Enhancing ICU Management and Addressing Challenges in Türkiye Through AI-Powered Patient Classification and Increased Usability With ICU Placement Software
Bibliographic record
Abstract
The increasing demand for intensive care unit (ICU) admissions, and the associated rising costs have urged the need for effective management strategies. In this research, we focus on the challenges faced by (1) hospitals in report generation and in their effort to properly allocate ICU patients, and (2) insurance organizations responsible for payments. We address the issues of misclassification and financial burden on hospitals and insurance organizations that arise from inefficient and subjective application of regulations while also considering the impact on medical personnel. Through existing literature analysis, as well as extensive discussions with critical care professionals and insights gained from university hospitals, we identified the need for a supportive machine learning model for ICU level classification of patients, and furthermore, we propose an easily deployable and highly interoperability software system specifically for placement of patients in various ICU levels. We aim to support healthcare professionals in their decision-making process with the supportive machine learning model and the software system that we named "heartbeat". This research aims to bridge the gap between hospitals and insurance institutions to ensure fair and objective patient classification and to improve the overall ICU management. The process has been tested using MIMIC-III version 1.4 dataset as a proof of concept to demonstrate the applicability and effectiveness of the developed system. Further testing using real data after official deployment and usage by various stakeholders.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.007 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".