Noninvasive Lung Cancer Early Detection via Deep Methylation Representation Learning

被引:0
|
作者
Cai, Xiangrui [1 ]
Tao, Jinsheng [2 ]
Wang, Shichao [1 ]
Wang, Zhiyu [1 ]
Wang, Jiaxian [1 ]
Li, Mei [1 ]
Wang, Hong [2 ]
Tu, Xixiang [2 ]
Yang, Hao [2 ]
Fan, Jian-Bing [2 ]
Ji, Hua [1 ]
机构
[1] Nankai Univ, Adv Med Data Res Ctr, CS, Tianjin, Peoples R China
[2] AnchorDx Med Co, Guangzhou, Guangdong, Peoples R China
来源
THIRTY-SIXTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE / THIRTY-FOURTH CONFERENCE ON INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE / TWELVETH SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE | 2022年
关键词
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Early detection of lung cancer is crucial for five-year survival of patients. Compared with the pathological analysis and CT scans, the circulating tumor DNA (ctDNA) methylation based approach is noninvasive and cost-effective, and thus is one of the most promising methods for early detection of lung cancer. Existing studies on ctDNA methylation data measure the methylation level of each region with a predefined metric, ignoring the positions of methylated CpG sites and methylation patterns, thus are not able to capture the early cancer signals. In this paper, we propose a blood-based lung cancer detection method, and present the first ever study to represent methylation regions by continuous vectors. Specifically, we propose DeepMeth to regard each region as a one-channel image and develop an auto-encoder model to learn its representation. For each ctDNA methylation sample, DeepMeth achieves its representation via concatenating the region vectors. We evaluate DeepMeth on a multicenter clinical dataset collected from 14 hospitals. The experiments show that DeepMeth achieves about 5%-8% improvements compared with the baselines in terms of Area Under the Curve (AUC). Moreover, the experiments also demonstrate that DeepMeth can be combined with traditional scalar metrics to enhance the diagnostic power of ctDNA methylation classifiers. DeepMeth has been clinically deployed and applied to 450 patients from 94 hospitals nationally since April 2020.
引用
收藏
页码:11828 / 11836
页数:9
相关论文
共 50 条
  • [1] Early Lung Cancer Detection Using Deep Learning Optimization
    Elnakib, Ahmed
    Amer, Hanan M.
    Abou-Chadi, Fatma E. Z.
    INTERNATIONAL JOURNAL OF ONLINE AND BIOMEDICAL ENGINEERING, 2020, 16 (06) : 82 - 94
  • [2] Noninvasive Diagnostics for Early Detection of Lung Cancer: Challenges and Potential with a Focus on Changes in DNA Methylation
    Farooq, Maria
    Herman, James G.
    CANCER EPIDEMIOLOGY BIOMARKERS & PREVENTION, 2020, 29 (12) : 2416 - 2422
  • [3] Exploiting histopathological imaging for early detection of lung and colon cancer via ensemble deep learning model
    Alotaibi, Moneerah
    Alshardan, Amal
    Maashi, Mashael
    Asiri, Mashael M.
    Alotaibi, Sultan Refa
    Yafoz, Ayman
    Alsini, Raed
    Khadidos, Alaa O.
    SCIENTIFIC REPORTS, 2024, 14 (01):
  • [4] DFCV: a framework for evaluation deep learning in early detection and classification of lung cancer
    Alsadoon A.
    Al-Naymat G.
    Osman A.H.
    Alsinglawi B.
    Maabreh M.
    Islam M.R.
    Multimedia Tools and Applications, 2023, 82 (28) : 44387 - 44430
  • [5] Establishing a deep cfDNA methylation sequencing-based signature for noninvasive early-stage lung cancer diagnosis
    Zhang, Zhihong
    Liang, Naixin
    Deng, Wanglong
    Wang, Chenyang
    Zheng, Tao
    Li, Wenjun
    Hu, Jian
    Su, Jinh
    Han Han-Zhang
    Li, Bingsi
    Liu, Hao
    Li, Shanqing
    CLINICAL CANCER RESEARCH, 2018, 24 (17)
  • [6] Lung Cancer Detection: A Deep Learning Approach
    Bhatia, Siddharth
    Sinha, Yash
    Goel, Lavika
    SOFT COMPUTING FOR PROBLEM SOLVING, 2019, 817 : 699 - 705
  • [7] Deep learning for lung Cancer detection and classification
    A. Asuntha
    Andy Srinivasan
    Multimedia Tools and Applications, 2020, 79 : 7731 - 7762
  • [8] Deep learning for lung Cancer detection and classification
    Asuntha, A.
    Srinivasan, Andy
    MULTIMEDIA TOOLS AND APPLICATIONS, 2020, 79 (11-12) : 7731 - 7762
  • [9] A multiplex DNA methylation assay for noninvasive early detection of gastric cancer.
    Ruan, Weimei
    Lei, Xiaotian
    Sha, Weihong
    Li, Bin
    Wang, Hong
    Xia, Yuying
    Shen, Junran
    Kong, Lingchen
    Ma, Juan
    Chen, Zhiwei
    Bibikova, Marina
    Cui, Chunhui
    Fan, Jian-Bing
    JOURNAL OF CLINICAL ONCOLOGY, 2023, 41 : 300 - 300
  • [10] Lung Nodule Detection via Deep Reinforcement Learning
    Ali, Issa
    Hart, Gregory R.
    Gunabushanam, Gowthaman
    Liang, Ying
    Muhammad, Wazir
    Nartowt, Bradley
    Kane, Michael
    Ma, Xiaomei
    Deng, Jun
    FRONTIERS IN ONCOLOGY, 2018, 8