Predicting low oxygen in COVID-19 patients isolating at home
Bibliographic record
Abstract
<bold>Introduction:</bold> During the COVID-19 pandemic, identifying which patients could safely isolate at home and which required hospitalization was a concern. We developed COVIDFree@Home, a mobile app and clinician dashboard, for remote monitoring of patients at home. This study aimed to determine if remotely collected measures could predict low oxygen saturation (SpO2). <bold>Methods:</bold> Patients diagnosed with COVID-19 were recruited from three hospitals in Toronto, Canada, 2020 - 2022. Twice a day they entered symptoms, temperature, heart rate, and oxygen saturation, which were monitored by clinicians. Baseline characteristics and remote monitoring variables were analysed to determine predictors of SpO2 ≤ 92% in the following two days using a random forest classifier. <bold>Results:</bold> Of 431 participants, 376 (87.2%) entered at least one measure. Forty-nine (13%) had low SpO2, and 19 (5.1%) were hospitalized. Older age, alpha/beta variant, obesity, and preexisting pulmonary disease as well as time varying features of dyspnea, severe fatigue, and temperature ≥ 38.0°C, were associated with SpO2 ≤ 92% in the next two days. Our model predicted low SpO2 with a sensitivity of 53%, specificity 70%, and AUC of 0.70. <bold>Conclusions:</bold> Remote monitoring, along with baseline characteristics, can predict low SpO2 in people with COVID-19 isolating at home with 70% specificity, helping identify those needing medical attention. However, the 53% sensitivity suggests that this model may have missed a proportion of patients who subsequently became hypoxemic. Future studies should explore strategies to improve sensitivity and assess its effectiveness for other viruses in vulnerable populations.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 teacher head, 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".