Jakarta, INTI – In the latest advancements in the medical field, artificial intelligence (AI) as demonstrated its ability to detect cancer with accuracy far surpassing conventional methods. An international team of scientists, including researchers from Charles Darwin University (CDU) in Australia, has developed a new AI model called ECgMPL, capable of identifying endometrial cancer with an accuracy rate of 99.26%.
Breakthrough in Cancer Diagnostics
This AI model is designed to analyze microscopic images of cells and tissues using histopathology technology. By enhancing image quality and highlighting crucial areas in tissue analysis, AI can detect early signs of cancer that may be difficult to identify with the human eye. Currently, human-based diagnostic methods for endometrial cancer have an accuracy rate of only about 78.91% to 80.93%, making this technology a highly promising innovation.
According to Dr. Asif Karim from CDU, the ECgMPL model has been optimized through various methods, such as ablation studies, self-attention mechanisms, and efficient training. This allows AI to work more effectively across multiple histopathology datasets, making it a robust and clinically applicable solution for endometrial cancer diagnosis.
Broad Applications in the Medical Field
Beyond endometrial cancer, this AI model can also be applied to detect various other types of cancer. Based on evaluations of multiple histopathology image datasets, the model has demonstrated high accuracy in diagnosing other cancers:
- Colorectal cancer: 98.57%
- Breast cancer: 98.20%
- Oral cancer: 97.34%
With such extensive capabilities, ECgMPL is not only a faster and more accurate diagnostic tool but also more accessible and affordable for patients. According to Associate Professor Niusha Shafiabady from Australian Catholic University, this technology holds great potential in improving patient treatment outcomes through early disease detection.
Collaboration Between AI and Medical Professionals
Although AI has demonstrated remarkable accuracy, it is not intended to replace medical professionals. Instead, AI will serve as an assistive tool for specialists, helping them make more precise decisions in diagnosing and monitoring cancer treatment effectiveness. AI can function as the "brain" of a software system that supports the diagnostic process, enhancing efficiency and reducing the likelihood of errors in cancer identification.
Conclusion
The development of AI in cancer diagnosis, such as the ECgMPL model, marks a significant advancement in the medical field. With near-perfect accuracy and the ability to detect various types of cancer, this technology has the potential to save countless lives through faster and more accurate early detection. AI is not just a tool for doctors but also a new beacon of hope for patients worldwide. As research continues to evolve, the future of cancer diagnosis looks increasingly promising for global health.
This research has been published in the journal Computer Methods and Programs in Biomedicine Update and is expected to be increasingly adopted in medical practice in the future. With these advancements, the hope for earlier and more accurate cancer diagnoses grows, ultimately improving survival rates for many patients worldwide.
Read More : Setiaji - #3 Pros and Cons of AI in Indonesia’s Healthcare Industry | INTI TALK