On 8 October, Thomas Wood of Fast Data Science presented the Clinical Trial Risk Tool, along with the Harmony project, at the AI and Deep Learning for Enterprise (AI|DL) meetup sponsored by Daemon. You can now watch the recording of the live stream on AI|DL’s YouTube channel below:
The Clinical Trial Risk Tool leverages natural language processing to identify risk factors in clinical trial protocols. The Clinical Trial Risk Tool is online at https://clinicaltrialrisk.org/tool.
Artificial Intelligence and Deep Learning for Enterprise is a meetup group in London dedicated to talks from people in the industry using developments in AI for exciting real world applications.
We initially developed the Clinical Trial Risk Tool to identify risk factors in HIV and TB protocols. Version 2 is coming soon, which will also make cost predictions (i.e. predict the cost of running a trial in dollars), and which will also cover further disease areas, such as Enteric and diarrheal diseases, Influenza, Motor neurone disease, Multiple sclerosis, Neglected tropical diseases, Oncology, COVID, Cystic fibrosis, Malaria, and Polio.
The project has been funded by the Bill and Melinda Gates Foundation and we have published a technical paper in the journal Gates Open Research:
The software is under MIT License, meaning that it is open source, and can be freely used for other purposes, both commercial and non-commercial, with no restrictions attached. The source code is on Github at https://github.com/fastdatascience/clinical_trial_risk.
[Fast Data Science]](https://fastdatascience.com/) is a leading data science consultancy firm providing bespoke machine learning solutions for businesses of all sizes across the globe, with a concentration on the pharmaceutical and healthcare industries.
Guest post by Youssef Soliman, medical student at Assiut University and biostatistician Designing a high-quality clinical trial protocol is critical for the success of any study. A protocol is the blueprint that outlines every aspect of a trial. In an ideal world, a flawless protocol would require no revisions and include only essential elements. In reality, however, the average protocol undergoes 2–3 amendments and often contains excessive data collection and overly complex entry criteria.
Clinical trials have long been the foundation of medical breakthroughs, but traditional methods often stumble over slow timelines, high costs, and difficulties in finding the right participants. Artificial intelligence (AI) — a technology ready to transform this landscape by making trials faster, more affordable, and smarter. The accelerating adoption of AI in clinical trials signals a major shift in healthcare research. It is already making significant strides in transforming clinical trials.
Guest post by Safeer Khan, Lecturer at Department of Pharmaceutical Sciences, Government College University, Lahore, Pakistan In clinical trials, a staggering 80% encounter delays during the startup phase and 37% struggle to meet enrollment targets. Read more Key clinical trial statistics. These figures highlight a critical, yet often underemphasized, aspect of clinical trials—the feasibility process. The feasibility process is essential for assessing the practicality of a clinical trial’s design, ensuring the study is prepared to tackle the challenges that may arise.
