Micrograph showing prostatic acinar adenocarcinoma - the most common form of...
Micrograph showing prostatic acinar adenocarcinoma - the most common form of prostate cancer.
Source: Wikipedia

Using AI to detect and grade prostate cancer

Researchers at Karolinska Institutet have developed a method based on artificial intelligence for histopathological diagnosis and grading of prostate cancer. The AI-system has the potential to solve one of the bottlenecks in today’s prostate cancer histopathology by providing more accurate diagnosis and better treatment decisions.

The study shows that the AI-system is as good at identifying and grading prostate cancer as world-leading uro-pathologists. “Our results show that it is possible to train an AI-system to detect and grade prostate cancer on the same level as leading experts,” says Martin Eklund, associate professor at the Department of Medical Epidemiology and Biostatistics at Karolinska Institutet who led the study. “This has the potential to significantly reduce the workload of uro-pathologists and allow them to focus on the most difficult cases.”

A problem in today’s prostate pathology is that there is a certain degree of subjectivity in the assessments of the biopsies. Different pathologists can reach different conclusions even though they are studying the same samples. This leads to a clinical problem where the doctors must pick treatment based on ambiguous information. In this context, the researchers see significant potential to use the AI-technology to increase the reproducibility of the pathological assessments.

To train and test the AI system, the researchers digitized more than 8,000 biopsies taken from some 1,200 Swedish men in the ages of 50–69 to high-resolution images using digital pathology scanners. About 6,600 of the samples were used to train the AI system to spot the difference between biopsies with or without cancer. The remaining samples, and additional sets of samples collected from other labs, were used to test the AI system. Its results were also compared against the assessments of 23 world-leading uro-pathologists. The study was conducted in collaboration with researchers at Tampere University in Finland.

Near-perfect detection

The findings showed that the AI-system was almost near-perfect in determining whether a sample contained cancer or not, as well as in estimating the length of the cancer tumor in the biopsy. When it comes to determining the severity of the prostate cancer, the so-called Gleason score, the AI system was on par with the international experts. “When it comes to grading the severity of the prostate cancer, the AI is in the same range as international experts, which is very impressive, and when it comes to diagnostics, to determine whether or not it is cancer, the AI is simply outstanding,” says Lars Egevad, professor in pathology at Karolinska Institutet and co-author of the study.

The initial findings are promising but more validation is needed before the AI system may be rolled out broadly in clinical practice, according to the researchers. That is why a multicenter study spanning nine European countries is currently underway with completion slated by the end of 2020. That study, which is financed by EIT Health, aims to train the AI-system to recognize cancer in biopsies taken from different laboratories, with different types of digital scanners and with very rare growth patterns. In addition, a randomized study starting in 2020 will examine how the AI-model may be implemented in Sweden’s health care system. “AI-based evaluation of prostate cancer biopsies could revolutionize future health care,” says Henrik Grönberg, professor in cancer epidemiology at Karolinska Institutet and head of the Prostate Cancer Center at S:t Göran Hospital in Stockholm. “It has the potential to improve the diagnostic quality, and thereby secure a more equitable care at a lower cost.”

“The idea is not that AI should replace the human involvement, but rather act as a safety net to ensure that pathologists don’t miss some cancers, and assist in standardization of grading,” says Martin Eklund. “It could also serve as an alternative in parts of the world where there is a complete lack of pathological expertise today.”

Subscribe to our newsletter

Related articles

Using AI to predict risk of thyroid cancer on ultrasound

Using AI to predict risk of thyroid cancer on ultrasound

Researchers from Thomas Jefferson University use machine learning on ultrasound images of thyroid nodules to predict risk of malignancy.

Hybrid microscope could bring digital biopsy to the clinic

Hybrid microscope could bring digital biopsy to the clinic

By adding infrared capability to the ubiquitous, standard optical microscope, researchers hope to bring cancer diagnosis into the digital era.

AI tracks down acute myeloid leukemia

AI tracks down acute myeloid leukemia

Artificial intelligence can detect one of the most common forms of blood cancer—acute myeloid leukemia (AML)—with high reliability.

ConvPath software uses AI to identify cancer cells

ConvPath software uses AI to identify cancer cells

A software tool uses artificial intelligence to recognize cancer cells from digital pathology images — giving clinicians a powerful way of predicting patient outcomes.

Sensor chip detects cancer biomarker in urine

Sensor chip detects cancer biomarker in urine

Researchers have used a chip-based sensor with an integrated laser to detect very low levels of a cancer protein biomarker in a urine sample.

Combination of AI and radiologists accurately identified breast cancer

Combination of AI and radiologists accurately identified breast cancer

An AI tool identified breast cancer with approximately 90 percent accuracy when combined with analysis by radiologists.

Machine learning improves diagnostics of head and neck cancers

Machine learning improves diagnostics of head and neck cancers

Researchers used artificial intelligence to develop a new classification method which identifies the primary origins of cancerous tissue based on chemical DNA changes.

AI classifies lung cancer slides at pathologist level

AI classifies lung cancer slides at pathologist level

Researchers have utilized machine learning capabilities to assist with the challenging task of grading tumor patterns and subtypes of lung adenocarcinoma.

Lab-on-a-chip detects cancer less invasively

Lab-on-a-chip detects cancer less invasively

A new ultrasensitive diagnostic device could allow doctors to detect cancer quickly from a droplet of blood or plasma, leading to timelier interventions and better outcomes for patients.

Popular articles