Charles Darwin University lend hand in developing new AI model to detect endometrial cancer with 99.26 per cent accuracy

An AI model could soon be used to help detect one of Australia’s most common gynaecological cancers, according to Charles Darwin University.

A new artificial intelligence model could soon be used to help detect one of Australia’s most common gynaecological cancers. Picture: Pema Tamang Pakhrin

A new artificial intelligence model could soon be used to help detect one of Australia’s most common gynaecological cancers, according to Charles Darwin University.

Researchers from Daffodil International University in Bangladesh, CDU, the University of Calgary and Australian Catholic University developed the AI, which can reportedly detect endometrial cancer with 99.26 per cent accuracy.

Endometrial cancer is the most common gynaecological cancer in Australia, and one of the most diagnosed cancers in Australian women, according to the Cancer Council.

Photometric augmentation is applied to the data. Picture: CDU

The model, called ECgMPL, examines histopathological images, which are microscopic images of tissue used in disease analysis, particularly for cancer diagnosis.

CDU say the model enhances the quality of the images, identifies the most important areas and analyses the tissue.

The current endometrial accuracy using automated diagnosis is reported to be about 78.91 to 80.93 per cent.

CDU/ACU Associate Professor and study co-author Niusha Shafiabady. Picture: Supplied

But the model could also have benefits outside of endometrial cancer diagnosis, according to Co-author and CDU adjunct Associate Professor Niusha Shafiabady, who is also an Associate Professor at Australian Catholic University.

“The same methodology can be applied for fast and accurate early detection and diagnosis of other diseases which ultimately leads to better patient outcomes,” Ms Shafiabady said.

“We evaluated the model on several histopathology image datasets. It diagnosed colorectal cancer with 98.57 per cent accuracy, breast cancer with 98.20 per cent accuracy, and oral cancer with 97.34 per cent accuracy.

“The core AI model developed through this research can be adopted as the brain of a software system to be used to assist the doctors for decision-making in cancer diagnosis.”

The images analysed by the AI model. Picture: CDU

Co-author and CDU Lecturer in Information Technology Dr Asif Karim said the model could enhance clinical processes.

“The proposed ECgMLP model outperforms existing methods by achieving 99.26 per cent accuracy, surpassing transfer learning and custom models discussed in the research while being computationally efficient,” Dr Karim said.

“Optimised through ablation studies, self-attention mechanisms, and efficient training, ECgMLP generalises well across multiple histopathology datasets thereby making it a robust and clinically applicable solution for endometrial cancer diagnosis.”

ECgMLP: A novel gated MLP model for enhanced endometrial cancer diagnosis, was published in the journal Computer Methods and Programs in Biomedicine Update.

Originally published as Charles Darwin University lend hand in developing new AI model to detect endometrial cancer with 99.26 per cent accuracy