AI enables early diagnosis – healing practice

Detect prostate cancer without uncomfortable exams?

To successfully treat prostate cancer, it is extremely important that the disease is diagnosed early. Artificial intelligence (AI) along with routine computed tomography could help greatly improve the early diagnosis of prostate cancer and save so many men’s lives.

The new program developed at RMIT University in Melbourne uses artificial intelligence (AI) to diagnose prostate cancer using computed tomography (CT). The technology was developed in collaboration with experts at St. Vincent’s Hospital Melbourne and analyzes CT scans for tell-tale signs of prostate cancer, which are very difficult to see even with a well-trained human eye. The results of the investigation can be found in the English-language specialist journal “Nature’s Scientific Reports“Can be read.

Computed tomography for cancer screening?

CT imaging is not suitable for regular cancer screening examinations due to the high radiation dose, but the AI ​​solution can also be used for a cancer check if men have their abdomen or pelvis scanned for other problems, the team explains.

Prostate cancer can be difficult to detect on CT scans

So-called CT scans are well suited to detect bone and joint problems, but even radiologists have difficulties recognizing prostate cancer in the images, reports study author Dr. Ruwan Tennakoon from RMIT. “We trained our software to see what the human eye cannot see, with the aim of detecting prostate cancer by chance,” explains the expert.

“It’s like training a tracker dog – we can teach the AI ​​to see things that we can’t see with our own eyes, just as a dog can smell things that human noses can’t,” adds Dr. Tennakoon in one Press release from RMIT University.

For the current study, CT scans of asymptomatic subjects, with and without prostate cancer, were analyzed. The team trained the AI ​​software to look for disease features in a variety of scans and to see exactly where to look without the need to manually crop the images.

AI better than radiologists

The AI ​​performed better than radiologists who viewed the same images. The AI ​​was actually able to detect canker sores within seconds, according to the researchers. In addition, the AI ​​improved with each scan, learning and adapting to read images from different devices and detect even the smallest irregularities in the process.

The study suggests how AI can and should be used for the common good, explains study author Professor John Thangarajah of RMIT University. The expert adds: “Our healthcare sector needs smarter solutions and AI can help, but we are only scratching the surface. There is a lot of good that Artificial Intelligence can bring to the world, which is our focus at RMIT, and this study is a big part of that. ”

Detecting prostate cancer early is important

The director of the CT diagnostic imaging department at St. Vincent’s Hospital, Dr. Mark Page goes on to say that early intervention for prostate cancer is key to better health. With the help of the new technology, the experts hope to identify early cases of prostate cancer in people who are actually being scanned for other reasons. For example, hospital emergencies that are already doing CT scans could be screened for prostate cancer at the same time.

If prostate cancer is detected earlier, people can be referred to prostate cancer professionals faster, which can make a significant difference in their prognosis, said Dr. Page. The technology could be used on a large scale in the future and possibly combined with a variety of diagnostic imaging devices (MRI and DEXA). However, further research is still pending. (as)

Author and source information

This text complies with the requirements of specialist medical literature, medical guidelines and current studies and has been checked by medical professionals.


  • RMIT University: New AI tech for early detection of prostate cancer (veröffentlicht 08.07.2021), RMIT University
  • Ruwan Tennakoon, Mark Page, Peter Brotchie, John Thangarajah, Cosmin Florescu et al.: Incidental detection of prostate cancer with computed tomography scans, in Nature’s Scientific Reports (veröffentlicht 12.04.2021), Nature’s Scientific Reports

Important NOTE:
This article is for general guidance only and is not intended to be used for self-diagnosis or self-treatment. He can not substitute a visit at the doctor.


Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.