ABSTRACT
In recent years, day surgery has established a strong reputation and popularity. It has a number of advantages over traditional surgery, including a shorter hospital stay, a lower risk of hospital-associated infections, and a higher cost efficiency. However, the patient admission criteria for day surgery were primarily dependent on manual guidelines defined by specialists, which lacked data driving from the real world and potentially wasted a lot of medical resources. In this paper, we proposed a day surgery patient admission algorithm, which is built by semi-supervised learning that combines both structured data and unstructured diagnoses to help surgeons make speedy admission decisions. We test this algorithm with the clinical data of day surgery patients who underwent laparoscopic cholecystectomy at West China Hospital and achieve an accuracy of 0.85 and an f1-score of 0.83, as well as reaching 0.97 on the precision score. The result is potentially broadly applicable to more day surgery types.
- Jiang L, Houston R, Li C, Siddiqi J, Ma Q, Wei S, Day surgery program at west China Hospital: exploring the initial experience. 2020;12(7). https://doi.org/10.7759/cureus.8961.Google Scholar
- Raita Y, Goto T, Faridi MK, Brown DF, Camargo CA, Hasegawa KJCc. Emergency department triage prediction of clinical outcomes using machine learning models. 2019;23(1):1-13. https://doi.org/10.1186/s13054-019-2351-7.Google Scholar
- Fenn A, Davis C, Buckland DM, Kapadia N, Nichols M, Gao M, Development and Validation of Machine Learning Models to Predict Admission From Emergency Department to Inpatient and Intensive Care Units. 2021. https://doi.org/10.1016/j.annemergmed.2021.02.029.Google ScholarCross Ref
- Luo L, Luo Y, Qin CJJoms. Admission control policies for surgery patients. 2017;41(8):1-16. https://doi.org/10.1007/s10916-017-0764-x.Google ScholarDigital Library
- Wang KY, Suresh KV, Puvanesarajah V, Raad M, Margalit A, Jain AJS. Using Predictive Modeling and Machine Learning to Identify Patients Appropriate for Outpatient Anterior Cervical Fusion and Discectomy. 2021;46(10):665-70. https://doi.org/10.1097/BRS.0000000000003865.Google Scholar
- Schwenker F, Trentin EJPRL. Pattern classification and clustering: A review of partially supervised learning approaches. 2014;37:4-14. https://doi.org/10.1016/j.patrec.2013.10.017.Google ScholarDigital Library
- Stekhoven DJ, Bühlmann PJB. MissForest—non-parametric missing value imputation for mixed-type data. 2012;28(1):112-8. https://doi.org/10.1093/bioinformatics/btr597.Google Scholar
- Goh KH, Wang L, Yeow AYK, Poh H, Li K, Yeow JJL, Artificial intelligence in sepsis early prediction and diagnosis using unstructured data in healthcare. 2021;12(1):1-10. https://doi.org/10.1038/s41467-021-20910-4.Google Scholar
- Jones KSJJod. A statistical interpretation of term specificity and its application in retrieval. 1972. https://doi.org/10.1108/eb026526.Google Scholar
- Mikolov T, Chen K, Corrado G, Dean JJapa. Efficient estimation of word representations in vector space. 2013.Google Scholar
- Devlin J, Chang M-W, Lee K, Toutanova KJapa. Bert: Pre-training of deep bidirectional transformers for language understanding. 2018.Google Scholar
- Reimers N, Gurevych IJapa. Sentence-bert: Sentence embeddings using siamese bert-networks. 2019.Google Scholar
- Zhang N, Jia Q, Yin K, Dong L, Gao F, Hua NJapa. Conceptualized representation learning for chinese biomedical text mining. 2020.Google Scholar
- Chen T, Guestrin C, editors. Xgboost: A scalable tree boosting system. Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining; 2016.Google ScholarDigital Library
- Desai RJ, Wang SV, Vaduganathan M, Evers T, Schneeweiss SJJno. Comparison of machine learning methods with traditional models for use of administrative claims with electronic medical records to predict heart failure outcomes. 2020;3(1):e1918962-e. https://doi.org/10.1001/jamanetworkopen.2019.18962.Google Scholar
- Xu Y, Yang X, Huang H, Peng C, Ge Y, Wu H, Extreme gradient boosting model has a better performance in predicting the risk of 90-day readmissions in patients with ischaemic stroke. 2019;28(12):104441. https://doi.org/10.1016/j.jstrokecerebrovasdis.2019.104441.Google Scholar
- Ke G, Meng Q, Finley T, Wang T, Chen W, Ma W, Lightgbm: A highly efficient gradient boosting decision tree. 2017;30:3146-54.Google Scholar
- Koleck TA, Dreisbach C, Bourne PE, Bakken SJJotAMIA. Natural language processing of symptoms documented in free-text narratives of electronic health records: a systematic review. 2019;26(4):364-79. https://doi.org/10.1093/jamia/ocy173.Google Scholar
- Abd-Alrazaq A, Alhuwail D, Househ M, Hamdi M, Shah ZJJomIr. Top concerns of tweeters during the COVID-19 pandemic: infoveillance study. 2020;22(4):e19016. https://doi.org/10.2196/19016.Google Scholar
- Islam MS, Sarkar T, Khan SH, Kamal A-HM, Hasan SM, Kabir A, COVID-19–related infodemic and its impact on public health: A global social media analysis. 2020;103(4):1621. https://doi.org/10.4269/ajtmh.20-0812.Google Scholar
Index Terms
- An Intelligent Patient Admission Model of Day Surgery Using Heterogeneous Data with Semi-Supervised Learning
Recommendations
Admission Control Policies for Surgery Patients
In China, day surgery has been promoted because its operation time and post-operative hospital stay are shorter than those of elective surgery. Day surgery can speed up the turnover of beds and operation rooms. Usually, the conditions of elective ...
Transductive Multilabel Learning via Label Set Propagation
The problem of multilabel classification has attracted great interest in the last decade, where each instance can be assigned with a set of multiple class labels simultaneously. It has a wide variety of real-world applications, e.g., automatic image ...
Inductive Semi-supervised Multi-Label Learning with Co-Training
KDD '17: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data MiningIn multi-label learning, each training example is associated with multiple class labels and the task is to learn a mapping from the feature space to the power set of label space. It is generally demanding and time-consuming to obtain labels for training ...
Comments