Artificial Intelligence and Machine Learning Archives

The Year Ahead for FSP: Open Source, AI, Global Reach, and Cost Efficiency

The Biometrics FSP outsourcing market is evolving faster than ever. Looking back, 2024 was a year of transition for our industry as we put the COVID bubble in our rearview mirror and focused on efficient delivery of our portfolios. Looking ahead now to 2025, Biometrics FSP is on track for continued growth, with a strong emphasis on open-source technologies, global reach, artificial intelligence, and cost efficiency.

Here, I touch a bit on these four areas and share our thoughts on the impact they will have in 2025 and beyond.

Embracing open-source technologies

While the adoption of open-source programming has been slower in the clinical research space, tools like R and Shiny are quickly gaining traction as cost-effective and reliable solutions for data analysis and submissions.

Cytel has been leveraging open-source software for big data aggregation, application development, and validation. We will continue to be a key contributor to the open-source ecosystem and help organizations solve key design and analysis challenges, while offering access to the industry’s top R/Shiny talent pool.

Offshoring hub strategy and cost-effective solutions

The demand for more cost-effective solutions continues to drive the use of offshore resources across the industry. Cytel’s expansion into Eastern Europe and continued growth of our South Africa– and India-based teams positions us well to support our sponsors in reducing costs while maintaining the high levels of quality they have come to expect of Cytel.

Artificial intelligence

AI is revolutionizing the biometrics space by enabling real-time data monitoring, automated code generation, and improving statistical accuracy in clinical trials. By flagging anomalies and potential errors, AI reduces the risk of data discrepancies and enhances overall data quality. AI-powered tools are streamlining biometric services, automating routine tasks and allowing researchers to focus on high-value activities.

Secure data collection and real-time monitoring

Innovations in data collection and real-time monitoring are improving privacy, security, and data integrity. Advanced authentication methods and AI integration are helping ensure the accuracy and confidentiality of data.

Automation is also playing a key role by extracting data from unstructured sources, such as medical records, and reducing human error during data transcription. This further enhances the efficiency of electronic data capture (EDC) systems and boosts the overall reliability of clinical data.

Final takeaways

I’m more excited about 2025 than any year in my career. In an industry that has been criticized for moving too slowly and cautiously, we sit an inflection point for rapid evolution of decades-old models. Change can be exciting and scary all the same. Reach out — myself and our FSP experts are eager to present content and engage with you throughout the year at events such as PHUSE, JSM, SCDM, and more.

AI’s Influence on SAS Programming

The advent of Artificial Intelligence (AI) has transformed numerous fields, and the domain of SAS (Statistical Analysis System) programming is no exception. From automating tedious tasks to enhancing decision-making processes, AI has made significant inroads into how SAS programmers work.

However, AI is not a substitute but a companion to programmers. While AI can help us focus our critical thinking, creativity, and problem-solving skills, AI needs our expertise. Domain expertise is still essential.

To understand this transformation better, here we explore key ways AI has impacted SAS programming, particularly by comparing skills of traditional to AI-assisted programming, examining the days before and after AI, and discussing the new responsibilities and skills required in the modern programming landscape.

Traditional SAS programming vs. AI-assisted SAS programming

Traditional SAS programming has long been a manual, code-intensive practice requiring a high level of expertise in statistical analysis and programming. In the earlier days, SAS programmers worked with well-defined, often repetitive tasks. The process of developing code required a deep understanding of the data and statistical methodologies, all while meticulously debugging and quality-checking code.

AI-assisted SAS programming introduces a new level of efficiency, allowing programmers to focus more on value-added tasks rather than repetitive work. Traditional SAS programming workflows are now supported by AI-driven automation tools that can generate code, optimize algorithms, and even offer suggestions for complex statistical analyses. For example, where traditional methods would require a programmer to sift through data to find patterns, AI can now analyze large datasets in seconds and offer insights that help in decision-making. This allows the SAS programmers to focus on more strategic and high-level interpretations.

In essence, the role of the SAS programmer is evolving from being a “code generator” to a “code curator” and they maintain control over every step, providing deep customization and understanding of the entire process.

AI as a companion, not a substitute

The fear of AI replacing jobs has become a common narrative, but in the case of SAS programming, AI should be viewed as a companion rather than a replacement. While AI can optimize code, automate reporting, or even suggest corrections, it is still far from replacing the creative and analytical skills of programmers. AI systems can generate insights based on patterns within datasets, but understanding the nuances of those patterns and making informed decisions based on them remains a unique programmer’s skill.

