Using AI to save lives. Depositphotos –
An international team of scientists including those from Australia’s Charles Darwin University (CDU) has developed a novel AI model known as ECgMLP, which can assess microscopic images of cells and tissue to identify endometrial cancer – one of the most common forms of reproductive tumors – with an impressive 99.26% accuracy. And the researchers say it can be adapted to identify a broad range of disease, including colorectal and oral cancer.
“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 the study’s co-author Dr. Asif Karim, from CDU. “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.”
What that “science-speak” means is that the well-trained model is able to look at these microscopic scans – histopathology images – and enhance image quality in order to identify early stages of cancer, homing in on certain areas of the scans to pinpoint problematic growth that may not be easily detected by the naked eye. Right now, current human-led diagnostic methods are around 78.91% to 80.93% accurate. Endometrial cancer is treatable and, if found in time, has a good five-year outcome for patients. However, once it spreads outside the uterus, it becomes difficult to effectively treat – which makes timely diagnosis critical in saving lives.
–
Currently, more than 600,000 Americans have battled the disease. And while this cancer may not personally impact half of the population, the scientists confirm that ECgMLP analysis has much broader application than what it has been trained on.
“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 co-author Niusha Shafiabady, an associate professor at ACU (Australian Catholic University). “We evaluated the model on several histopathology image datasets. It diagnosed colorectoral cancer with 98.57% per cent accuracy, breast cancer with 98.20% accuracy, and oral cancer with 97.34% accuracy.”
–
[For the balance of this extremely important article, please visit: https://newatlas.com/cancer/ai-cancer-diagnostic/]
–
The study was published in the journal Computer Methods and Programs in Biomedicine Update.
Source: Charles Darwin University
–
Tags
