UPMC Develops AI That Can Identify Prostate Cancer With Near 100% Accuracy - NewsRadio 1020 KDKA

PITTSBURGH (Newsradio 1020 KDKA) In a study published today in the Lancet Digital Health, UPMC and the University of Pittsburgh School of Medicine has demonstrated the highest accuracy in recognizing and characterizing prostate cancer by using artificial intelligence (AI).

 “Humans are good at recognizing anomalies, but they have their own biases or past experience,” senior author Rajiv Dhir, M.D., M.B.A., chief pathologist and vice chair of pathology at UPMC Shadyside and professor of biomedical informatics at Pitt said.

Dhir said that using AI adds an element of standardizing care.

In training the AI to recognize prostate cancer, Dhir and his colleagues provided more than a million images of stained tissue slides from patient biopsies.

Each image was labeled by an expert to teach the AI how to identify healthy or abnormal tissue.

The algorithm was tested on 1,600 different slides taken from 100 consecutive patients seen at UPMC suspected prostate cancer.

The AI demonstrated 98% sensitivity and 97% specificity at detecting prostate cancer, much higher than previous tests were showing.

This algorithm is the first to go beyond cancer detection; The algorithem is able to detect important clinical features such as tumor grading, sizing, and invasion of surrounding nerves.

Dhir did notify that the AI is not particularly superior to humans, in some cases a pathologist could see enough evidence of malignancy from other samples from a patient.

The AI is good failsafe for less experienced pathologists.

“Algorithms like this are especially useful in lesions that are atypical,” Dhir said. “A nonspecialized person may not be able to make the correct assessment. That’s a major advantage of this kind of system.”

While these results look promising, Dhir stated that new algorithms still have to be trained to detect other forms of cancer, but he is optimistic that the AI can adapt and learn.

Additional authors on the study include Liron Pantanowitz, M.B.B.Ch., of the University of Michigan; Gabriela Quiroga-Garza, M.D., of UPMC; Lilach Bien, Ronen Heled, Daphna Laifenfeld, Ph.D., Chaim Linhart, Judith Sandbank, M.D., Manuela Vecsler, of Ibex Medical Analytics; Anat Albrecht-Shach, M.D., of Shamir Medical Center; Varda Shalev, M.D., M.P.A., of Maccabbi Healthcare Services; and Pamela Michelow, M.S., and Scott Hazelhurst, Ph.D., of the University of the Witwatersrand.

___
Listen to KDKA Radio on the RADIO.COM App
Tell your smart speaker to "play News Radio 10 20 KDKA"
Receive email updates from KDKA Radio
Follow KDKA Radio: Facebook | Twitter | Instagram