Figure - PMC

Skip to main content

An official website of the United States government

Here's how you know

Here's how you know

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

View full-text article in PMC

. 2023 May 17;11:306–317. doi: 10.1109/JTEHM.2023.3276943

Search in PMC
Search in PubMed
View in NLM Catalog
Add to search

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/

PMC Copyright notice

Fig. 4. — Split the data into inputs and target, and inputs’ embedding. The patient represented made eight visits, and the diff_dgn set is . We have records extending until 1,140 days after this patient’s sleep-apnea diagnosis. The goal is to predict the visit cost over the last year ( ) on the basis of the previous visits’ records (from the date of G47.3 diagnosis to day 776). Therefore, the time indices for the inputs and target are the maximum between 0 and 776 and the minimum between 776 and the last visit point, day 1,140. In line with the definition and , the inputs for this patient are the records from day 0 to day 690 from diagnosis, and the targets are the visit costs from day 810 to 1,140. The autoregressive prediction mechanism of Transformer is shown at right.

Inline graphic — Split the data into inputs and target, and inputs’ embedding. The patient represented made eight visits, and the diff_dgn set is . We have records extending until 1,140 days after this patient’s sleep-apnea diagnosis. The goal is to predict the visit cost over the last year ( ) on the basis of the previous visits’ records (from the date of G47.3 diagnosis to day 776). Therefore, the time indices for the inputs and target are the maximum between 0 and 776 and the minimum between 776 and the last visit point, day 1,140. In line with the definition and , the inputs for this patient are the records from day 0 to day 690 from diagnosis, and the targets are the visit costs from day 810 to 1,140. The autoregressive prediction mechanism of Transformer is shown at right.