{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/attention-based-multi-instance-neural-network","title":"Attention-based Multi-instance Neural Network for Medical Diagnosis from Incomplete and Low Quality Data","arxiv_id":"1904.04460","date":"2019-04-09","proceeding":null,"authors":["Zeyuan Wang","Josiah Poon","Shiding Sun","Simon Poon"],"abstract":"One way to extract patterns from clinical records is to consider each patient\nrecord as a bag with various number of instances in the form of symptoms.\nMedical diagnosis is to discover informative ones first and then map them to\none or more diseases. In many cases, patients are represented as vectors in\nsome feature space and a classifier is applied after to generate diagnosis\nresults. However, in many real-world cases, data is often of low-quality due to\na variety of reasons, such as data consistency, integrity, completeness,\naccuracy, etc. In this paper, we propose a novel approach, attention based\nmulti-instance neural network (AMI-Net), to make the single disease\nclassification only based on the existing and valid information in the\nreal-world outpatient records. In the context of a patient, it takes a bag of\ninstances as input and output the bag label directly in end-to-end way.\nEmbedding layer is adopted at the beginning, mapping instances into an\nembedding space which represents the individual patient condition. The\ncorrelations among instances and their importance for the final classification\nare captured by multi-head attention transformer, instance-level multi-instance\npooling and bag-level multi-instance pooling. The proposed approach was test on\ntwo non-standardized and highly imbalanced datasets, one in the Traditional\nChinese Medicine (TCM) domain and the other in the Western Medicine (WM)\ndomain. Our preliminary results show that the proposed approach outperforms all\nbaselines results by a significant margin.","url_abs":"http://arxiv.org/abs/1904.04460v1","url_pdf":"http://arxiv.org/pdf/1904.04460v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"attention-based-multi-instance-neural-network","repo_url":"https://github.com/Zeyuan-Wang/AMI-Net","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"medical-diagnosis","task_name":"Medical Diagnosis"}],"methods":[{"method_slug":"attention","method_name":"Attention"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}