Evidence practice · research lineage
One project, three records
How to describe a preprint, a conference abstract, and open-source software without turning three public manifestations into three scientific claims.
Status: Primary-source evidence guide · Current public-record snapshot · TotalSpineSeg lineage · No new canonical work · No new empirical result · No journal peer-review claim · No clinical-deployment claim · No citation-count claim
The record type controls the sentence
A search for TotalSpineSeg can surface a public preprint, an ISMRM conference record, an open-source repository, package releases, and third-party references. These records are related, but they answer different questions. A preprint reports a study, a conference identifier records a conference contribution, and a source repository exposes an implementation. None silently inherits the status of another.
The evidence-safe approach is to identify the research lineage once, attach each claim to the record that supports it, and preserve the status of every manifestation.
Three records, three bounded claims
| Public record | What it supports | What it does not establish |
|---|---|---|
| ResearchGate preprint 10.13140/RG.2.2.31318.56649 | The March 2025 record identifies TotalSpineSeg: Robust Spine Segmentation with Landmark-Based Labeling in MRI, lists Emanuel Shirbint among the authors, and exposes the study manuscript. The page classifies it as a preprint or early-stage research record whose peer-review status is not established there. | A peer-reviewed journal publication, present employment, clinical approval, or deployment in patient care. |
| ISMRM conference record 10.58530/2025/0624 | The persistent identifier records the 2025 conference contribution TotalSpineSeg: Robust Spine Segmentation and Labeling Across Multiple MRI Contrasts. It supports a conference-work and authorship statement tied to that record. | That the related preprint became a journal article, or that conference inclusion independently validates every claim in the manuscript. |
| Open-source implementation neuropoly/totalspineseg | The project repository supports the existence of a public implementation and documents its software workflow, citation target, and release lineage. | Clinical effectiveness, regulatory clearance, institutional adoption, a production service operated by Emanuel Shirbint, or a new scholarly work. |
Why this remains one catalogue lineage
Titles can differ across a manuscript, conference abstract, and software documentation while still referring to the same underlying project. Deduplication therefore uses more than title similarity: author overlap, topic, timing, persistent identifiers, citation instructions, and explicit cross-links are considered together. In the canonical catalogue, the TotalSpineSeg preprint, ISMRM contribution, and implementation remain connected manifestations rather than three independently counted research works.
This does not erase the conference record or reduce the software to a citation. It preserves both while preventing double counting. The conference work remains conference work; the preprint remains a preprint; the repository remains software evidence.
A claim ladder for mixed research outputs
- Existence: the linked record exists and exposes the stated title or implementation.
- Authorship or contribution: the specific record lists Emanuel Shirbint; attribution stays work-specific.
- Reported result: numerical or empirical statements are attributed to the preprint, not restated as independently verified performance.
- Dissemination: the conference DOI supports conference presentation status only.
- Availability: the repository supports public software availability, not clinical readiness.
- Higher-status claims: journal peer review, production validation, clinical use, regulatory status, present employment, or advisory authority require separate direct evidence.
Current registry interpretation
On 8 September 2026, a targeted Crossref created-date check returned no new exact-author-family record since 1 September. Europe PMC returned 14 author-query records and PubMed returned four exact-author records. The current Preprints.org pages for the two Version 2 works in the catalogue still identify those versions as latest and not peer-reviewed. No new canonical work or version was promoted.
Crossref and Europe PMC citation metadata was inspected as time-sensitive registry metadata only. Citation counts remain excluded from public claims because coverage, version deduplication, and counting rules have not been normalized into the canonical claim ledger.
Claim boundaries across the public profile
Peer-reviewed journal articles, conference work, preprints, conceptual research, the bounded empirical preprint, software records, the locally evidenced EWCC production system, employment, and advisory work remain separate evidence classes. This guide reports no new scientific finding and makes no claim of journal acceptance, clinical deployment, regulatory approval, present employment, advisory engagement, institutional representation, or citation impact.
Primary records
- ResearchGate DOI record — TotalSpineSeg public preprint
- ISMRM 2025 DOI record — TotalSpineSeg conference work
- NeuroPoly GitHub repository — TotalSpineSeg implementation
- PyPI project — public software distribution history
- Europe PMC current author-query record set
Evidence checked 8 September 2026 (Asia/Jerusalem). Search results were used for discovery only; claim support was checked against the linked publication, conference, registry, and software records.