PROMETRIKA's Statistical Programming Team weighs in on CDISC's new Analysis Results Standard (ARS) initiative: https://lnkd.in/dejxW8g5
PROMETRIKA, LLC’s Post
More Relevant Posts
-
I'm happy to announce the publication of our paper, "Predicting Subgroup Treatment Effects for a New Study: Motivations, Results, and Learnings from Running a Data Challenge in a Pharmaceutical Corporation". This paper summarises a data challenge we conducted in 2021/22, focusing on the topic of subgroup identification. The challenge was a success, bringing together 30 teams and around 100 participants. It wasn't just a competition but a comprehensive learning journey, introducing participants to advanced data science tools (git, R, Python) and providing hands-on experience in a new SCE. The collaborative effort yielded a number of insights, particularly in addressing treatment effect heterogeneity. Key Takeaways: ❇ Participants enjoyed the blend of competition and learning, enhancing their understanding of statistical/machine learning methods. ❇ Most participants gained firsthand experience with cutting-edge data science tools, marking their initiation onto a new SCE platform. ❇ The challenge helped develop further methodological insights into practical guidance on treatment effect heterogeneity. We invite you to learn more on our motivations, the challenge's execution, and our learnings by reading the full paper here https://lnkd.in/dhFmvcSN Bjoern Bornkamp , Silvia Zaoli , Michela Azzarito , Ruvie Martin , Carsten Philipp Mueller , Conor Moloney , Giulia Capestro , David Ohlssen , Janice Branson
To view or add a comment, sign in
-
We are excited to announce the latest release of DataHowLab! DataHows AI-enabled bioprocess development and data analytics solution supports the efficient development of high-performing bioprocesses by maximising process learning and insight. The solutions features and applications are continuously being improved to support the development goals of our users within a curated user-friendly environment. We have introduced several powerful new features in our latest release, including: • Dynamic Experimental Design Optimization; • Improved Data Management for Projects; • DataHow SDK for direct API connection; • Calculated Variables in Dataset Editor & Visualization Boards; • Confidence Interval Control for Models. Would you like to learn more about each feature? Click the link to explore - https://lnkd.in/dm3ZazU6 or contact one of our experts. #data #dataanalytics #bioprocessing #artificialintelligence #machinelearning #digitalplatform #digitaltransformation #pharmaindustry #pharmaceuticalindustry #lifesciences #processdevelopment #processoptimization #datahowlab #release #releasenotes
Release Notes - DataHowLab v3.2
datahow.ch
To view or add a comment, sign in
-
FYI... ExpoSeq is a powerful pipeline for processing and analyzing FASTQ files from high-throughput sequencing of samples from display techniques such as phage display. The pipeline was mainly designed for analyzing phage display campaigns, but RNA-Seq data for analysing T-cell receptor data can also be given as input. It utilizes MiXCR to align and assemble the data which you can subsequently analyze in multiple plots. The pipeline offers a huge flexibility for various sequencing strategies and targets which are defined by the presets which are offered by MiXCR and are implemented in the pipeline. The pipeline focuses on analysing the quality of your sequencing and the sequence relations between your samples. Further, you can upload binding data to the pipeline to analyse these with various clustering techniques. https://lnkd.in/eNQPti-N
GitHub - DigBioLab/ExpoSeq: ExpoSeq is a pipeline to process and analyze in various visualizations ngs data from phage display campaigns
github.com
To view or add a comment, sign in
-
𝐎𝐛𝐬𝐞𝐫𝐯𝐚𝐭𝐢𝐨𝐧: 80% of Data Scientists do ad-hoc ML projects, while 20% develop and maintain libraries. In ad-hoc ML projects you have one business problem to solve, develop and deploy a model, and then move on to a different project. I've seen this happening not only in consultancy or customer teams but also in product teams. These teams tend to skip testing, code reviews, or good software practices because they focus on quick delivery. In contrast, the Data Science teams that develop and maintain libraries usually have a better understanding of software development. I'm curious to hear the observations from people in my audience.
