News | Prostate Cancer | July 28, 2020

Artificial Intelligence Identifies Prostate Cancer with Near-perfect Accuracy

Prostate biopsy with cancer probability (blue is low, red is high). This case was originally diagnosed as benign but changed to cancer upon further review. The AI accurately detected cancer in this tricky case. Image courtesy of Ibex Medical Analytics

Prostate biopsy with cancer probability (blue is low, red is high). This case was originally diagnosed as benign but changed to cancer upon further review. The AI accurately detected cancer in this tricky case. Image courtesy of Ibex Medical Analytics

July 28, 2020 — A study published in The Lancet Digital Health by UPMC and University of Pittsburgh researchers demonstrates the highest accuracy to date in recognizing and characterizing prostate cancer using an artificial intelligence (AI) program. The AI technology used in the study is from of Ibex Medical Analytics.

"Humans are good at recognizing anomalies, but they have their own biases or past experience," said 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. "Machines are detached from the whole story. There's definitely an element of standardizing care."

To train the AI to recognize prostate cancer, Dhir and his colleagues provided images from more than a million parts of stained tissue slides taken from patient biopsies. Each image was labeled by expert pathologists to teach the AI how to discriminate between healthy and abnormal tissue. The algorithm was then tested on a separate set of 1,600 slides taken from 100 consecutive patients seen at UPMC for suspected prostate cancer.

During testing, the AI demonstrated 98% sensitivity and 97% specificity at detecting prostate cancer -- significantly higher than previously reported for algorithms working from tissue slides.

Also, this is the first algorithm to extend beyond cancer detection, reporting high performance for tumor grading, sizing and invasion of the surrounding nerves. These all are clinically important features required as part of the pathology report.

AI also flagged six slides that were not noted by the expert pathologists.

But Dhir explained that this doesn't necessarily mean that the machine is superior to humans. For example, in the course of evaluating these cases, the pathologist could have simply seen enough evidence of malignancy elsewhere in that patient's samples to recommend treatment. For less experienced pathologists, though, the algorithm could act as a failsafe to catch cases that might otherwise be missed.

"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 are promising, Dhir cautions that new algorithms will have to be trained to detect different types of cancer. The pathology markers aren't universal across all tissue types. But he didn't see why that couldn't be done to adapt this technology to work with breast cancer, for example.

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.

Funding for this study was provided by Ibex, which also created this commercially available algorithm. Pantanowitz, Shalev and Albrecht-Shach report fees paid by Ibex, and Pantanowitz and Shalev serve on the medical advisory board. Bien and Linhart are authors on pending patents US 62/743,559 and US 62/981,925. Ibex had no influence over the design of the study or the interpretation of the results.

For more information: www.medschool.pitt.edu

Related Content

A sonographer at Northwestern Memorial Hospital using the Caption Health AI to optimize ultrasound imaging.

A sonographer at Northwestern Memorial Hospital using the Caption Health AI to optimize ultrasound imaging.

News | Ultrasound Imaging | October 28, 2020
October 28, 2020 — Northwestern Memorial Hospital is the first hospital in the United States to purchase Caption Heal
Vinai Gondi, M.D., the Director of Research at the Northwestern Medicine Chicago Proton Center, Co-director of the Brain Tumor Center at Northwestern Medicine Cancer Center, Warrenville, and lead author of the NRG-BN001 abstract.

Vinai Gondi, M.D., director of research at the Northwestern Medicine Chicago Proton Center, co-director of the Brain Tumor Center at Northwestern Medicine Cancer Center, Warrenville, and lead author of the NRG-BN001 abstract.

News | Intensity Modulated Radiation Therapy (IMRT) | October 27, 2020
October 27, 2020 — A...
"Stereotactic radiosurgery versus whole-brain radiation therapy for patients with 4-15 brain metastases: A phase III randomized controlled trial" was presented by Jing Li, M.D., The University of Texas MD Anderson Cancer Center, at ASTRO20
News | Stereotactic Body Radiation Therapy (SBRT) | October 26, 2020
October 26, 2020 — Results of a new randomized phase III trial suggest that...
"Stereotactic ablative fractionated radiotherapy versus radiosurgery for oligometastatic neoplasia to the lung: A randomised phase II trial," presented by Shankar Siva, Ph.D., Peter MacCallum Cancer Centre, was presented at ASTRO20
News | Stereotactic Body Radiation Therapy (SBRT) | October 26, 2020
October 26, 2020 — A new study, conducted across 13 medical centers in Australia and New Zealand, strengthens the cas
"Initial report of a randomized trial comparing conventional- vs conventional plus fluciclovine (18F) PET/CT imaging-guided post-prostatectomy radiotherapy for prostate cancer" was presented by Ashesh Jani, M.D., Winship Cancer Institute of Emory University, at ASTRO20
News | PET Imaging | October 26, 2020
October 26, 2020 — Adding the advanced PET radiotracer...
A randomized phase II/III study comparing 24Gy in 2 stereotactic body radiotherapy (SBRT) fractions versus 20Gy in 5 conventional palliative radiotherapy (CRT) fractions for patients with painful spinal metastases was presented by Arjun Sahgal, M.D., University of Toronto, today at ASTRO20
News | Radiation Therapy | October 26, 2020
October 26, 2020 — A new study shows using fewer and higher doses of high-precision...
Nearly 40 MRIdian-focused presentations will be part of the ASTRO Scientific Sessions
News | Magnetic Resonance Imaging (MRI) | October 26, 2020
October 26, 2020 — ViewRay, Inc. announced that the company's MRIdian Linac MRI-guided radiation therapy system will
Treating lung cancer patients with proton therapy may help reduce the risk of radiation-induced heart diseases, suggests a new study from Penn Medicine. In a retrospective trial of more than 200 patients, mini-strokes were significantly less common among patients who underwent proton therapy versus conventional photon-based radiation therapy. Proton therapy patients also experienced fewer heart attacks.
News | Proton Therapy | October 25, 2020
October 25, 2020 — Treating lung cancer patients with prot...