Clinical trials are good, fast, inexpensive and with great knowledge
“One of the biggest challenges in my career is enrolling patients in education,” says Nicholas Borys, chief medical officer at Lawrenceville, NJ, a Celsion-based technology company that manufactures next-generation chemotherapy for cervical and cervical cancer and other types. brain tumors. Borys estimates that less than 10% of cancer patients register for medical examinations. “If we were to get 20% or 30%, we would probably have a number of cancers by now.”
Clinical trials test drugs, equipment, and new methods to determine if they are safe and effective before they are approved for use. But the process from making the study to being accredited is long, twisted, and expensive. Nowadays, researchers are using artificial intelligence and data analytics to speed up the process, reduce costs, and provide faster medical care for those who need it. And they are using what has been used but what is growing rapidly: more of the patients from previous trials
Making external controls
Clinical trials usually involve at least two groups, or “arms”: a test or trial that receives the investigated treatment, with the control hand not receiving it. The control hand will not receive any treatment, facilities or services currently available, depending on the type of medication being prescribed and what is being compared to the research program. It is easy to see the problem of co-workers studying cancer and other serious illnesses: patients with a serious illness need help now. While he may be in danger of helping out at the new hospital, “the last thing he wants is a change from the authorities,” Borys says. Including these doubts is the need to consult patients with rare diseases – for example, a type of breast cancer that is characterized by genetic predisposition – and the full enrollment period can last months, or years. Nine out of 10 clinical trials worldwide — not just cancer but all sorts of cultures — will not find enough people in their time. Some tests fail together due to lack of sufficient participants.
What if the researchers did not have to look for a control group and were able to provide test support to anyone who agreed to participate in the study? Celsion is exploring a similar approach in New York, headquartered in Medidata, which provides more than half of all clinical trials and surveillance programs worldwide, by deploying major medical and medical companies. Founded by the French software company Dassault Systèmes in 2019, Medidata has written a “great” tool: more from more than 23,000 tests and about 7 million patients who have returned in about 10 years.
The idea is to use data from patients in previous trials to create “external correction instruments.” These groups also work similarly to traditional weapons, but can be used in areas where the control group is difficult to find: due to a rare disease, for example, or a disease such as cancer, which is at high risk. It can also be used to test the “one-handed” test, which makes the control group impossible: for example, to measure using a fixed instrument or surgery. Perhaps their most valuable use is to do some initial testing, to see if the drug should be followed until it is fully tested.
Medidata uses artificial intelligence to create its own database and locate patients who were in control in previous tests of a different type of treatment to develop external external devices. “We can carefully select past patients and compare the current trial arm with the test history,” says Arnaub Chatterjee, senior vice president of marketing, Acorn AI at Medidata. (Acorn AI is a Medidata component of distribution and analytics.) These trials with patients correspond to the objectives of the study — referred to at the end, such as the reduction in mortality or the length of time patients remain cancer-free — and in other design settings, such as the type of data collected at the beginning of the study and along the route.
Using the external monitoring hand, “We do our best to emulate well-conducted experiments,” said Ruthie Davi, vice president of data science, Acorn AI at Medidata. The first step is to look at resuscitation for people who may need to have an arm amputated using the following diagnostic methods: for example, the type of cancer, the severity of the disease and its condition, and whether it is the first time the patient is being treated. It is the same method used to select patients for clinical trials — except those listed in the initial clinical trials, not the current ones, used to determine eligibility, Davi says. “We are finding patients who would be judged if they were present today.”
Download the full report.
This was created by Insights, the hand of material contained in the MIT Technology Review. It was not written by the authors of the MIT Technology Review.