Related Experiment Video
Updated: May 27, 2025
08:12
Detection of a Circulating MicroRNA Custom Panel in Patients with Metastatic Colorectal Cancer
Published on: March 14, 2019
5.4K
Leveraging Radiomics and Hybrid Quantum-Classical Convolutional Networks for Non-Invasive Detection of Microsatellite
T Buvaneswari1, M Ramkumar2, Prabhu Venkatesan3
1Department of Computer Secience and Engineering, Annapoorana Engineering College (Autonomous), NH_47, Sankari Main Road, Periyaseeragapaddi, Salem, Tamil Nadu, 636 308, India. buvanamuruga2008@gmail.com.
Molecular Imaging and Biology
|February 20, 2025
Summary
This study introduces a new AI framework for identifying microsatellite instability (MSI) status in colorectal cancer. The advanced radiomics and deep learning model achieves 99% accuracy, improving cancer diagnosis and patient care.
Area of Science:
- Oncology
- Medical Imaging
- Artificial Intelligence
Background:
- Colorectal cancer (CRC) diagnosis relies on accurate assessment of microsatellite instability (MSI) status.
- Current methods for MSI status determination can be time-consuming and may lack precision.
- Developing automated, accurate tools for MSI status identification is crucial for personalized cancer treatment.
Purpose of the Study:
- To develop and validate a novel framework for identifying MSI status in colorectal cancer.
- To leverage advanced radiomics and deep learning techniques for enhanced diagnostic accuracy.
- To improve clinical decision-making and patient outcomes in oncology.
Main Methods:
- Utilized histopathological slide images from the NCT-CRC-HE-100K and PAIP 2020 databases.
- Implemented self-attentive adversarial stain normalization for data standardization.
- Employed a Slimmable Transformer for tumor delineation and a hybrid quantum-classical neural network for radiomics feature extraction.
Main Results:
- The developed system achieved 99% accuracy in identifying colorectal cancer MSI status.
- The model demonstrated high capability in differentiating between MSI and MSS tumors.
- The findings suggest potential for real-world clinical application in cancer care.
Conclusions:
- The novel framework significantly improves upon existing methods for colorectal cancer MSI status determination.
- Optimized processing and analysis of tissue features enhance the system's utility.
- This technology holds promise for improving patient care decisions in oncology.
Keywords:
Axial-attentionMicrosatellite instabilityQuantum–classical convolutional pre-trained neural networkSlimmable transformer
