News, Analysis, Trends, Management Innovations for
Clinical Laboratories and Pathology Groups

Hosted by Robert Michel

News, Analysis, Trends, Management Innovations for
Clinical Laboratories and Pathology Groups

Hosted by Robert Michel
Sign In

UCLA Spinoff Develops AI Tool That Improves Accuracy of Prostate Cancer Assessments

Software analyzes imaging scans and clinical laboratory data to help oncologists and anatomic pathologists visualize a tumor’s extent

Anatomic pathologists understand that, along with breast cancer, diagnostic testing for prostate cancer accounts for a high volume of clinical laboratory tests. Thus, a recent study indicating that a new artificial intelligence (AI)-based software tool can dramatically improve physicians’ ability to identify the extent of these cancers will be of interest.

The software, known as Unfold AI, was developed by Avenda Health, a University of California Los Angeles (UCLA) spinoff company. Unfold AI, according to Avenda, predicts focal therapy success by an increase of 77% over standard methods.

“The study found that Unfold AI’s patient-specific encapsulation confidence score (ECS), which is generated based on multiple patient data points, including MRI scans, biopsy results, PSA [prostate-specific antigen] data, and Gleason scores, is critical for predicting treatment success,” an Avenda press release states. “These findings emphasize the importance of Unfold AI’s assessment of tumor margins in predicting treatment outcomes, surpassing the predictive capability of conventional parameters.”

“Unfold AI’s ability to identify tumor margins and provide the ECS will improve treatment recommendations and allow for less-invasive interventions,” said study co-author Wayne Brisbane, MD, a urologic oncologist and UCLA medical professor, in another press release. “This more comprehensive approach enhances our ability to predict treatment outcomes and tailor interventions effectively to individual patient needs.”

The UCLA researchers published their findings titled, “Artificial Intelligence Improves the Ability of Physicians to Identify Prostate Cancer Extent,” in The Journal of Urology. Results were also presented at the 2024 American Urological Association annual meeting.

“This study is important because it shows the ability of AI to not only replicate expert physicians, but to go beyond human ability,” said study co-author Wayne Brisbane, MD (above), a urologic oncologist and UCLA medical professor, in a press release. “By increasing the accuracy of cancer identification in the prostate, more precise and effective treatment methods can be prescribed for patients.” Clinical laboratories that work with anatomic pathologists to diagnose prostate and other cancers may soon have a new AI testing tool. (Photo copyright: UCLA.)

How Unfold AI Works

To gauge the extent of prostate tumors, surgeons typically evaluate results from multiple diagnostic methods such as PSA tests and imaging scans such as MRIs, according to a UCLA press release. However some portions of a tumor may be invisible to an MRI, causing doctors to underestimate the size.

Unfold AI, originally known as iQuest, was designed to analyze data from PSA, MRI, fusion biopsy, and pathology testing, according to a company brochure. From there, it generates a 3D map of the cancer. Avenda’s website says the technology provides a more accurate representation of the tumor’s extent than conventional methods.

“Accurately determining the extent of prostate cancer is crucial for treatment planning, as different stages may require different approaches such as active surveillance, surgery, focal therapy, radiation therapy, hormone therapy, chemotherapy, or a combination of these treatments,” Brisbane said in the UCLA press release.

Putting AI to the Test

In the new study, the UCLA researchers enlisted seven urologists and three radiologists to review 50 prostate cancer cases. Each patient had undergone prostatectomy—surgical removal of all or part of the prostate—but might have been eligible for focal therapy, a less-aggressive approach that uses heat, cryotherapy, or electric shocks to attack cancer cells more selectively.

The physicians came from five hospitals and had a wide range of clinical experience from two to 23 years, the researchers noted in The Journal of Urology.

They reviewed clinical data and examined MRI scans of each patient, then “manually drew outlines around the suspected cancerous areas, aiming to encapsulate all significant disease,” the press release states. “Then, after waiting for at least four weeks, they reexamined the same cases, this time using AI software to assist them in identifying the cancerous areas.”

The researchers analyzed the physicians’ work, evaluating the accuracy of the cancer margins and the “negative margin rate,” indicating whether the clinicians had identified all of the cancerous tissue. Using conventional approaches, “doctors only achieved a negative margin 1.6% of the time,” the press release states. “When assisted by AI the number increased to 72.8%.”

The clinicians’ accuracy was 84.7% when assisted by AI versus 67.2% to 75.9% for conventional techniques.

They also found that clinicians who used the AI software were more likely to recommend focal therapy over more aggressive forms of treatment.

“We saw the use of AI assistance made doctors both more accurate and more consistent, meaning doctors tended to agree more when using AI assistance,” said Avenda Health co-founder and CEO Shyam Natarajan, PhD, who was senior author of the study.

