Specifically, Readout plans to use the cash to expand its team, accelerate its product development and “deepen its AI sophistication” according to a public release.
Readout’s AI tool is designed to automate data analysis and describe the results in a range of formats including table captions and abstracts. This is typically done by biostatisticians and medical writers manually, and can take weeks in the case of complex clinical trials.
Users can upload preclinical and clinical trial data on Readout’s website and use its AI-guided algorithms to automatically detect what types of data it is and what statistics are needed. The tool then describes the findings and builds data tables based on the input data. The users can pay to download the results, and the data is soon deleted by Readout’s system to maintain data privacy.
One aim of this tool is to let researchers crunch the numbers from clinical trials more quickly to make development decisions. It can also assist in preparing clinical study reports (CSRs), comprehensive reports about the trial and results for regulatory authorities, sponsors and investigators.
"One application that's often overlooked is technology to improve trial progression and performance – most people just focus on the start, patient recruiting, and at the end, final results. But Readout's AI can also help analyze and report on progression, which is a novel improvement," said Michael Shleifer, co-founder of Readout AI, in a public statement.
The market for AI tools for clinical trials is predicted to grow from $1.5 billion in 2022 to $4.8 billion in 2027, spurred by the push to cut costs of clinical development in the pharmaceutical industry. Some of the main players in the space include Insilico Medicine, Unlearn.AI, IBM and more.
The launch of the AI-driven chatbot ChatGPT in late 2022 led to speculations that AI models like ChatGPT could be used to speed up the design and organization of clinical trials. Experts note that these tools would likely support and not entirely replace healthcare professionals in these complex studies.
Even with the availability of AI-based tools, there are still challenges for companies aiming to adopt them in clinical trials. For example, clinical trial data can come from a variety of systems and formats and require a lot of manual effort to standardise. Additionally, databases often need to be built and rebuilt in between trials, which can create a challenge for streamlining the process.