Data-Driven Site Selection: Leveraging Machine Learning

Selecting the right sites for clinical trials is a critical factor that can significantly influence the success of a study. Traditionally, site selection has been based on historical performance, expert opinion, and manual analysis of various factors. However, this approach can be subjective, time-consuming, and prone to errors. With the advent of machine learning (ML), data-driven site selection is becoming a reality, offering a more efficient, accurate, and scalable method to identify optimal sites for clinical trials.

1. Welcome Data Driven Site Selection : Leveraging Machine Learning 10/18/2022 www.clinosol.com | follow us on social media @clinosolresearch POTLAPALLY LAXMI M. Pharmacy (PhD)

2. Contents Study site & selection Introduction Overview of data driven paradigm in CT using ML Site selection Data Data driven site selection with ML Conclusion 10/18/2022 www.clinosol.com | follow us on social media @clinosolresearch 2

3. Study site & selection A clinical trial site is  a controlled environment where clinical research, development and testing can take place safely with appropriate oversight . Common site selection strategies  use past trial data to assess how well a site would perform in a prospective clinical trial  and different standardized and objective methods have been developed across industry and academia. Site Qualification Visit (SQV):  A meeting with a representative from a sponsor representative to ensure the institution is fully capable and equipped to run a specific clinical trial. 10/18/2022 www.clinosol.com | follow us on social media @clinosolresearch 3

4. Clinical trials are struggling, just look at the statistics. Despite all of the innovation and breakthroughs in drug discovery over the last several years, it still takes 10 to 15 years and about 1 billion to develop a new therapy. Globally, more than 80% of trials fail to meet enrollment deadlines, resulting in costly extensions, addition of new study sites, or outright study failure..  There are many reasons for this, of course. First, there is the enormity and complexity of the challenge of safely bringing a new drug to market. Recently , a new solution was designed to change that status by implementing a real-world data-driven approach to the clinical trial site selection process. 10/18/2022 www.clinosol.com | follow us on social media @clinosolresearch 4 Introduction

5. Site selection data 10/18/2022 www.clinosol.com | follow us on social media @clinosolresearch 5 The data types that need to be gathered: Demographic Data Economic Indicators Geographic and Environmental Factors Infrastructure and Accessibility Competitive Landscape

6. Overview of Data driven paradigm in CT using ML 10/18/2022 www.clinosol.com | follow us on social media @clinosolresearch 6

7. There is an increasing demand from sponsors for technology and data-driven solutions to improve site selection, study start-up and patient recruitment. With the explosion of data and improvements in analytical techniques, including emerging technologies that incorporate machine learning and artificial intelligence (AI), sponsors are also demanding greater transparency about how these processes are carried out. Improving the use of data analytics and technology would allow for an increased ability to re-use site information on multiple studies and reduce the time between site selection and initiation. In addition, it could support increased engagement and partnering with sites with known patient populations . 10/18/2022 www.clinosol.com | follow us on social media @clinosolresearch 7 Data driven site selection with ML

8. 10/18/2022 www.clinosol.com | follow us on social media @clinosolresearch 8 Overview of data processing in site selection using ML

9. Conclusion The ML technology have made significant progress in medicine and healthcare. Ongoing progresses in research from ML areas are creating the recent paradigm shift in medicine and healthcare. However, the implementation of these technologies in real-world medical fields is still in its infancy and huge concern. These technologies are in the early stages of development. All recent AI–based research technologies need to scale up for use. At the same time, all the technology needs proper validation and implementation. Therefore, the technologies must be scalable and applicable. Similarly, during technological implementation and use, there should be collaboration among computer scientists, physicians and data scientists, healthcare providers, and engineers, which is essential now. At the same time, new and diversified algorithms are needed with numerous medical applications and to improve the quality of diagnostics and therapeutics. 10/18/2022 www.clinosol.com | follow us on social media @clinosolresearch 9

10. Thank You! www.clinosol.com (India | Canada) 9121151622/623/624 [email protected] 10/18/2022 www.clinosol.com | follow us on social media @clinosolresearch 10