SAS programmers have a deep understanding of the data they work with, including the context, limitations, and real-world implications of their findings. While AI can handle the heavy lifting in terms of data processing and analytics, the role of the programmer is to interpret these findings, cross-check their accuracy, and ensure the outputs are aligned with business goals or research questions.

Additionally, AI’s suggestions aren’t always perfect, especially when dealing with edge cases or complex datasets with nuanced relationships. In such scenarios, a programmer’s oversight is crucial to prevent AI-driven errors from propagating throughout the analysis.

Before and after AI

The landscape of SAS programming before the integration of AI was characterized by manual coding, exhaustive debugging processes, and labor-intensive quality control procedures. Let’s break down the key changes AI has brought to these areas:

Code development

Before AI, coding was manual and depended heavily on a programmer’s syntax knowledge and experience to ensure that the code adhered to best practices for efficiency and performance. This could be a time-consuming process, especially when dealing with large, complex datasets.

In the post-AI era, code development is becoming more efficient through AI-assisted coding tools. These tools can automatically suggest code snippets based on previous coding patterns or even generate entire blocks of code tailored to the dataset. AI-driven auto-complete features and advanced libraries that recommend the best statistical models or data manipulation techniques have significantly sped up the development process.

Debugging

Debugging used to be a meticulous and painstaking part of the SAS programmer’s job. Identifying errors in code or incorrect outputs is often required by going through large blocks of code line by line, manually reviewing logic and syntax.

AI has revolutionized debugging by identifying errors in real time, suggesting fixes, and even automatically correcting syntax errors. AI tools can also track changes in code and predict where potential issues might arise based on past errors, significantly reducing debugging time and enhancing code accuracy.

Quality control (QC)

Before AI, the QC process was often manual or semi-automated, prone to missed errors, and involved peer reviews, statistical validations, and rigorous testing to ensure that the code met the necessary standards. This was particularly important in industries such as healthcare or finance, where data accuracy is critical.

Today, AI-driven QC tools can automatically verify the integrity of datasets, flag inconsistencies, and ensure that statistical models meet predefined accuracy thresholds. These tools can run tests much faster than human reviewers, allowing for quicker validation cycles and better compliance with industry standards.

AI doubles productivity, without replacing the need for programmer’s intuition and expertise, so we can opt for other developmental activities like enhancing the client outcomes, learning new skills, and mentoring to strengthen the overall team.

New responsibilities and skills for SAS programmers in the AI age

New responsibilities and skills required for AI platforms

To understand how to work along with AI tools
To adopt AI-driven workflows for faster development cycles
To learn to guide and review AI-generated code
Additional skills like data literacy, critical thinking, and ethical AI considerations are also required

Industry AI tools

Tabnine: AI-powered code predictions
Snyk: AI-driven security checks
DeepCode: Real-time AI code review
SAS Viya: Integrate existing code with AI tools

Final takeaways

AI tools are transforming the role of SAS programmers, making them faster and more effective, but human expertise remains crucial in directing AI and ensuring high-quality outcomes. The future of programming likely lies in a hybrid approach that leverages both human expertise and AI-driven efficiencies.

Interested in learning more about AI in clinical development? Watch our recent webinar:

Driving Innovation in Clinical Trial Design: Open Source, Commercial Software, and AI in 2025

As we usher in a new year, we reflect on 2024’s prominent trends in simulation software for clinical trial design that will continue to drive innovation in the coming year. The two main areas of growth and innovation we see taking the lead in 2025 are:

The combination of open source with commercial software solutions
The increasing use of AI to generate open-source code and augment clinical trial design

Commercial software: Confident and quick design capabilities

Commercial software remains a common and popular choice for clinical trial design, with many sponsors opting for these tools. This choice allows for confident and quick design through validated workflows and pre-coded and verified design types. As an accepted choice with a wealth of trial design options, biostatisticians can easily and quickly design and compare a variety of trials. Furthermore, users enjoy access to expert professional support in addition to frequent software releases that ensure updates to methodologies and design types.

