0
0
A groundbreaking artificial intelligence (AI) model has demonstrated near-perfect accuracy in diagnosing endometrial cancer, potentially revolutionizing early detection and improving patient outcomes, according to new research.
The study, published in the journal Computer Methods and Programs in Biomedicine Update, reveals that the AI model, developed by researchers from Daffodil International University, Charles Darwin University (CDU), the University of Calgary, and Australian Catholic University, achieved an impressive 99.26% accuracy in detecting endometrial cancer.
Endometrial cancer, the most common gynecological cancer in Australia, poses a significant health challenge. The new AI model, named ECgMLP, analyzes histopathological images—microscopic tissue images used for disease diagnosis—with remarkable precision. It enhances image quality, pinpoints crucial areas, and meticulously analyzes the tissue.
This advancement significantly surpasses current automated diagnostic methods, which typically achieve accuracy rates between 78.91% and 80.93%.
“The proposed ECgMLP model outperforms existing methods by achieving 99.26 percent accuracy, surpassing transfer learning and custom models discussed in the research while being computationally efficient,” said Dr. Asif Karim, co-author and CDU Lecturer in Information Technology. “Optimized through ablation studies, self-attention mechanisms, and efficient training, ECgMLP generalizes well across multiple histopathology datasets, thereby making it a robust and clinically applicable solution for endometrial cancer diagnosis.”
The model’s potential extends beyond endometrial cancer. Researchers also tested ECgMLP on other histopathology image datasets, yielding high accuracy rates: 98.57% for colorectal cancer, 98.20% for breast cancer, and 97.34% for oral cancer.
“The same methodology can be applied for fast and accurate early detection and diagnosis of other diseases which ultimately leads to better patient outcomes,” said Associate Professor Niusha Shafiabady, co-author and CDU adjunct Associate Professor, who is also an Associate Professor at Australian Catholic University. “The core AI model developed through this research can be adopted as the brain of a software system to be used to assist doctors with decision-making in cancer diagnosis.”
The researchers believe this AI model could significantly enhance clinical processes, leading to earlier and more accurate diagnoses, ultimately improving patient care.
More information: Md. Alif Sheakh et al, ECgMLP: A novel gated MLP model for enhanced endometrial cancer diagnosis, Computer Methods and Programs in Biomedicine Update (2025). DOI: 10.1016/j.cmpbup.2025.100181
Disclaimer: While this research demonstrates promising results, it is important to note that AI diagnostic tools are intended to assist, not replace, medical professionals. Clinical diagnosis should always be made by a qualified healthcare provider, considering all relevant factors. Further research and validation are necessary before widespread clinical implementation.
