Endometrium cancer, one of the most common gynecological cancers, can be detected earlier and more accurately thanks to Artificial Intelligence (AI). The new model is expected to go beyond the diagnosis of this type of cancer.
A study in the latest edition of Computer Methods and Programs in Biomedicine revealed a new model of AI capable of holding cancers with 99.26% accuracy.
As writes in a Charles Darwin University (CDU), in Australia, which was part of the investigation, the model – called ECgMPL – Examine histopathological images (microscopic tissue images used in disease analysis), being able to identify the most important areas and analyze the tissue.
“The proposed ECGMLP model exceeds existing methods by reaching 99.26%accuracy, surpassing transfer learning and other personalized models, while computationally efficient,” he has praised Asif Karimstudy co -author and technology teacher at CDU.
“ECGMLP is generalized well through multiple histopathological data sets, thus making it a robust and clinically applicable solution to Diagnosis of endometrium cancer“He added.
Endometrium cancer is currently the most common gynecological cancer in the world. If detected on time, it is perfectly treatable.
However, When it spreads out of the womb, the diagnosis is complicated. The timely diagnosis is therefore fundamental to save lives. And this new AI model is excellent news for the medical community.
Effective in other types of cancer
Niusha Shafiabady of the Australian Catholic University said the model will also bring Benefits beyond the diagnosis of endometrium cancer.
“The same methodology can be applied to early detection and rapid and accurate diagnosis of other diseases, which ultimately leads to better results for patients,” he said.
“We evaluated the model in various sets of histopathological images. Colorectal Cancro com 98,57% precision, the Cancro da mother com 98,20% of precision and the cancro oral com 97,34% of precision.
“The central model of AI developed through this investigation can be adopted as the brain of a software system to be used to help doctors make decisions in the diagnosis of cancer,” added the expert, quoted in the announcement.
This is not a tool designed to replace health professionals, but to be used in collaboration with cancer specialists to accurately detect the disease and control the success of treatment.
