SG-Transunet: A segmentation-guided Transformer U-Net model for KRAS gene mutation status identification in colorectal cancer
These videos have been matched automatically. Contact us if they are not relevant.
These videos have been matched automatically. Contact us if they are not relevant.
View abstract on PubMed
Summary
This summary is machine-generated.Identifying Kirsten rat sarcoma virus (KRAS) gene mutation status in colorectal cancer (CRC) is crucial for targeted therapy. Our novel SG-Transunet model accurately identifies KRAS mutations by integrating spatial and frequency domain features, improving treatment decisions.
Area Of Science
- Oncology
- Medical Imaging
- Bioinformatics
Background
- Accurate Kirsten rat sarcoma virus (KRAS) gene mutation identification is vital for colorectal cancer (CRC) treatment selection.
- Current deep learning methods often overlook frequency domain features and are susceptible to redundant data, impacting KRAS mutation status identification accuracy.
- Existing spatial feature extraction methods may not fully capture the complexities required for precise KRAS mutation status identification.
Purpose Of The Study
- To develop an advanced deep learning model for accurate KRAS gene mutation status identification in colorectal cancer (CRC).
- To enhance the model's ability to integrate both spatial and frequency domain features for improved diagnostic performance.
- To address limitations of existing methods, including redundant feature handling and neglect of frequency domain information.
Main Methods
- Proposed a segmentation-guided Transformer U-Net (SG-Transunet) model combining CNNs and Transformers.
- Integrated a frequency domain supplement block to capture essential frequency domain features alongside spatial features.
- Employed an encoder-decoder for lesion segmentation to guide mutation identification, a pre-trained Xception block to prevent overfitting, and an aggregate attention module to enhance feature discriminability.
- Utilized a mutual-constrained loss function to optimize both segmentation and identification tasks simultaneously.
Main Results
- The SG-Transunet model demonstrated superior performance in identifying KRAS gene mutation status compared to existing state-of-the-art methods.
- The integration of spatial and frequency domain features significantly improved the discriminative power for KRAS mutation status.
- Segmentation-guided lesion localization enhanced the accuracy of the identification task.
Conclusions
- The proposed SG-Transunet model offers a robust and accurate approach for KRAS gene mutation status identification in colorectal cancer.
- The model's ability to leverage both spatial and frequency domain information provides a significant advancement in the field.
- This method holds promise for improving personalized treatment strategies for CRC patients based on their KRAS mutation status.
These videos have been matched automatically. Contact us if they are not relevant.
These videos have been matched automatically. Contact us if they are not relevant.
Related Concept Videos
The Ras-gene-encoded proteins are regulators of signaling pathways controlling cell proliferation, differentiation, or cell survival. The Ras-gene family in humans constitutes three primary members—the HRas, NRas, and KRas. These genes code for four functionally distinct yet closely related proteins—the HRas, NRas, KRas4A, and KRas4B. The involvement of mutant Ras genes in human cancer was first discovered in 1982 and is among the most common causes of human tumorigenesis.
Ras is a...
Tumor progression is a phenomenon where the pre-formed tumor acquires successive mutations to become clinically more aggressive and malignant. In the 1950s, Foulds first described the stepwise progression of cancer cells through successive stages.
Colon cancer is one of the best-documented examples of tumor progression. Early mutation in the APC gene in colon cells causes a small growth on the colon wall called a polyp. With time, this polyp grows into a benign, pre-cancerous tumor. Further...