Prostate cancer is the second most common cancer found in Indian males with an incidence rate of 9-10 cases per one lakh population, according to data from the Indian Council of Medical Research (ICMR).?
This is higher than other parts of Asia and Africa and not far behind western countries with larger caseload. Globally, prostate cancer is the fifth leading cause of death by cancer and recorded more than a million caseload and 3.5 lakh deaths in 2018.
It usually affects men in the age group of 65+ years, however, there are an increasing number of cases in younger men residing in metropolitan cities. Prostate cancer often shows no symptoms in its early stages -- when it’s still confined to the prostate gland -- and its diagnosis techniques are either invasive, resource intensive, or has low efficacy, making widespread early detection an arduous task.
If detected in its early stages, the survival rates of patients are high due to slow progression of the disease. Most men with prostate cancer often die of unrelated causes, without ever knowing that they had the disease.
Now, Australian researchers have developed an artificial intelligence (AI) program that can detect early signs of prostate cancer by analysing routine computed tomography (CT) scans.
The study, published in Nature's Scientific Reports, saw researchers from RMIT University and St Vincent's Hospital Melbourne train the AI algorithm by analysing CT scans of asymptomatic patients -- with and without prostate cancer -- for tell-tale signs of the disease.
Researchers said that CT scans are great for detecting bone and joint problems but even radiologists struggle to spot prostate cancers on the images. High radiation doses which may lead to adverse effects in the long term also make this process unsuitable for regular cancer screening. The AI solution, on the other hand, could be used to run a cancer check whenever men have their abdomen or pelvis scanned for other issues.
"We've trained our software to see what the human eye can't, with the aim of spotting prostate cancer through incidental detection," staff author Dr Ruwan Tennakoon said in a statement. "It's like training a sniffer dog -- we can teach the AI to see things that we can't with our own eyes, in the same way a dog can smell things human noses can't."
The team trained the AI software to look for features of disease in a variety of scans and where exactly to look for them, avoiding the need to manually crop the images.
The machine’s performance was evaluated by comparing a dataset with 571 CT scans of the abdomen and pelvic region using cross-validation techniques with that of expert radiologists. Researchers found that not only did the AI perform better but also detected cancerous growths in just seconds.
Moreover, the AI improved with each scan, learning and adapting to read images from different machines to spot even the smallest irregularities.
Dr Mark Page from St Vincent's Hospital Melbourne said that this technology could allow healthcare workers to catch prostate cancer cases early in patients.
"For example, emergency patients who have CT scans could be simultaneously screened for prostate cancer. If we can detect it earlier and refer them to specialist care faster, this could make a significant difference to their prognosis," he added.