Harnessing Big Data to Optimize an Algorithm for Rapid Diagnosis of Pulmonary Tuberculosis in a Real-World Setting
Bibliographic record
Abstract
Background The prompt diagnosis of pulmonary tuberculosis (PTB) remains a challenge in clinical practice. The present study aimed to optimize an algorithm for rapid diagnosis of PTB in a real-world setting. Methods 28,171 adult inpatients suspected of having PTB in China were retrospectively analyzed. Bronchoalveolar lavage fluid (BALF) and/or sputum were used for acid-fast bacilli (AFB) smear, Xpert MTB/RIF (Xpert), and culture. A positive mycobacterial culture was used as the reference standard. Peripheral blood mononuclear cells (PBMC) were used for T-SPOT.TB. We analyzed specimen types’ effect on these assays’ performance, determined the number of smears for diagnosing PTB, and evaluated the ability of these assays performed alone, or in combination, to diagnose PTB and nontuberculous mycobacteria (NTM) infections. Results Sputum and BALF showed moderate to substantial consistency when they were used for AFB smear or Xpert, with a higher positive detection rate by BALF. 3-4 smears had a higher sensitivity than 1-2 smears. Moreover, simultaneous combination of AFB and Xpert correctly identified 44/51 of AFB+/Xpert+ and 6/7 of AFB+/Xpert- cases as PTB and NTM, respectively. Lastly, when combined with AFB/Xpert sequentially, T-SPOT showed limited roles in patients that were either AFB+ or Xpert+. However, T-SPOTMDC (manufacturer-defined cut-off) showed a high negative predicative value (99.1%) and suboptimal sensitivity (74.4%), and TBAg/PHA (ratio of Mycobacterium tuberculosis-specific antigens to phytohaemagglutinin spot-forming cells, which is a modified method calculating T-SPOT.TB assay results) ≥0.3 demonstrated a high specificity (95.7%) and a relatively low sensitivity (16.3%) in AFB-/Xpert- patients. Conclusions Concurrently performing AFB smear (at least 3 smears) and Xpert on sputum and/or BALF could aid in rapid diagnosis of PTB and NTM infections in a real-world high-burden setting. If available, BALF is preferred for both AFB smear and Xpert. Expanding this algorithm, PBMC T-SPOTMDC and TBAg/PHA ratios have a supplementary role for PTB diagnosis in AFB-/Xpert- patients (moderately ruling out PTB and ruling in PTB, respectively). Our findings may also inform policy makers’ decisions regarding prevention and control of TB in a high burden setting.
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 | 0.005 | 0.012 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.004 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.003 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| 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".