Bibliographic record
Abstract
Background: One of the hurdles in osteoarthritis (OA) drug discovery and the improvement of therapeutic approaches is the early identification of patients who will progress. It is therefore crucial to find efficient and reliable means of screening OA progressors. Although the main risk factors, age, gender and body mass index (BMI), are important, they alone are poor predictors. However, serum factors could be potential biomarkers for early prediction of knee OA progression. Objectives: In a first step toward finding early reliable predictors of OA progressors, this study aimed to determine, in OA individuals, the optimum combination of serum levels of adipokines/related inflammatory factors, their ratios, and the three main OA risk factors for predicting knee OA infrapatellar fat pad (IPFP) volume, as this tissue has been associated with knee OA onset and progression. Methods: Serum and magnetic resonance images (MRI) were from the Osteoarthritis Initiative at baseline. Variables (48) comprised the 3 main OA risk factors (age, gender, BMI), 6 adipokines, 3 inflammatory factors, and their 36 ratios. IPFP volume was assessed on MRI with a neural network methodology. The best variables and models were identified in Total cohort (n=678), High-BMI (n=341) and Low-BMI (n=337), using an artificial intelligence selection approach: the adaptive neuro-fuzzy inference system embedded with fuzzy c-means clustering (ANFIS-FCM). Performance was validated using uncertainty analyses and statistical indices. Reproducibility was done using 80 OA patients from a clinical trial (female, n=57; male, n=23). Results: For the three groups, 8.44E+14 sub-variables were investigated and 48 models were selected. The best model for each group included five variables: the three risk factors and adipsin/C-reactive protein combined for Total cohort, adipsin/chemerin; High-BMI, chemerin/adiponectin high molecular weight; and Low-BMI, interleukin-8. Data also revealed that the main form of the ratio used for the model was justified, as the use of the inverse form slightly decreased the performance of the model in both training and testing stages. Further investigation indicated that gender improved (13-16%) the prediction results compared to the BMI-based models. For each gender, we then generated a pseudocode (an evolutionary computation equation) with the 5 variables for predicting IPFP volume. Reproducibility experiments were excellent (correlation coefficient: female 0.83, male 0.95). Conclusion: This study demonstrates, for the first time, that the combination of the serum levels of adipokines/inflammatory factors and the three main risk factors of OA could predict IPFP volume with high reproducibility, and superior performance with gender separation. By using the models for each gender and the pseudocodes for OA patients provided in this study, the next step will be to develop a predictive model for OA progressors. Acknowledgments: This work was funded by the Chair in Osteoarthritis of the University of Montreal, the Osteoarthritis Research Unit of the University of Montreal Hospital Research Centre, the Groupe de recherches des maladies rhumatismales du Québec and by ArthroLab Inc., all from Montreal, Quebec, Canada. Disclosure of Interests: Hossein Bonakdari: None declared, Ginette Tardif: None declared, François Abram Employee of: ArthroLab Inc., Jean-Pierre Pelletier Shareholder of: ArthroLab Inc., Grant/research support from: TRB Chemedica, Speakers bureau: TRB Chemedica and Mylan, Johanne Martel-Pelletier Shareholder of: ArthroLab Inc., Grant/research support from: TRB Chemedica
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 | 0.000 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 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".