Open-source code offers a high degree of flexibility

Although commercial software provides numerous benefits to biostatisticians, there are also drawbacks to this choice in isolation. In a complex scientific field, biostatisticians often encounter idiosyncratic problems that require unique and custom solutions. In these cases, validated commercial software may prove insufficient, and custom code must be developed to address the problem at hand. In fact, this need for flexibility is at the root of the rise of open-source software for custom coding using industry-accepted languages like R, Python, or Julia. These languages afford biostatisticians a degree of creativity in their work and go hand-in-hand with the collaborative nature of this highly academic field. Over the years, many code packages have been developed and shared as solutions to unique design aspects, helping to drive and shape industry trends.

However, with this near-limitless flexibility come several drawbacks. Vetting or developing a bespoke solution can be complicated and resource intensive. Time is required for collecting requirements, writing code, testing, and validating a custom open-source design option. This approach relies on a set of expertise in both software development and statistical methodology. While biostatisticians have deep knowledge and experience in statistics and clinical trial design, they are not typically trained in best practices for software development and programming. These best practices are crucial in developing reliable, robust solutions that can easily be shared with others and that apply to a wide array of trials. Finally, the results derived from open-source code require additional resources for both design selection and communication of results, in the context of a multidisciplinary team. The biostatistician’s attention is thus diverted from providing valuable strategic input to the clinical development team towards software development and implementation tasks.

Combining open-source code with commercial software

Acknowledging these challenges, the industry is quickly adopting a combined-capabilities approach that incorporates the established, validated backbone of commercial software with the added creativity afforded by open-source code. This approach allows biostatisticians to augment elements of the design such as the choice of analysis type, statistical test, or the distributions used to generate various design inputs, without the need to code an entire design. In addition, clinical trial design professionals benefit from the cloud computing power embedded in some commercial software solutions, eliminating the need for maintaining an expensive internal computational grid. We believe that this integrated future of study design harnesses the benefits of both commercial software and open-source solutions while limiting the drawbacks experienced with each approach individually.

The use of artificial intelligence in generating code for clinical trial design

Along with the intensive use of R and other coding languages, we believe that we will see increased interest in using AI solutions for a variety of clinical trial design and execution activities. These applications of AI may include data transformation and cleaning; statistical analysis; protocol writing; clinical data reporting; trial management practice; and efficient code generation and validation for clinical trial design. For the latter, AI solutions powered by Large Language Models (LLMs) can be harnessed to produce analysis-ready custom code based on project specifications. Indeed, over the past few months, Cytel has introduced an AI-driven coding assistant in its newest clinical trial design software to augment study designs with novel approaches via custom code. This approach holds several advantages, among them: the ability to generate code faster; the potential for efficient code validation and editing; and the ability to generate code using natural language prompts.

With the great promise that such tools hold, there are also potential drawbacks and concerns expressed by biostatisticians working in the field. AI-supported code generation requires close review by trained coders to ensure the code created using these tools is sound and applicable to the purpose for which it was created. While code generated by AI can save considerable resources, it requires close supervision and review for validation and application in practice. Over-reliance on code-generation tools may, over time, change the way in which statisticians think through complex coding problems, and limit creativity in this field.

Final takeaways

The landscape of clinical trial design is poised for significant advancements in 2025, driven by the integration of commercial software and open-source solutions, as well as the innovative application of AI for code generation. By leveraging the strengths of commercial software — validated workflows, expert support, and computational power — and combining them with the flexibility and creativity of open-source coding, biostatisticians can overcome traditional challenges and design trials more efficiently. Furthermore, AI-powered tools promise to streamline the generation, validation, and customization of code, empowering teams to focus on strategic decision-making and innovation. These trends signal a promising era of collaboration, efficiency, and enhanced capabilities in clinical trial design.

Cytel’s East Horizon Platform now includes open-source integration points, allowing users to inject custom analysis types, statistical tests, and patient outcome generation into existing software workflows. In addition, the software includes an advanced AI-driven coding assistant that can generate compatible custom R code using plain language queries for integration in study designs. These new features, in combination with Cytel’s advanced trial simulation tools and cloud computing capabilities offer a potent, comprehensive solution for clinical trial design and optimization.

The Future of Drug Development: Data Science, AI, and the Evolution of the Clinical Trial

The year 2025 is poised to be a turning point in clinical development, driven by a convergence of trends that are reshaping the way we generate evidence and bring new treatments to patients.

A key theme emerging from industry discussions is the need to modernize the clinical trial process by embracing the power of data science, advanced analytics, and emerging technologies. While the traditional RCT remains a cornerstone of drug development, there’s growing recognition that it can be enhanced and augmented to meet the demands of the 21st century.

