Artificial intelligence as a digital colleague improves prostate diagnosis

Medical Facts Editorial / Janine Padding January 19, 2022 – 08:14

Computer is better than pathologist, but together they are better

Artificial intelligence (AI) is technically ready for use in a prostate cancer analysis clinic, thanks to a global programming competition published in Nature Medicine. The algorithms are superior to the average pathologist, but even better is the combination of artificial intelligence and the pathologist, where the computer gives a digital second opinion. This was demonstrated by doctoral student Wouter Bolten in his thesis.

To diagnose prostate cancer, which is the most common type of cancer in men, microscopic analysis of suspicious prostate tissue by a pathologist is critical. But pathologists are subjective and often differ in their assessment. With potentially serious consequences for the patient. Wouter Bulten, a doctoral student at Radboudumc, demonstrated in his thesis that artificial intelligence (AI) assesses prostate tissue as well as pathologists. He investigated this with project leader Geert Letgens, on a data set of more than 1,000 patients from Radbodomic, thanks in part to support from the NWO and KWF Alpe d’HuZes.

a challenge

Based on this finding, Bulten, Litjens and their team, the Karolinska Institute and Google Health, organized a major follow-up study, the PANDA Challenge. The goal is to test whether computers can improve diagnosis around the world. Radboudumc and Karolinska combined their databases to build the largest and freely accessible data set in pathology, with images of more than ten thousand prostate cancer biopsies. They then challenged AI experts around the world to develop an algorithm for digital analysis in three months.

“We had over a thousand participants,” Bolten says. “Everything was common to all participants, which allowed them to work together to come up with very powerful solutions. After ten days, one of the algorithms was already at the level of an ordinary pathologist and then even better. The creators of the three best algorithms received a cash prize. Then we followed up with The top 15 for more comprehensive and independent analysis. For example, we have developed a blueprint for a large-scale assessment of AI in pathology. AI is now ready for the technology for clinical application.”

finest

The PANDA study underscores the power of AI in diagnosis on a global scale. Will the computer soon replace the coroner? Bolten: I hear this question regularly in the media. But I don’t think so. Our follow-up research shows that pathologists using AI diagnose prostate cancer better and more consistently. AI alone does a better job than a normal pathologist, but a pathologist with AI is superior. The same applies here: two know more than one.

Bulten sees an important role for AI in improving diagnoses by pathologists: “Quality improves when a pathologist uses AI as a digital second opinion, especially with less experienced clinicians. Thanks to AI, we’ve got better pathologists who agree a lot. AI can also Speeding up diagnosis, making healthcare more affordable. Plus, you can use AI to assess many tissues very quickly in a screening population. Then the pathologist only has to look at suspicious biopsies.”

workflow

The question is how long it will take before AI is applied standard in clinical practice. “Now we need to demonstrate that more accurate diagnoses using AI actually improves outcomes in prostate cancer. And that costs of care are declining. To this end, we want to integrate AI into the current workflow. At the moment, only 10 percent of hospitals are in use. All over the world the digital images, while the other 90 percent see biopsies only through a microscope. So we can make huge gains in that.”

According to the researchers, the algorithm can show not only whether the structure is neoplastic, but also how aggressive the tumor is, the so-called tumor classification. Promoter Jeroen van der Laak: “Our programs therefore provide a great deal of information. eosin, a division of Radboudumc, is now continuing to work on software for various applications of artificial intelligence in pathology in the clinic. This is how we create value for the patient.”

Source: Radboudumc

Medical Facts Editorial / Janine Padding

We have specialized in interactive news for healthcare providers, so that healthcare providers are informed every day of news that may be relevant to them. Placing both news and news for healthcare providers and prescriptions. Social media, women’s health, patient advocacy, patient empowerment, personalized medicine and care 2.0, and the social sphere are some of the most important things you should pay more attention to.

I have studied physical therapy and healthcare business management. I am also a registered independent client supporter and unofficial sponsorship broker. I have extensive experience in various positions in healthcare, social field, medicine and pharmaceutical industry both nationally and internationally. They have extensive medical knowledge of most healthcare specialties. and the health care laws by which health care is regulated and financed. Every year I attend most of the leading medical conferences in Europe and America to keep my knowledge up-to-date and keep abreast of the latest developments and innovations. I am currently studying applied psychology.

My messages in this weblog do not reflect the strategy, policy or direction of my employer, nor are they for business or for a client or business owner.

See also  Smart glove improves post-stroke rehabilitation - ICT and Health