When it comes to cancer diagnosis, machines are just as good as human doctors, according to a new report from researchers at the National Institutes of Health.
The findings, published in the journal PLoS ONE, indicate that a growing number of new technologies are providing faster and more accurate diagnoses than ever before.
The study also points to a growing trend of AI-based cancer diagnosis.
“We see the emergence of an emerging technology to rapidly assess a person’s cancer status, in the same way that we would have an opinion about a person using the eye test,” said study lead author Dr. Daniela R. Roeser, a postdoctoral researcher at the Institute for Healthcare Engineering and Applied Science at Johns Hopkins.
“There are a number of technologies that we have been using to diagnose cancers in the past, but we’ve never been able to use them to identify the true cancer.
We can now use the technology to help us diagnose cancer.”
Roeser’s team used a robotic arm called a robot prostate cancer test (RPOCT) to help diagnose a patient with prostate cancer.
Using the RPOCT, a team of experts would collect information about the patient’s anatomy and their age and gender, and use the data to determine a person was at risk of developing prostate cancer or other cancers.
Roesers team found that using this approach to diagnose a person with prostate or lung cancer, it was able to identify approximately 1 in 7 cancer cells as tumors.
However, when it came to detecting lung cancer or cervical cancer, the RPoCT was only able to detect around 1 in 5 cancer cells.
The researchers noted that while the accuracy of the Rpoct is improved over the eyes test, the accuracy with which it can differentiate between human cancer cells and cancerous tissue remains low.
The researchers also pointed out that although this technology can help diagnose cancers, the benefits of the technology are limited in the long term, and not available to patients in all cases.
“There is still a lot we don’t know about the accuracy and reliability of this technology and it may not be feasible for patients in the future,” said Roesers senior author Drs.
Ananth B. Srivastava and Aniruddha P. Narasimhan.
“However, these improvements could be important in the short term, in order to identify more cases of cancer.”
In the study, the team used an RPOct in conjunction with a robotic prosthetic arm called the Proterus to help assess a patient’s prostate cancer risk.
They found that the Protersus was able, in conjunction to a second method, to identify around one in five prostate cancer cells in the prostate as potential tumors.
While the Protsus was capable of identifying a significant number of prostate cancer tumors, the overall accuracy of diagnosis remained low.
Despite the limitations of this approach, the researchers hope the findings will help spur the development of other similar approaches.
“I think the most interesting thing about this is that it’s actually being used to help people who are in the early stages of cancer and they’re also at high risk of the disease,” Roesering said.
“And this is something that is already happening.
We see a lot of progress with this technology, and it’s an important step in the right direction.”
While the Protesus can diagnose prostate cancer in a matter of minutes, it’s still only capable of performing a quick test.
The scientists said that in order for the RTOCT to be useful for detecting prostate cancer, further research is needed to determine how it could be applied to other types of cancers.
“The main question is whether this technology could be used to detect breast cancer, lung cancer and other cancers in more specific ways, such as diagnosing the cancer itself, and that will be very important,” said Srivas.
“We also need to look at how it might be used for other types.
We need to make sure we can actually use it to diagnose cancer, not just test for it.
We have some work to do.”
Source: PLoS ONE DOI: 10.1371/journal.pone.0028608