LC/NC Pipeline for Training and Operationalising Segmentation Models in a Data Scarce Domain: De-arraying Tissue MicroArrays
Bibliographic record
Abstract
Abstract Here we present a new approach to training and operationalizing segmentation models for de-arraying Tissue Micro Arrays (TMAs). The scarcity of large, high-quality datasets in sensitive domains such as human tissue samples, coupled with strict privacy regulations to protect doner interests, poses significant obstacles to training robust and generalised segmentation models. To address these challenges, we introduce a new Low-Code/No-Code (LCNC) Domain-Specific Language (DSL) integrated into the Cinco de Bio (CdB) platform. The DSL consists of multiple Service-Independent Building Blocks (SIBs), each providing a distinct functionality essential to creating a pipeline. LCNC enables biologists to train and deploy de-arraying models without writing code. Our methodology incorporates a domain-specific data augmentation technique that generates pseudo-synthetic samples from a minimal set of real data. It also leverages AutoML techniques, including Neural Architecture Search (NAS) and hyperparameter optimisation, to automate the model development process. Furthermore, we present an architectural update to the Cinco de Bio platform, adopting a “Model as Data” paradigm that treats neural network models as dynamic, versioned data assets that can be used as inputs to SIBs. This work provides a practical solution to the challenges of distribution shift and data scarcity in sensitive health domains, where building sufficiently sized datasets to train generalise robust models is infeasible. The proposed LCNC DSL and accompanying pipeline enables domain experts to effectively leverage Artificial Intelligence (AI) technologies and tailor them to their own data.
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 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.008 | 0.004 |
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".