these limitations.

A clinical site workload prediction model with machine learning lifecycle

In clinical trial monitoring, substantial resources are allocated to perform source data verification (SDV). SDV ensures accurate and reliable transcription of trial participant information. Clinical site visits are typically scheduled at a fixed frequency for SDV, without objectively factoring in individual site workload. This often results in wasted resources and directly influences clinical trial cost. We leveraged historical data from several hundred clinical trials to predict SDV workload at trial sites using machine learning. Specifically, we adopted cross industry standard process for data mining (CRISP-DM) process model and devised a novel deep learning pipeline for longitudinal clinical trial data. The deep learning pipeline, which comprises recurrent neural network-based encoder and decoder, ingests multivariate sequence data from study sites and predicts SDV workload for future months. We also developed an efficient model enhancement workflow, in a data science platform that facilitates machine learning operations best practices, for timely adaptation of new features and changes. Several enhancement iterations have been performed since the launch of first SDV workload prediction model, resulting in a more accurate latest model compared to previous versions. We discuss these enhancements in the context of CRISP-DM phases. In conclusion, the SDV workload prediction model has enabled informed planning and optimization of resources within clinical trial operations.

Psychiatryai.com: An exploratory online Artificial Intelligence and Data Science platform to enable near real-time psychiatry and mental health evidence-based medical research and data dissemination

IntroductionPsychiatryai.com is a prototype Artificial Intelligence (AI) and Data Science (DS) platform and research project developed for my Evidence-Based Healthcare (EBHC) course at University of Oxford in MSc studies (Kellogg College). This is a singular, multi-disciplinary, and beta-testing project in Computing Science, Psychiatry, and Mental Health for oral presentation at EPA 2023.ObjectivesAI and DS in Psychiatry and Mental Health have emerged as important research areas in the post Covid-19 pandemic era. This prototype University project (Psychiatryai.com) was launched on 22nd November 2021. It aims to develop a free, secure, and open access platform in near real-time about psychiatry and mental health evidence-based research - for healthcare professionals, doctors, and researchers in psychiatry. The project also aims to integrate findings from the Goldacre Review (2022) into practice and develop novel computing solutions utilising AI and DS, and present findings.MethodsA WordPress site (Psychiatryai.com) was developed with syndicated RSS feeds across 330 psychiatry topics and refreshed by data servers hourly, 24 hours a day, 7 days a week. A total of 43 WordPress plugins were utilised to develop this secure platform. The site is powered by intuitive data modelling and analytics in near real-time and available in open access coding format for peer-review, future development, and research. The primary sources of live evidence for the project are PubMed and University of Helsinki, Finland. The server performance data analytics will be available for poster presentation at EPA 2023. This includes full statistical results and discussion since its inception and launch (including traffic, MESH tags, and PubMed ID), and robust technical analysis and performance outcomes - available freely online to promote research in psychiatry and mental health.ResultsKnowledge Synthesis and Dissemination:Total Words: 4356886 *Live Psychiatry and Mental Health Citations from PubMed: Exceeds 325000Total Evidence Alerts Published: 54391 *Total Algorithms/Topics: 330Total site visitors: 8023 ** Since launch of Psychiatryai.com on 22 November 2021 inclusive to 31 October 2022ConclusionsPsychiatryai.com was able to demonstrate succesful development of an effective and viable platform to study AI and DS in Psychiatry and Mental Health, as evidenced by results table. The platform has also incorporated findings from the Goldacre Review (2022) and aims to continue to collect valuable insights towards full real-time data analytics and dissemination of peer-reviewed current evidence in the future. The emergence of these technologies will be useful in settings such as disaster psychiatry, psychiatry e-training and research, and e-mental health awareness/promotion ahead.Disclosure of InterestP. Naik Consultant of: Dr Paras Naik (Psychiatryai.com) Non-profit University Project

Study on the Optimal Reaction Parameters of C4 Olefin Prepared by Ethanol Catalyzed Reaction Based on XGBoost Regression Prediction Algorithm

The production of C4 olefin by ethanol coupling catalytic reaction has a broad development prospect in the field of renewable energy. In this paper, the influencing factors of C4 olefin preparation process were studied, the influence of temperature and catalyst type on the reaction was analyzed, and the multivariate nonlinear regression fitting model was established under the assumption of ignoring the small probability of experimental error. Then, based on the grey correlation analysis model, the correlation degree between each variable and ethanol conversion rate and C4 olefin selection rate was studied.Then, using the catalyst parameter variables considered above, the relationship equations between four catalyst parameter variables, temperature and C4 olefin yield were calculated, and the overall constructor model of C4 olefin yield under the joint action of five variables was established by using the constructor method. Finally, using MPai data science platform software based on XGBoost regression prediction algorithm, the established constructor was cross-validated by 50% fold, and the optimal solution of reaction parameters was finally obtained.Finally, five experiments were selected to improve the experimental control group data and verify the accuracy of the theoretical value of the optimal solution. By changing the catalyst parameters and temperature, a relative control group was added to the experimental group, and the conclusion of the theoretical parameter optimal solution under unrestricted and temperature restrictions was experimentally verified.