/model-selection

name: model-selection
description: Use when choosing the model for a medical-imaging study. Picks a paper-grounded architecture family (CNN/ViT, U-Net/nnU-Net, detection, SAM/foundation, GNN), then vets the concrete repo or checkpoint for licence, version pin, weight provenance and benchmark overlap.
metadata:
  triggers: "architecture zoo, model sourcing, which architecture, choose a model, model selection, ResNet vs ViT, U-Net vs nnU-Net, what backbone, foundation model for, transfer learning choice, MedSAM, TotalSegmentator, DINO, MAE, self-supervised, graph neural network, GNN, brain connectome, GCN, GAT, GraphSAGE, BrainGNN, population graph, paper to architecture, reference implementation, when to use ViT, segmentation architecture, classification backbone, nnU-Net ResEnc, MedNeXt, STU-Net, nnInteractive, VISTA3D, SAM-Med3D, Mamba, U-Mamba, interactive segmentation, labelling acceleration, promptable segmentation, nnDetection, lesion detection, ConvNeXt, YOLO, YOLOv8, RT-DETR, DETR, RetinaNet, detection architecture, RETFound, UNI, CONCH, RAD-DINO, Merlin, medical foundation model, pathology foundation model, domain transfer, diffusion model, latent diffusion, ControlNet, MAISI, image synthesis, GAN, CycleGAN, Pix2Pix, source a model, vet a model, pick a model, model provenance, model dossier, pretrained weights, checkpoint, HuggingFace model, GitHub model, model licence, weight provenance, is this model independent, benchmark overlap, trained on my test set, data contamination, model version pin, third-party model, can I use this model"

Model-Selection Skill

Two questions, in this order. Phases 1–3 answer a literature question with a stable answer: which architecture **family** suits the task. Phases 4–6 answer a provenance question: which concrete **artifact** — repository, revision, checkpoint — will be run, and what its numbers may claim. The skill writes decision notes and runs one stdlib gate; it never downloads, runs, trains or benchmarks a model, and it describes archetypes, not a live SOTA leaderboard.

Elsewhere: building the repo → `/model-scaffold`; profiling data and planning preprocessing → `/imaging-data`; validation design, metrics, uncertainty, explainability → `/model-assessment`; documenting a model you built → `/model-card`; an LLM/MLLM, including benchmark contamination in that setting → `/mllm-eval`; AI-vs-expert reader study → `/design-ai-benchmarking`.

Workflow

Phase 1 — Frame the question

State the **task** (classification / segmentation / detection / synthesis / transfer), the **modality and dimensionality** (2-D vs 3-D volume), the **labelled-data scale** (events / structures, not just images), **label availability** (lots / few / unlabelled pool), and the constraints (class imbalance, small structures, interpretability, deployment compute).

Phase 2 — Walk the decision tree, read the family card

Follow `${CLAUDE_SKILL_DIR}/references/index.md` (task → constraints → default pick). It routes to one card; each gives the source paper, core idea, when-to-use, medical-imaging use, reference implementation, and the typical validation setup for that class.

| Card | Families | |---|---| | `${CLAUDE_SKILL_DIR}/references/classification.md` | ResNet / DenseNet / EfficientNet / Inception / ConvNeXt / ViT / Swin / DeiT | | `${CLAUDE_SKILL_DIR}/references/segmentation.md` | U-Net / 3-D U-Net / V-Net / Attention & Residual U-Net / nnU-Net (+ ResEnc, MedNeXt, STU-Net) / SegResNet / Swin-UNETR / Mask R-CNN | | `${CLAUDE_SKILL_DIR}/references/detection.md` | nnDetection / Faster R-CNN + FPN / Mask R-CNN / RetinaNet / YOLO / DETR | | `${CLAUDE_SKILL_DIR}/references/synthesis.md` | Pix2Pix / CycleGAN / SPADE / diffusion (DDPM, latent) / VAE / fastMRI reconstruction | | `${CLAUDE_SKILL_DIR}/references/foundation_models.md` | SAM / MedSAM / MedSAM2 / nnInteractive / VISTA3D / TotalSegmentator / SegVol / BiomedCLIP / DINO / MAE / SimCLR / MoCo | | `${CLAUDE_SKILL_DIR}/references/graph.md` | GCN / GraphSAGE / GAT / GIN / BrainGNN for connectomes and population graphs (PyTorch Geometric / DGL; not scaffolded by `/model-scaffold`) |

Never recommend an architecture for a modality or data scale it does not suit (a from-scratch ViT on a few hundred images, 2-D slices for a volumetric structure) — the decision-tree constraints exist to prevent exactly that. If asked for "the best" model, say the zoo is a curated archetype map, not a current SOTA ranking.

Phase 3 — Write the architecture decision note

Write `decisions/architecture_choice.md`: the task, the chosen architecture, its **source paper** (mandatory — the Methods cite it), the reason against the constraints, the runner-up and why not, and the matching `/model-scaffold` template. Never invent a benchmark number or paper claim: cite it (verify via `/search-lit`), or write `[VERIFY]` and ask. A reference without a `/search-lit`-confirmed DOI/PMID is marked `[UNVERIFIED - NEEDS MANUAL CHECK]`.

Phase 4 — Write the model dossier for the concrete artifact

When a concrete candidate exists (a GitHub repo, a Hugging Face checkpoint, a paper's released weights), write `model_dossier.json` recording what is **known**; leave unknowns unstated, never guessed — the gate turns an absence into a finding.

{
  "model": "OrganSeg-3D v2.5.1",
  "source": {"kind": "github", "url": "...", "version": "v2.5.1", "commit": "abc1234"},
  "licence": {"spdx": "Apache-2.0", "verified_from": "LICENSE at commit abc1234"},
  "intended_use": "research",
  "weights": {"pretrained": false},
  "task": {"model": "3d_ct_organ_segmentation", "study": "3d_ct_organ_segmentation"},
  "reported_validation": [{"dataset": "ExampleBench", "metric": "Dice", "source": "J Ex 2021"}],
  "developed_on": ["ExampleBench"],
  "evaluation_arms": [{"name": "external", "dataset": "OtherCohort-2026"}],
  "hardware": {"claimed": "any CUDA GPU", "verified_on": "GTX 1080 Ti", "verified": true}
}

Read each field from the artifact, not fro