"Sakura-san, please set up the protocol, database design, and electronic medical record link for Company A's new clinical trial."
"Yes, yes, I'm back."
The bio venture where I work has a real-world database built by combining our own AI algorithms with clinical trial data from various pharmaceutical companies, treatment data collected from the National Hospital Organization and medical groups across Japan, receipt data, and prescription data. We are also developing a simulator of biological reactions called virtual twins. Furthermore, we are contracted to develop new generation drugs and conduct post-marketing surveillance using these.
Since most of the work required for drug development is carried out on quantum computers, if there is an engineer who has mastered how to use the system, almost all of the development process can be carried out by a few people, and the collection of subject data can be kept to a minimum, making it possible to efficiently carry out clinical development with minimal burden on patients, and the system we developed is very popular with pharmaceutical companies.
In addition, with regard to the collection of subject data, the concept of monitoring has changed dramatically as all electronic medical records in the world can be easily connected to the clinical data repository (database) thanks to an AI-equipped electronic medical record data automatic mapping system developed about two years ago.
All data entered into the electronic medical record at the facility is automatically transferred to the clinical data repository, and tasks that were previously performed by clinical development monitors visiting the facility, such as risk assessment of safety information, signal detection, abnormal value determination, and query issuance, are now automatically performed by AI. Since data entry at the facility is no longer required, facility visits and SDV are no longer necessary, and all monitoring is now performed centrally. As a result, in order to monitor clinical data, the skill of reviewing data across the entire trial is now required more than ever before.
In particular, the skill of clinical development monitors and data managers is shown by how quickly they can perform tasks that AI cannot cover, such as comparing the settings of analysis conditions and the output results to set more appropriate analysis parameters, or correcting data bias based on the judgment of the analysis results.The ability to shorten the time it takes to reach results is an indicator of whether the person in charge is top-notch or not.
It's been about 20 years since EDC dramatically improved the efficiency of data collection in clinical development and replaced paper CRFs, but the electronic medical record data automatic mapping system replaced that EDC and even changed the concept of clinical monitoring.
"Well, now that the protocol design is done, I'll set up the AI to map it with the medical records and then go home."
Rock in the photo frame on the desk looked like it was smiling again today.
========================================================================
"Huh?"
I must have had a long dream, because when I woke up, it was almost 7pm. I had fallen asleep with my face down on the desk.
"Sakura, it's time for medicine."
Rock, who was moving around busily at my feet, said this.
"What are you busy doing?"
"I'm dancing Monster gymnastics number two...."