“These results demonstrate a marked change in how physicians will be able to diagnose and recommend treatment for prostate cancer patients,” said Natarajan in a company press release. “By increasing the confidence in which we can predict a tumor’s margins, patients and their doctors will have increased certainty that their entire tumor is treated and with the appropriate intervention in correlation to the severity of their case.”

Already Cleared by FDA

Avenda received FDA 510(k) clearance for Unfold AI in November 2022. On July 1, 2024, the American Medical Association (AMA) implemented a CPT [Current Procedural Terminology] Category III code for the software, enabling insurance reimbursement for services that employ the technology, the company said in a press release.

The AMA describes CPT Category III as “a temporary set of codes for emerging technologies, services, procedures, and service paradigms.”

In the same press release, Avenda revealed that the federal Centers for Medicare and Medicaid Services (CMS) had assigned a national payment rate for Unfold AI.

UCLA’s study found that AI can outperform doctors both in sensitivity (a higher detection rate of positive cancers) and specificity (correctly detecting the sample as negative). That’s relevant and worth watching for further developments.

Pathologists and clinical laboratory managers should consider this use of AI as one more example of how artificial intelligence can be incorporated into diagnostic tests in ways that allow medical laboratory professionals to diagnose disease earlier and more accurately. This will improve patient care because early intervention for most diseases leads to better outcomes.

—Stephen Beale

Related Information:

New Study Proves AI Enhances Physicians’ Ability to Identify Prostate Cancer Extent with 84 Percent Accuracy

New Study Demonstrates Avenda Health’s Unfold AI to Better Predict Focal Therapy Success by 77% as Compared to Standard Methods

AI Model May Yield Better Outcomes for Prostate Cancer

Artificial Intelligence Improves the Ability of Physicians to Identify Prostate Cancer Extent

Artificial Intelligence Detects Cancer with 25% Greater Accuracy than Doctors in UCLA Study

Study Finds Unfold AI Better Predicts Focal Therapy Success in Prostate Cancer Patients

First AI-Powered Precision Oncology Platform for Prostate Cancer Care, iQuest Receives FDA Clearance

UPMC Researchers Develop Artificial Intelligence Algorithm That Detects Prostate Cancer with ‘Near Perfect Accuracy’ in Effort to Improve How Pathologists Diagnose Cancer

Working from tissue slides similar to those used by surgical pathologists, the algorithm accurately detects prostate cancer with an impressive 98% sensitivity

It could be that a new milestone has been reached on the road to using artificial intelligence (AI) to help anatomic pathologists diagnose cancer and other diseases. A research collaboration between a major American university and an Israeli company recently published a study about the ability of an AI algorithm to correctly diagnose prostate cancer.

The collaboration involved researchers at the University of Pittsburgh Medical Center (UPMC) and at Ibex Medical Analytics of Israel. The research team created an AI algorithm dubbed the Galen Prostate (part of the Galen Platform). In the study, the Galen Prostate AI accurately detected prostate cancer with 98% sensitivity and 97% specificity.

Researchers noted that this level of diagnostic sensitivity and specificity was significantly higher, compared to previously tested cancer-detecting algorithms that utilized tissue slides. The UPMC scientists published their findings in The Lancet Digital Health, titled, “An Artificial Intelligence Algorithm for Prostate Cancer Diagnosis in Whole Slide Images of Core Needle Biopsies: A Blinded Clinical Validation and Deployment Study.”

AI Show and Tell in Anatomic Pathology

The scientists trained the Galen Prostate AI to recognize prostate cancer by having it examine images from over a million parts of stained tissue slides taken from patient biopsies. Expert pathologists labeled each image to teach the algorithm how to distinguish between healthy and abnormal tissue. The AI was then tested on 1,600 different tissue slide images that had been collected from 100 patients seen at UPMC who were suspected of having prostate cancer.  

“Humans are good at recognizing anomalies, but they have their own biases or past experience,” said Rajiv Dhir, MD, Chief Pathologist and Vice Chair of Pathology at UPMC Shadyside Hospital, Professor of Biomedical Informatics at University of Pittsburgh, and senior author of the study, in a UPMC news release. “Machines are detached from the whole story. There’s definitely an element of standardizing care.”

Ibex Galen Prostate AI solution
The image above is “of prostate cancer (represented by the heatmap) detected by the Ibex Galen Prostate [AI] solution on a biopsy that was previously diagnosed as benign by the pathologist,” stated an Ibex news release announcing the UPMC study. (Photo copyright: Ibex.)

UPMC Algorithm Goes Beyond Cancer Detection, Exceeds Human Pathologists

The researchers also noted that this is the first algorithm to extend beyond cancer detection. It reported high performance for tumor grading, sizing, and invasion of surrounding nerves—clinically important features of pathology reports.  

“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.”

The algorithm also flagged six slides as potentially containing abnormal tissue that were not flagged by human pathologists. However, the researchers pointed out that this difference does not mean the AI is better than humans at detecting prostate cancer. It is probable, for example, that the pathologists simply saw enough evidence of malignancy elsewhere in the patients’ samples to recommend treatment.