To view or add a comment, sign in
-
Immediate joiner | Data Science | Python | SQL | Power BI | Analysis | Machine Learning | Statistics | NLP | Deep Learning | Django | Advance Excel | Post Graduate | Ex BharatTech, Upsolve.in
Hello Connection 👋 Step into the world of medical data exploration with our project, Medical Dataset Analysis💊: Python, SQL, and Insights" is an interesting field that I explore. With Python as my trusted tool and SQL as my analytical compass In my project, three important datasets are examined: hospitalization_details: Contains an abundance of data regarding patient admissions. medical_examinations: Offers details about the diagnosis and medical background of patients. names: Connects patients with the appropriate information in the other databases. It is divided into two main modules: 1. Data Cleaning and preprocessing : This is an important but thorough step. I make sure the data is accurate, consistent, and structured by thoroughly cleaning and preprocessing it. Imagine it like cleaning a diamond to show off its genuine radiance! 2. SQL-Powered Insights: Once the data shines, I unleash the power of SQL queries. With each query, I extract valuable insights, uncover hidden patterns, and answer important questions like: What are the average hospitalization charges? Which surgeries are most prevalent? How do charges vary across different cities? Is there a link between smoking and specific health issues? How does BMI impact healthcare outcomes? And many more. Through this project, I Gained hands-on experience with Python and SQL in a real-world healthcare context. Honed my data analysis skills, learning to clean, explore, and interpret medical data responsibly. Uncovered valuable insights that could potentially inform healthcare decisions and improve patient outcomes. This project, provided by the amazing HiCounselor platform, is just the beginning! I'm excited to continue exploring the power of data in healthcare and share my journey with you. Stay tuned for more updates and insights! Github link : https://lnkd.in/dic2grAd
To view or add a comment, sign in
-
Edit check programs Edit check programs are essential components of data validation in clinical research and healthcare settings. They help ensure the accuracy, consistency, and integrity of the data collected during clinical trials or healthcare studies. 1 Data Validation with PROCs: Utilize PROC MEANS, PROC UNIVARIATE, and PROC FREQ to validate data. PROC FREQ provides frequency counts of unique character values for variables like gender, diagnosis, adverse events, etc. PROC MEANS and PROC UNIVARIATE help identify outliers and summarize statistical measures such as mean, minimum, maximum, and percentiles. 2 Checking for Invalid Character Values: Employ PROC FREQ to identify and list invalid character values in the dataset. 3 Listing Invalid Data Values: Utilize PROC PRINT with WHERE statement to list observations with invalid data values, such as systolic and diastolic blood pressure falling outside specified ranges. 4 Identifying Outliers: Utilize PROC MEANS, PROC UNIVARIATE, and PROC TABULATE to identify outliers. PROC MEANS and PROC UNIVARIATE provide summary statistics and graphical displays like stem-and-leaf plots and box plots to visualize outliers. 5 Range Checking with PROC FORMAT: Utilize PROC FORMAT to define and apply formats for range checking, ensuring data falls within specified ranges. 6 Data Analysis with Data Step Operations: Employ DATA step operations like SET, MERGE, UPDATE, KEEP, and DROP for data analysis and manipulation. 7 Creating Datasets: Use PROC IMPORT and DATA step to create datasets from flat files, ensuring data integrity and consistency. 8 Extracting Data: Utilize LIBNAME statement to extract data from external sources into SAS datasets for analysis. 9 Statistical Analysis with SAS/STAT Procedures: Employ procedures like PROC ANOVA and PROC REG for statistical analysis, including analysis of variance and regression modeling. 10 Duplicate Data Handling: Use PROC SORT with NODUPKEY or NODUP to identify and remove duplicate observations based on specified variables. 11 Creating Analysis Datasets: Utilize PROC FORMAT, RENAME, and LENGTH statements to prepare and transform raw datasets into analysis-ready datasets, ensuring data quality and consistency. By implementing these techniques, SAS programmers can effectively validate, clean, and analyze data, ensuring its integrity and reliability for further analysis and decision-making in clinical research and healthcare settings. #sas #sasprogramming #cdisc #adam #sdtm #tlf #r #clinicaltrial #advancesas #clinicaltrialprogrammimg #biostatastic #datavalidation #clinicalresearch #healthcare
To view or add a comment, sign in
-
📊 Visualizing Sales Trends: A Dynamic Dashboard for Pharmaceutical Sales Analysis Excited to unveil a recent Python data analysis project focused on unraveling sales trends for one of the leading pharmaceutical companies! 📈 Through the power of Python and data visualization, I've created a dynamic dashboard that provides invaluable insights at various levels, from national down to territory specifics. Project Overview: I developed a comprehensive dashboard using Python's data analysis and visualization libraries. This dashboard offers a holistic view of sales performance, enabling stakeholders to delve deep into key performance indicators (KPIs) across different levels – national, regional, and territory. 🔍 Multi-Level Analysis: Dive into sales data with granularity, exploring trends and patterns at the national, regional, and territory levels. 📈 Performance Insights: Identify successes and challenges through insightful KPIs, facilitating informed decision-making and strategy formulation. By harnessing the power of data visualization, pharmaceutical companies can gain actionable insights into sales performance, identify growth opportunities, and optimize resource allocation. #Python #DataAnalysis #DataVisualization #salestrends #PharmaceuticalSales
To view or add a comment, sign in
-
Nievgen (training@nievgen.com) offers the CERTIFIED DATA SCIENCE ASSOCIATE training and certification course (Instructor-led, activity-based via ZOOM, PHP 5,000.00) this September 9, 16, 23, 30 | 6:30 PM - 10 PM. Please email us at training@nievgen.com for more information. SCHEDULE: September 9, 16, 23, 30 | 6:30 PM - 10 PM INSTRUCTOR: Engr. John C. Placente, MBA, MSc EE, CMBB Competencies of Learning: CDSA-01. Understand basic data science operations. CDSA-02. Understand data science applications. CDSA-03. Perform descriptive, predictive and prescriptive analysis. CDSA-04. Perform comparative analysis. CDSA-05. Use data science software in analysis. CDSA-06. Extract, store, transform and load data. CDSA-07. Perform data cleansing, merging and pivoting. CDSA-08. Use data science software for visualization. CDSA-09. Perform data clustering and anomaly detection. CDSA-10. Perform decision tree analysis. CDSA-11. Perform random forest modeling. CDSA-12. Perform Neural Network modeling. CDSA-13. Perform Deep Learning modeling. CDSA-14. Perform general linear modeling (GLM). CDSA-15. Perform polynomial regression. CDSA-16. Perform scoring, model testing and selection. CDSA-17. Perform basic optimization techniques. CDSA-18. Apply data science in case studies. #nievgen #datascience
To view or add a comment, sign in
2,636 followers