Several factors are driving this shift. First, the sheer volume and variety of data available to researchers is exploding. We are awash in data from electronic health records, genomic databases, wearable sensors, and even social media platforms. This data deluge presents both opportunities and challenges, requiring new tools and techniques to extract meaningful insights.

Second, the regulatory and reimbursement landscape is evolving rapidly. Payers are increasingly demanding evidence of value in parallel with regulators, with both encouraging the use of real-world data (RWD) to support submissions. In Europe, for example, the EU Joint Clinical Assessment (JCA) is creating common standards to expedite the HTA process and requiring sponsors to consider payer perspectives much earlier in the development process.

Third, the rise of precision medicine and targeted therapies requires more sophisticated trial designs to identify patient subpopulations most likely to benefit from treatment. Adaptive designs, master protocols, and the use of biomarkers as surrogate endpoints are all gaining traction.

Here’s a glimpse of what we might see in 2025

AI and machine learning will play an even more prominent role across the entire clinical development lifecycle.
- Algorithms will be used to identify promising drug targets, screen compounds, and optimize lead candidates.
- AI-powered tools will automate routine tasks in data management, statistical programming, and medical writing, freeing up researchers to focus on higher-value activities.
- AI will also power advanced analytics, enabling the development of predictive models that can forecast trial outcomes, personalize treatment decisions, and identify safety signals.

Simulation-guided design will become the norm. Sophisticated platforms will enable sponsors to evaluate a wider range of trial design options in silico, optimize resource allocation, and improve hypothesis generation.

The lines between different data sources will continue to blur. RWD will be routinely integrated with clinical trial data, requiring new statistical approaches like causal inference and quantitative bias analysis to address issues of bias and confounding.

Digital endpoints and biomarkers will move to the forefront. Wearable sensors, imaging technologies, and “omics” data will provide richer and more patient-centric insights into disease progression and treatment response.

One of the most exciting areas of innovation is the emergence of agentic AI and the use of digital twins and synthetic data.

These technologies have the potential to revolutionize clinical trials by:

Automating key trial processes: Agentic AI systems, trained on vast amounts of data, could manage tasks such as patient recruitment, data collection, and safety monitoring, potentially reducing costs and accelerating timelines.

Creating virtual patient populations: Digital twins, virtual representations of real patients built using diverse data sources, could be used to simulate the effects of different treatments, optimize trial designs, and even identify new drug targets.

Enhancing control arms: Synthetic data, generated by algorithms trained on real patient data, could be used to create external control arms, reducing the need to recruit control patients and potentially making trials more efficient and ethical.

The convergence of these trends will require a collaboration of clinical data scientists

— ones who not only master statistical techniques but are also fluent in data science, machine learning, epidemiology, and domain-specific knowledge of drug development. These individuals will be key to unlocking the full potential of the data revolution, translating complex insights into actionable strategies, and guiding the industry toward a future of more efficient, patient-centric, and data-driven clinical trials.

However, as we embrace these powerful new technologies, we must also be mindful of the ethical implications. Ensuring algorithmic accountability, transparency, and fairness will be paramount. The role of statisticians and data scientists will be crucial in guiding the responsible use of AI and ensuring that it benefits patients and society as a whole.

The year 2025 promises to be a pivotal year in the evolution of clinical development. By embracing innovation and collaboration, we can harness the power of data to accelerate the development of new treatments and improve the lives of patients worldwide.

Opportunities and Pitfalls of Using AI/ML in Clinical Development

The future of clinical development is set for significant change, driven by the integration of new digital data sources, advanced computing power for detecting meaningful patterns using artificial intelligence (AI) and machine learning (ML), and increasing regulatory support through new collaborations.[1] The United States Food and Drug Administration (FDA) has noted a significant increase in the use of AI/ML in drug development, with over 100 submissions in 2021 alone.[2] These submissions cover the entire drug development process, from drug discovery and clinical research to post market safety and advanced pharmaceutical manufacturing. Here, however, we concern ourselves with how AI and ML approaches can impact the way clinical trials are designed, conducted, and analyzed, offering potential enhancements in efficiency, reduced costs, and improved patient outcomes.

In this article, I explore the opportunities and challenges of AI/ML in clinical development, particularly in the context of Cytel’s operations.

