Artificial intelligence and machine learning in clinical development: a translational perspective

Abstract

Future of clinical development is on the verge of a major transformation due to convergence of large new digital data sources, computing power to identify clinically meaningful patterns in the data using efficient artificial intelligence and machine-learning algorithms, and regulators embracing this change through new collaborations. This perspective summarizes insights, recent developments, and recommendations for infusing actionable computational evidence into clinical development and health care from academy, biotechnology industry, nonprofit foundations, regulators, and technology corporations. Analysis and learning from publically available biomedical and clinical trial data sets, real-world evidence from sensors, and health records by machine-learning architectures are discussed. Strategies for modernizing the clinical development process by integration of AI- and ML-based digital methods and secure computing technologies through recently announced regulatory pathways at the United States Food and Drug Administration are outlined. We conclude by discussing applications and impact of digital algorithmic evidence to improve medical care for patients.

The need for a system view to regulate artificial intelligence/machine learning-based software as medical device

“…Regulators may also simply require clinical trials of the AI/ML-based SaMD as used in actual planned clinical settings. Moreover, regulators could also request data collected outside traditional clinical trials such as from Fitbits and other wearables capturing users' behavioral changes over time as well as electronic health records capturing all decisions that may be related to the use of an AI/ML-based SaMD 20 .…”

Artificial intelligence in glioma imaging: challenges and advances

“…Although machine and deep learning techniques have exhibited great potential in analyzing glioma images, their implementation in clinical care remains an elusive goal. Several recent reviews have broadly summarized the challenges in applying AI to clinical medicine (some of which also apply to neuro-oncology) [26,27,28]. These challenges involve the full life-cycle of developing an AI model, from 1) obtaining the training data, to 2) training the AI models, to 3) evaluating and deploying the AI model to clinical settings.…”

Translational Informatics Connects Real‐World Information to Knowledge in an Increasingly Data‐Driven World

“…As an evolving community, we also need to explore the potential for informatics to overcome challenges in storing, linking, and leveraging the rapidly growing resources of molecular and clinical data while retaining patient privacy. 4…”