AITRICS's Research Paper on Patient Deterioration Prediction Model to be Accepted
[July 18, 2023] AITRICS (CEO Kwang joon Kim, Jin-gyu Yoo), a company specializing in artificial intelligence (AI) technology announced on 18th that AITRICS's research paper on patient deterioration prediction model was accpeted as an outstanding paper at MLHC 2023.
The research paper aims to address the challenges associated with the use of multi-modality in EHR. Existing clinical sites using EHR provide abundant information through various modalities, but as the modalities used increased, the amount of calculations also increased and the data input cycle was irregular.
In this study, the ▲Unified Multi-Modal Set Embedded (UMSE) module and ▲Flexible Multi-Modal Attachment with Skip Botleneck developed by AITRICS solved irregular data entry problems using only original data, and improved performance in patients with some missing data.
Accordingly, the model developed by AITRICS, confirmed superior performance compared to other existing models in predicting mortality, use of Vasopressor, and intubation that may occur in patients within 12 hours.
The paper was selected for the Oral Presentation, which is given only to a few excellent studies among the papers submitted to MLHC 2023, and will be presented at a conference held at Columbia University in New York from August 11 to 12.
Kwan-hyung Lee, a researcher at AITRICS, said, "Through this study, we were able to confirm that AI also improves the accuracy of predicting patient conditions by comprehensively utilizing biosignals, X-Ray images, and clinical note data in a similar way to actual medical staff." In particular, more than three EHR multimodal deep fusion are the first cases, adding, "I think the oral presentation is an opportunity to show AI technology and research results worldwide. In the future, AITRICS will continue to make efforts to help medical staff effectively select high-risk patients in the clinical field and make decisions quickly"
Meanwhile, MLHC is the largest academic conference in AI technology using medical big data.The conference held every year since 2011, active exchanges and discussion of artificial intelligence, machine learning, machine learning and medical experts.