EDI Global’s Post

The vastness of #bigdata and #realworlddata can be daunting. Designing studies to minimize bias, either when estimating causal effects (e.g. https://lnkd.in/g4XzMSTB) or correlations, are often discussed and, if designed properly, will result in outcomes that could benefit patients. However, the key to unlocking the true potential of #bigdata always starts with: ▶ Are you asking the right question? ▶ Has your question been answered already? ▶ Have you assured good provenance for your data source(s)? Get in touch with our team at 📣 DiscoverNOW@imperialcollegehealthpartners.com, whether you are looking to be in a better position at the start of these questions, or if you are in need of study design support or #realworlddata.

🚀Glad to share that my research paper, "PCA-Based Dimensionality Reduction for IoMT Prenatal Data Analytics," which I presented at an IEEE conference in April this year, is now published on IEEE Xplore! 📑🎉 This milestone represents a significant step in the journey of integrating advanced analytics into prenatal healthcare.🌟📊 Dive in and explore the details here: https://lnkd.in/gdupZiB8 #Research #IEEE #IoMT #DataAnalytics

Replicability is at the centre of open data and open science. In today's featured #DigitalDiscovery article, Hattrick-Simpers et al. attempt to replicate a selection of results related to inorganic property prediction. In response to the challenges they faced, they propose specific action items for the design of accessible materials informatics tools. Read the #openaccess paper here: https://lnkd.in/eUSgYGDk

Do you want to join dynamic workshop about SICRIT- Plasmion ion source technique?? listen to the scientists users over the world 🌎 talking about real experience, asking questions and see how it works.. Don't miss the event... #massspectrometry #ionsource #ionization #lcmsms #plasmion

Interesting article presenting some recommendations for accurate reporting in medical research statistics Some points recommended: - Report mean and SD or median and IQR for quantitative variables depending on distribution. - Include histograms or tables for variable visualization. - Validate model assumptions - Display precise p values - Avoid stating results as no effect - Interpret results based on clinical significance, with 95% CIs. - Identify confounders using causal directed acyclic graphs, not significance tests. - Address missing data using methods beyond simple record exclusion. - Handle sparse-data bias in ratio estimates using appropriate methods. - Use risk ratios or risk differences for high outcome frequency. - Evaluate additive interactions even in multiplicative models. Read more on: https://lnkd.in/dQa_bcQW The Lancet #statistics #medicalresearch #modeling #datascience

Read #FeaturePaper: "Criticality Analysis: Bio-Inspired Nonlinear Data Representation" by Tjeerd V. Olde Scheper. See more details at: https://lnkd.in/ghXPg73r #bioinspiredcomputing #machinelearning #selforganisedcriticality

𝐂𝐮𝐫𝐢𝐨𝐮𝐬 𝐚𝐛𝐨𝐮𝐭 𝐃𝐚𝐧𝐢𝐬𝐡 𝐫𝐞𝐚𝐥-𝐰𝐨𝐫𝐥𝐝 𝐝𝐚𝐭𝐚? 🔍 Here are some of the things that makes our data unique: 1. Fast access to comprehensive data In January 2024, the processing time at the Danish Health Data Authority decreased to just 34.9 days, providing swift access to a wealth of national-level data. 2. Rich clinical quality databases Denmark boasts 80 nationwide clinical quality databases, offering invaluable insights beyond administrative registries. These databases include detailed information on symptoms, clinical verifications, imaging, prognostic parameters, and lifestyle risk factors. 3. Utilization of national registries Leveraging national registries, researchers can extract populations for patient record identification, access biological material, and gather data from various registry types. 4. Seamless integration with research machines Researchers in Denmark have the advantage of uploading their own data and seamlessly merging it with register data on dedicated research machines. 5. Continuous expansion of database access Denmark is committed to enhancing research opportunities by consistently making new databases available to researchers, such as the national children database. Get in touch for more information on how Quantify Research can help on your projects! Quantify Research info@quantifyresearch.com www.quantifyresearch.com

📝 A concise and informative 2-pager to introduce the Quantum project which is developing a data quality label for the secondary use of #healthdata: 🔗 https://t.co/MIgx28W31y

Data Sharing! Over the past several years, there has been greater recognition of the importance of data sharing in research. The benefits of broad data sharing include permitting secondary data analysis, optimizing resource use, avoiding unnecessary study duplication, ensuring reproducibility, and increasing statistical power. However, researchers may be reluctant to share their data, and current mechanisms for sharing human data through Data Access Committees generally rely on manual workflows. Recent papers published in Biopreservation and Biobanking explore the critical issue of incentives for data sharing and the use of automated decision support tools to enhance the efficiency and effectiveness of decision-making by Data Access Committees for access to data. Read this collection on data sharing in BIO #datasharing #biobanking #FAIR #benefits #dac https://lnkd.in/e2cahSzs

[ICIS 2024: Call for Papers Deadline – May 1] Data Analytics for Business and Societal Challenges Track chairs: Jingjing Li, Esko Penttinen, Chuan-Hoo Tan This track seeks to gather contemporary research works that have applied and/or developed novel data science and analytics methods, techniques, algorithms, and theories, to address challenging and real-world problems. We are particularly interested in developing research works that would benefit and impact how we do business, live, and socialize. While we do not limit the spectrum of domains, some of the suggested domains could be small and medium-sized enterprises, digital business and innovation, healthcare and wellness, community-based social services, judicial system and democracy, fraud and deception, social cohesion and inclusion, privacy and security, social media and politics, sustainability and climate change, etc. Contributions may be motivated by the shortcomings of state-of-the-art approaches in addressing practical challenges or may apply novel data science tools to existing problems. This track is open to all types of research, including conceptual, theoretical, analytical, and/or empirical. Submit your research by May 1 and join ICIS 2024!! #InformationSystems  #ICIS2024 #DataScience #Analytics #dataanalytics See more info. https://lnkd.in/gq47KZBq