We published DigiSpan paper on memory testing

🚀 Excited to share DigiSpan, an innovative tool for assessing short-term and working memory developed by Harvey Dillon Christian Boyle Shrutika Gaikwad Ponsuang Luengtaweekul and Sharon Cameron ! 🎧🧠 Macquarie University Hearing DigiSpan is a computer-administered memory test, replacing traditional digit-span tests. It includes both forward and backward span tasks, starting with simpler trials and adapting based on performance for precision and reliability. 🔍 Why DigiSpan stands out: ➕ Standardised stimuli presentation through headphones ➕ Automatic administration  ➕ Inclusion of adaptive trials for increased accuracy ➕ Monitors response consistency for reliable results ➕ Instantaneous age-appropriate standardised scoring This tool is particularly beneficial for assessing memory in individuals, especially children, with listening difficulties. It helps identify whether memory deficits may lead to their challenges, ensuring accurate diagnosis and intervention. Partners Sound Scouts Macquarie University The University of Manchester #MemoryAssessment #DigiSpan #CognitiveTesting #ChildDevelopment #Psychology

P values, confidence intervals are ubiquitously simple yet their relevance to clinical trial outcomes can be a prone to misinterpretation. Differences in the interpretation of frequentist vs Bayesian methods adds a layer of complexity when directly comparing results. Our blog post: "P-Values, Confidence Intervals, and Clinical Trials."provides an overview of P-values and confidence intervals, the correct interpretation of which is critical for interpreting clinical trial outcomes and making informed decisions. Read the full post here: (https://lnkd.in/eK2ykt7r) #Biostatistics #ClinicalTrials #FrequentistStatistics #Research #AnatomiseBiostats https://lnkd.in/esjGAae9

You may know that discrimination and calibration are key properties of clinical prediction models, but do you understand and keep them straight? After you read this paper, hugely cited for clarity and accessibility, you will return to refresh if you forget. https://lnkd.in/eyy-hPvd

**NEW pre-print** "Sample size for developing a prediction model with a binary outcome: targeting precise individual risk estimates to improve clinical decisions and fairness" + new Stata module: pmstabilityss https://lnkd.in/dtSJpeCW Feedback welcome

Clinical Reasoning in Physical Therapy. Not a novel concept. Extracting methods and mechanisms of decision-making in the clinical setting as one of the ways to benchmark and optimize performance. Rather than a one-size-fits-all mentality, we can learn clinical excellence by learning from successful traits of successful clinicians. An important read:

…but if you are a healthcare professional with experience in the care of individuals living with PCOS, it is worth it! 💪🙏 https://lnkd.in/g5AHRu9c 👉 PhD research study: review AI-generated advice for individuals living with #PCOS, and compare that to evidence-based information

Just sharing this study, as it was mentioned by Professor Kartik Hosanagar this afternoon on how the study's small sample size can't be the weight of advice which is what the authors of the study appropriately caution that the results of this study “should not be interpreted to indicate that LLMs should be used for diagnosis autonomously without physician oversight." And stated that they were fully aware the headline would be the LLM outperformed physicians, and it was. Just another instance of human-in-the-loop media inferences for that changes the conversation of trust.. where LLMs can support doctors and the use of CDSS. https://lnkd.in/gycejr7D

A new #COVID19 pub from @UUtah researchers! Check it out: Characterization of Post-COVID-19 Definitions and Clinical Coding Practices: Longitudinal Study https://lnkd.in/gvhYXj2m @UofUResearch

There are so many good reasons to use Bayesian methods in clinical trials. If I had to choose only one reason, it is that I can formally include previous knowledge in statistical analysis using a prior distribution. This is a very powerful tool to make better inferences, and when it comes to power analysis, save sponsors valuable research dollars. PS: As an added bonus, I also get to kick Null Hypothesis Significance Testing (NHST) and its annoying kid, p-values, to the curb.

Understand the importance of sample size in research! 📊 Discover how the right sample size can enhance statistical power, precision, and generalizability in your studies. Learn more in our latest blog post: https://lnkd.in/dMT-GT-d #StatisticalAnalysis #SampleSize #StatisMed #MedicalResearch