126 The Canadian Neonatal Brain Platform: A three-pillar approach
Bibliographic record
Abstract
Brain injury and abnormal maturation in the neonatal period is associated with long-term changes underlying significant cognitive, motor, language and behavioural deficiencies. Our understanding of clear cerebral disruptors of brain development and the extent of their impact are still limited, mostly due to the lack of robust non-invasive biomarkers, difficulties in conducting studies in newborns and the use of small isolated cohorts. To address all these issues, we have created the Canadian Neonatal Brain Platform (CNBP). The aims of CNBP are threefold: 1) To create and manage a national neonatal MRI registry to support prospective multicenter trials; 2) To make accessible to the community standardized sets of tools and protocols for image acquisition and analysis specifically designed for neonatology; 3) To better identify the concerns of parents of preterm infants by building a robust partnership. Aim 1: We modelled our architecture with a semi-decentralized overall structure composed of a site-specific infrastructure for initial localized data aggregation and anonymization, and a central server for post-processing and ensuring long-term scaleability. Aim 2: We conducted experiments to simulate human rater image quality assessment with machine learning. In addition, we performed brain segmentation using various open source neonatal neuroimaging analysis software to qualitatively validate their results against human rater golden standard. Aim 3: We deployed Mieux Agir au Quotidien (http://developpementenfant.ca), a web-based educational and support program for parents of preterm infants that incorporates state-of-the-art teaching modules on developmentally supportive care. Aim 1: Our infrastructure (see Figure 1) was successfully deployed in Compute Canada. We used DICOMTransit to collect, aggregate, anonymize and centralize data from MRI scanners, Canadian Neonatal Network and Canadian Neonatal Follow-up Network. We implemented LORIS (https://dev.cnbp.ca) to build a clinical neonatal imaging registry. Aim 2: Our pipeline reached a 75% sensitivity and 85% positive predictive value to control for quality. Our assessment of the segmentation tools revealed that MANTIS provides the most robust segmentation results. Aim 3: Mieux Agir au Quotidien reached over 700K visits in 2017. We have established a special partnership with parents of preterm infants, which has enabled us to gather first-rate information on parents’ concern and knowledge about the disorders of preterm infants, now available in english. CNBP has successfully progressed towards achieving its aims by establishing an online data processing and integration portal integrated with numerous neonatal specific analysis software while providing social and knowledge transfer to the general public.
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 | 0.007 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.004 | 0.004 |
| Science and technology studies | 0.005 | 0.003 |
| Scholarly communication | 0.007 | 0.004 |
| Open science | 0.004 | 0.009 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.034 | 0.014 |
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".