Multiple instance learning method based on convolutional neural network and self-attention for early cancer detection
Authors:
- 7 December 2024
- journal article
- research article
- Published by Taylor & Francis in Computer Methods in Biomechanics and Biomedical Engineering
- p. 1-16
Abstract
Early cancer detection using T-cell receptor sequencing (TCR-seq) and multiple instances learning methods has shown significant effectiveness. We introduce a multiple instance learning method based on convolutional neural networks and self-attention (MICA). First, MICA preprocesses TCR-seq using word vectors and then extracts features using convolutional neural networks. Second, MICA uses an enhanced self-attention mechanism to extract relational features of instances. Finally, MICA can extract the crucial TCR-seq. After cross-validation, MICA achieves an area under the curve (AUC) of 0.911 and 0.946 on the lung and thyroid cancer datasets, which are 7.1% and 2.1% higher than other methods, respectively.Keywords
Funding Information
- Natural Science Foundation of Shandong Province (ZR2020MF134)
- Yantai School Land Integration Development Project (2021PT02)
This publication has 48 references indexed in Scilit:
- Implementation of machine learning techniques with big data and IoT to create effective prediction models for health informaticsBiomedical Signal Processing and Control, 2024
- Healthcare As a Service (HAAS): CNN-based cloud computing model for ubiquitous access to lung cancer diagnosisHeliyon, 2023
- From patterns to patients: Advances in clinical machine learning for cancer diagnosis, prognosis, and treatmentCell, 2023
- Breast cancer detection using deep learning: Datasets, methods, and challenges aheadComputers in Biology and Medicine, 2022
- LungNet: A hybrid deep-CNN model for lung cancer diagnosis using CT and wearable sensor-based medical IoT dataComputers in Biology and Medicine, 2021
- A comparative study of multiple instance learning methods for cancer detection using T-cell receptor sequencesComputational and Structural Biotechnology Journal, 2021
- TCR repertoire profiling of tumors, adjacent normal tissues, and peripheral blood predicts survival in nasopharyngeal carcinomaCancer Immunology, Immunotherapy, 2018
- T Cell Dysfunction in CancerCancer Cell, 2018
- T-cell receptor-engineered T cells for cancer treatment: current status and future directionsProtein & Cell, 2017
- Hallmarks of Cancer: The Next GenerationCell, 2011