AnchorDx Achieves Academic Breakthrough with Research Paper on Learning to Represent Methylation in Depth Accepted by 36th AAIA Conference
GUANGZHOU, China, December 9, 2021 / PRNewswire / – AnchorDx, a leading global developer of cancer screening and early detection solutions, was accepted on December 1, 2021, to present his research article “Noninvasive Lung Cancer Early Detection via Deep Methylation Representation Learning” (DeepMeth) at 36e AAAI Conference on Artificial Intelligence (AAAI-22) next year at Vancouver. Jointly submitted by AnchorDx and the Advanced Medical Data Research Center at Nankai University, the article was among a record 9,251 submissions. the AAAI’22 received 9,020 reviewed articles and only 1,349 accepted articles, resulting in an overall acceptance rate of just 15%, the lowest in AAAI history.
DeepMeth is one of the first studies in the world to represent methylation regions by continuous vectors based on deep learning. This is Also the core algorithm of AnchorDx’s early lung cancer diagnosis product, PulmoSeek™ More. Like an integrated diagnostic model combining methylation testing with artificial intelligence (AI) LDCT imaging, PulmoSeek ™ Plus could help facilitate the accurate diagnosis of early stage lung cancer in patients with lung nodules and guide clinical decisions. DeepMeth stands out from nearly ten thousand Research papers attach a high appreciation and validation by the global artificial intelligence community for the company’s achievements in the early detection of cancer. It’s also a breakthrough in the optimism that the convergence of AI and precision medicine in healthcare promises.
Moreover, back at In April of this year, AnchorDx’s research paper on its methylation model for the early diagnosis of blood-based lung cancer, PulmoSeek™, had already been published in The Journal of Clinical Investigation. These two high distinctions once again demonstrate the company’s global leadership position in the early detection of lung cancer.
Breaking the bottleneck of early detection of cancer by methylation of ctDNA
Methylation of circulating tumor DNA (ctDNA) in the blood is non-invasive, cost-effective, and offers organizational traceability, making it valuable for the early detection of cancers. Thereby, cDNA methylation analyzes is one of the most promising methods for the early detection of lung cancer. However, the abundance of traces of ctDNA in plasma has always been a technical challenge in the early detection of cancer. The Department of Computer Science at Nankai University and AnchorDx have worked collaboratively for years to overcome this challenge to develop a new deep learning model – DeepMeth – which represents each methylation sample of ctDNA by concatenating vectors. of the region.
Unlike previous studies, DeepMeth for the first time focused on information about the co-methylation pattern in the region instead of methylated sites in a region.. Considering each methylation region independently and developing an auto-encoder to learn CpG rich semantics, for example, the positions of methylated CpG sites, hidden in the region layouts, DeepMeth can largely avoid the curse of dimensionality.
DeepMeth has been validated by extensive experiments on a clinical data set collected from 14 hospitals. Experiences show that DeepMeth outperforms four leading baselines in terms of AUC. Additionally, DeepMeth has also been shown to be combined with traditional metrics to improve prediction performance. It is encouraging to note that since april 2020, PulmoSeek based on DeepMeth™ Plus has been clinically applied to patients at 94 hospitals for the early detection of lung cancer and has shown satisfactory performance.
Founded in 2015, AnchorDx is one of the world’s leading developers of cancer screening and early detection solutions based on NGS methylation technology. AnchorDx strives to revolutionize technologies and products for the early detection, detection and early detection of single / multi / pan-cancer cancers. The current product pipeline encompasses over 70% of all cancers. In addition, AnchorDx has a proprietary database and artificial intelligence platform designed for the early detection, diagnosis and treatment of lung cancer.
About AAAI and the AAAI Conference
The Association for the Advancement of Artificial Intelligence (AAAI) is an international scientific society dedicated to promoting the research and responsible use of artificial intelligence. The organization was founded in 1979 by the winners of the Turing Prize Marvin minsky and John mccarthy. AAAI’s annual conference is the world’s largest academic conference in the field of artificial intelligence, and it is also a Category A conference designated by the China Computer Federation (CCF). AAIA is the leading scientific society for artificial intelligence (AI), with over 200 elected members and over 6,000 members. AAIA members are known as the “Hall of Fame” and include the latest two Turing Prize winners, Deep Learning Pioneers, Yoshua Bengio and Yann LeCun.