M-CNN-RF: A hybrid deep learning model for accurate pediatric skeletal age estimation using hand bone radiographs
Bibliographic record
Abstract
Precise and accurate skeletal age estimation using medical imaging is a pivotal and challenging task in the healthcare sector, particularly for identifying potential bone growth issues in infants and newborns. Therefore, this study addresses the pervasive challenges associated with assessing bone abnormalities in pediatric patients, including injuries and infections. Given the importance of early and precise detection of skeletal development, a novel hybrid model is proposed that integrates a modified convolutional neural network (M-CNN) with a robust machine learning (ML) model, specifically random forest (RF), resulting in the M-CNN-RF framework. This model is designed to enhance pediatric bone health assessment by providing an effective method for skeletal age estimation. The M-CNN-RF model is tailored to accurately evaluate hand bone maturation, overcoming the inherent difficulties in skeletal age assessment. The model utilizes the bone age dataset from the Radiological Society of North America that includes 14,236 left-hand radiological images, focusing on the development of a robust model for a precise evaluation based on hand skeleton guidelines. In addition, to enhance the prediction and generalization of the model, data augmentation techniques were employed to increase the size of the dataset. The M-CNN-RF exhibits exceptional performance using numerous performance measures, achieving an accuracy of 97% and precision and recall exceeding 94%. In addition, the model reaches an F1 score of 97%, highlighting the ability of the model to ensure a balance between precision and recall. Furthermore, low mean absolute error (MAE) and mean square error (MSE) values of 0.0141 and 0.0327, respectively, were computed for the proposed model, which demonstrates its notable efficacy in predicting skeletal age. The findings of this study not only contribute valuable information for clinical applications but also underscore the potential of the adopted approach to address the challenges associated with pediatric bone health assessment.
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 | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
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".