Accuracy of a mobile sensor-based system for the detection of chemotherapy toxicity in older adults with cancer.
Bibliographic record
Abstract
1515 Background: Older adults receiving chemotherapy (CT) are at a high risk of toxicity (Tox) and of functional decline. Promptly acting upon CT Tox in older adults is difficult, particularly in developing countries where triage systems and personnel are limited. We previously showed that monitoring older patients using an accelerometer-equipped smartphone is feasible, and that a decline in the number of steps/day may represent a marker of Tox. We aimed to evaluate the diagnostic accuracy of an objective patient-centered measure of physical function (steps/day) for the remote detection of Tox in older adults receiving CT. Methods: We included consecutive patients aged ≥65 years starting first line CT for solid tumors at a single center in Mexico City. Patients underwent a geriatric assessment and were provided with an accelerometer-equipped smartphone. Daily steps were recorded for ≥7 days pre-CT initiation, and median number of pre-CT steps/day was calculated. Patients with ≤600 median pre-CT steps/day, and those using walking aids, were excluded (16% of recruited patients). Steps were monitored daily, compared with median pre-CT steps/day, and % decline was calculated. Patients were called daily, and Tox was assessed using PRO-CTCAE questionnaires. The % decline in steps/day from pre-CT was considered as the index test for Tox, while patient report was considered the reference standard. The association between % step decline and moderate/severe Tox was examined using generalized linear-mixed models. AUC was calculated and Youden’s index used to choose cutoff points for Tox detection. Results: 116 patients were included (96 development cohort, 20 validation cohort). Median age was 73y (range 65-91), 55% were female, and 65% had ≤high school education. The most common tumors were colon (21.5%), pancreas (17.5%), and gastric (12%). 28% of participants had never used a smartphone. The median number of pre-CT steps/day was 2979. Patients were followed for 6764 days, with Tox detected on 64.4% of days. Moderate/severe self-reported toxicity was detected on 1245 days (22.1%), while mild/no toxicity was detected on 4378 days (77.9%). AUC analysis for the development cohort demonstrated that a 32% decrease in steps/day from pre-CT median showed a sensitivity of 77.6% and a specificity of 67.3% for detecting moderate/severe Tox. Sensitivity and specificity of the cutoff in the validation cohort were 75.8% and 69.6%, respectively. Tox associated with a decline in steps/day included fatigue, pain, and nausea. Conclusions: A decline in the number of steps/day measured using an accelerometer-equipped smartphone was useful for the remote detection of moderate/severe Tox in older adults with cancer receiving CT, with a high sensitivity and specificity. This patient-centric measure could be used in clinical practice and research to detect and act promptly on Tox and, potentially, improve outcomes.
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 | 0.008 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 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 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".