Other Studies Where AI Detected Prostate Cancer

The UPMC researchers are not the first to use AI to detect prostate cancer. In February, The Lancet Oncology published a study from researchers at Radboud University Medical Center (RUMC) in the Netherlands who developed a deep learning AI system that could determine the aggressiveness of prostate cancer in certain patients. 

For that research, the RUMC scientists collected 6,000 biopsies from more than 1,200 men. They then showed the biopsy images along with the original pathology reports to their AI system. Using deep learning, the AI was able to detect and grade prostate cancer according to the Gleason Grading System (aka, Gleason Score), which is used to rate prostate cancer and choose appropriate treatment options. The Gleason Score ranges from one to five and most cancers obtain a score of three or higher. 

“Systems such as ours can be used in different ways. First, it can be used to screen biopsies and to filter out the easy (benign) cases. This could reduce the workload for pathologists,” said Wouter Bulten, a PhD candidate at Radboud who worked on the study, in an interview with HemOnc Today. “Second, the system can be used as a second opinion after the pathologist’s initial read. The system can flag a case if its opinion differs from that of the pathologist. It also can give feedback during the first read, showing the pathologist where to look. In this case, the pathologist needs only to confirm the opinion of the AI system.” 

Can Today’s AI Outperform Human Pathologists?

In their research, the Radboud team discovered that their AI system was able to achieve pathologist-level performance and, in some cases, even performed better than human pathologists. However, they do not foresee AI replacing the need for pathologists, but rather emerging as another method to use in cancer detection and treatment.  

“We see our system as an additional tool that the pathologist can use. Although our system performs very well, it still makes mistakes,” stated Bulten. “These mistakes are often different from those a human would make. We believe that when you merge the expertise of the pathologist with the second opinion of an AI system, you get the best of both worlds.” 

According to the American Cancer Society, prostate cancer is the second most common cancer among men in the US, after skin cancer. The organization estimates there will be approximately 191,930 new cases of prostate cancer diagnosed and about 33,330 deaths from the disease in the US in 2020. 

Though the UPMC study focused only on prostate cancer, the scientists believe their algorithm can be trained to detect other types of cancer as well. AI in clinical diagnostics is clearly progressing, however more studies will be required. Nevertheless, if AI can truly become a useful tool for anatomic pathologists to detect cancer earlier, we may see a welcomed reduction in cancer deaths.   

—JP Schlingman

Related Information:

Newly Developed AI Capable of Identifying Prostate Cancer with “Near-perfect Accuracy”

An Artificial Intelligence Algorithm for Prostate Cancer Diagnosis in Whole Slide Images of Core Needle Biopsies: A Blinded Clinical Validation and Deployment Study

Artificial Intelligence Identifies Prostate Cancer

The Lancet Reports Outstanding Performance of Ibex Medical Analytics’ AI-based Algorithm in a Study at UPMC

Prostate Cancer Can Now be Diagnosed Better Using Artificial Intelligence

AI System Outperforms Pathologists in Identifying Prostate Cancer Aggressiveness

Automated Deep-learning System for Gleason Grading of Prostate Cancer using Biopsies: A Diagnostic Study

New AI Technology Helps Pathologists Spot Cancer

Hospitals Worldwide Are Deploying Artificial Intelligence and Predictive Analytics Systems for Early Detection of Sepsis in a Trend That Could Help Clinical Laboratories, Microbiologists

CMS Considers Using Artificial Intelligence to Battle Fraud; Medical Laboratories Must Ensure Billing Practices Comply with New Federal Affiliation Regulations

Mobile Device Software Companies Are Developing Smartphone Apps That Use Artificial Intelligence to Test for COVID-19, Potentially Bypassing the Clinical Laboratory Altogether

Attention Pathologists! New Prostate Cancer Test Has CPT Code, NCCN Guideline Recommendation, and Potential Market of One Million Prostate Biopsies Annually

OPKO Health’s 4Kscore test predicts the rate of high-risk prostate cancer and may become a useful business case study for other labs developing proprietary diagnostic tests

Clinical laboratories and biotech companies with new medical laboratory tests are struggling to win coverage by Medicare and private payers. How big is this problem? There are currently tens of thousands of molecular diagnostic assays and genetic tests offered for clinical use.

Any lab company seeking to obtain an appropriate Current Procedural Terminology (CPT) code, favorable coverage guidelines, and adequate reimbursement from health insurers for its new lab test faces three big challenges, and they are related. First, payers are simply overwhelmed with requests to review new genetic tests. The flood of new test submissions exceeds the capability of payers to respond.

Most Payers May Not Have Right Scientific Expertise to Evaluate Genetic Tests

Second, most health insurance plans lack physicians and medical professionals who have the necessary experience in laboratory medicine, molecular diagnostics, and genetic medicine to evaluate these lab test submissions in a knowledgeable way. (more…)

;