Artificial Intelligence Applications in HEOR

Written by Reza Jafar, Omar Irfan, and Maria Rizzo

Recent advancements in machine learning (ML) and artificial intelligence (AI) can offer tremendous potential benefits to health economics and outcomes research (HEOR), such as in cohort selection, feature selection, predictive analytics, causal inference, and economic evaluation.[1] The use of ML and AI has been previously explored in systematic literature reviews (SLRs), real-world evidence (RWE), economic modeling, and medical writing.[2-4]

In this article, we assess the evolving landscape of evidence and developments attributed to AI in HEOR, reflecting on recent insights and developments presented at the 2024 US conference for The Professional Society for Health Economics and Outcomes Research (ISPOR) in Atlanta. Read more »

Navigating the Clinical Development Landscape: Insights for Success in 2024

After explosive and frenetic activity in the clinical trial industry during the COVID era, the past two years have seen challenging market dynamics and a drop-off in activity. Every one of us working in clinical development has felt this slowdown, but as we begin 2024, there is reason for optimism. The future looks promising. Here are some things to consider as you go forward.
Read more »

Living SLRs and the Rise of Digital Technology

Written by Marie Diamond and Maria Rizzo

Systematic literature reviews (SLR) are essential to informing healthcare decision-making and are pivotal for reimbursement submissions to health technology assessment (HTA) bodies. One limitation of SLRs, however, is that they quickly become outdated due to the continuous publication of new literature. The living systematic review (LSR) model can overcome this challenge by incorporating relevant new evidence as it is published. Read more »

Looking to the Future — Improving Diagnosis and Prognosis of Eye Conditions with Artificial Intelligence

Written by Alind Gupta, Cytel; Haridarshan Patel, Horizon Therapeutics; and Jason Simeone, Cytel

Ophthalmology is well-suited to using artificial intelligence (AI) methods because clinical decisions often rely on complex data-rich information from medical images of the eye and patient health status. AI has revolutionized image data analysis over the last decade and is promising to improve healthcare delivery and clinical decision-making while reducing healthcare costs. In fact, some AI-based diagnostic platforms for early detection of diabetic retinopathy have received FDA clearance and have been introduced in resource-constrained healthcare settings worldwide to screen for early signs of disease and to accelerate patient access to correct therapies. Read more »

Embracing AI and ML in Medical Devices: FDA’s Total Product Lifecycle-Based Regulatory Framework

Written by Fei Tang, RWE Senior Research Consultant, and Paul Arora, Assistant Professor (Status), Dalla Lana School of Public Health, University of Toronto

The Food and Drug Administration (FDA) has long been committed to innovative approaches to the regulation of medical device software and other digital health technologies. One such innovation is the use of artificial intelligence (AI) and machine learning (ML) in software, which has the potential to learn from real-world use and experience and improve its performance. The FDA’s vision is that AI/ML-based Software as a Medical Device (SaMD) will deliver safe and effective software functionality that improves the quality of care that patients receive. Read more »

Newer Posts

Discovery

Phase I-III Clinical Trials

Commercialization

Real-World Evidence Solutions

Clinical Trial Design

Trial Delivery

Advanced Analytics

Specialty Areas

Strategic Consulting

Beyond Functional Service Provider

Project-Based Analytical Solutions

Trial Design Software

Trial Implementation and Decision Support Software

LiveSLR® Software for Systematic Literature Reviews

Our Solutions

Quick Links

Quick Links

Quick Links

Quick Links

Quick Links

Quick Links

Quick Links

Quick Links

Quick Links

Quick Links

Quick Links

Quick Links

Quick Links

Quick Links

About Us

Quick Links

Insights

Quick Links

Careers

Quick Links

Embracing open-source technologies

Offshoring hub strategy and cost-effective solutions

Artificial intelligence

Secure data collection and real-time monitoring

Final takeaways

Traditional SAS programming vs. AI-assisted SAS programming

AI as a companion, not a substitute

Before and after AI

Code development

Debugging

Quality control (QC)

New responsibilities and skills for SAS programmers in the AI age

New responsibilities and skills required for AI platforms

Industry AI tools

Final takeaways

Interested in learning more about AI in clinical development? Watch our recent webinar:

Commercial software: Confident and quick design capabilities

Open-source code offers a high degree of flexibility

Combining open-source code with commercial software

The use of artificial intelligence in generating code for clinical trial design

Final takeaways

Here’s a glimpse of what we might see in 2025

One of the most exciting areas of innovation is the emergence of agentic AI and the use of digital twins and synthetic data.

The convergence of these trends will require a collaboration of clinical data scientists