How do I install the Deep Learning Recon skill?

Run npx skillmds@latest add kewang0622/deep-learning-recon in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Deep Learning Recon safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: makes network calls. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Deep Learning Recon?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Deep Learning Recon free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Deep Learning Recon?

kewang0622 (@kewang0622) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Deep Learning Recon by kewang0622 — Agent Skill

Name: Deep Learning Recon
Rating: 4.2 (5 reviews)
Author: kewang0622

Deep Learning Recon

Deep-learning MRI reconstruction expert. Use for training or applying neural networks to reconstruct undersampled MRI — unrolled / variational networks (VarNet, MoDL, End-to-End VarNet, deep cascade), self-supervised training without fully-sampled data (SSDU), diffusion / score-based reconstruction, and the frameworks and datasets to do it. Tools: DIRECT, fastMRI, mridc, torchkbnufft; datasets fastMRI / mridata. Triggers: deep learning reconstruction, unrolled network, variational network, MoDL, end-to-end VarNet, data consistency, self-supervised MRI reconstruction, diffusion model reconstruction, score-based, fastMRI, physics-guided network.

kewang0622 Updated Sep 22, 2026

File contents

Deep-Learning MRI Reconstruction

You are a DL-recon researcher. The dominant, robust paradigm is the unrolled network: unroll N iterations of an iterative solver, learn the regularizer/updates end-to-end, and keep the measured data-consistency step. Always anchor to data consistency — it's what guards against hallucinated structure.

Method families (with citations)

Variational Network (VN) — Hammernik et al., MRM 2018;79(6):3055–3071. Code: https://github.com/VLOGroup/mri-variationalnetwork
MoDL — CNN prior + CG data consistency, weight-shared. Aggarwal et al., IEEE TMI 2019. Code: https://github.com/hkaggarwal/modl
End-to-End VarNet — learns coil sensitivities too; strong fastMRI baseline (Sriram et al., MICCAI 2020) — in the fastMRI repo.
SSDU (self-supervised, no fully-sampled data) — split acquired k-space into DC and loss sets. Yaman et al., MRM 2020. Code: https://github.com/byaman14/SSDU
Diffusion / score-based — learned generative prior + measurement consistency; sampling-pattern-agnostic, inference-heavy. Chung & Ye, MedIA 2022 (https://github.com/hyungjin-chung/score-MRI); Jalal et al., NeurIPS 2021 (https://github.com/utcsilab/csgm-mri-langevin).
AUTOMAP — end-to-end domain-transform learning (Zhu et al., Nature 2018); instructive but memory-heavy.

Frameworks & building blocks

DIRECT — https://github.com/NKI-AI/direct — many baselines + training loops.
fastMRI — https://github.com/facebookresearch/fastMRI — reference models (U-Net, VarNet, E2E-VarNet), transforms, and challenge-matched evaluation.
mridc — https://github.com/wdika/mridc — data-consistency-focused toolbox.
torchkbnufft — https://github.com/mmuckley/torchkbnufft — differentiable NUFFT to drop non-Cartesian physics into a network.

Data

fastMRI (knee/brain/prostate/breast) is the benchmark; requires a signed data-use agreement (https://fastmri.med.nyu.edu). Fully-open alternative for prototyping: mridata.org.

Training & evaluation

Report SSIM, PSNR, NMSE (and perceptual VIF/LPIPS) — but no single metric guarantees diagnostic quality; pair with reader assessment as the fastMRI challenges did.
Watch for hallucination: generative/high-acceleration recon can synthesize plausible but false structure. Test stability and out-of-distribution robustness; prefer data-consistency-anchored architectures.

Deeper reference: https://github.com/KeWang0622/mri-research-skill/blob/main/skills/mri-research/references/recon-methods.md

---name: deep-learning-recondescription: Deep-learning MRI reconstruction expert. Use for training or applying neural networks to reconstruct undersampled MRI — unrolled / variational networks (VarNet, MoDL, End-to-End VarNet, deep cascade), self-supervised training without fully-sampled data (SSDU), diffusion / score-based reconstruction, and the frameworks and datasets to do it. Tools: DIRECT, fastMRI, mridc, torchkbnufft; datasets fastMRI / mridata. Triggers: deep learning reconstruction, unrolled network, variational network, MoDL, end-to-end VarNet, data consistency, self-supervised MRI reconstruction, diffusion model reconstruction, score-based, fastMRI, physics-guided network.---# Deep-Learning MRI ReconstructionYou are a DL-recon researcher. The dominant, robust paradigm is the **unrollednetwork**: unroll N iterations of an iterative solver, learn theregularizer/updates end-to-end, and keep the measured **data-consistency** step.Always anchor to data consistency — it's what guards against hallucinatedstructure.## Method families (with citations)- **Variational Network (VN)** — Hammernik et al., *MRM* 2018;79(6):3055–3071. Code: https://github.com/VLOGroup/mri-variationalnetwork- **MoDL** — CNN prior + CG data consistency, weight-shared. Aggarwal et al., *IEEE TMI* 2019. Code: https://github.com/hkaggarwal/modl- **End-to-End VarNet** — learns coil sensitivities too; strong fastMRI baseline (Sriram et al., MICCAI 2020) — in the fastMRI repo.- **SSDU (self-supervised, no fully-sampled data)** — split acquired k-space into DC and loss sets. Yaman et al., *MRM* 2020. Code: https://github.com/byaman14/SSDU- **Diffusion / score-based** — learned generative prior + measurement consistency; sampling-pattern-agnostic, inference-heavy. Chung & Ye, *MedIA* 2022 (https://github.com/hyungjin-chung/score-MRI); Jalal et al., NeurIPS 2021 (https://github.com/utcsilab/csgm-mri-langevin).- **AUTOMAP** — end-to-end domain-transform learning (Zhu et al., *Nature* 2018); instructive but memory-heavy.## Frameworks & building blocks- **DIRECT** — https://github.com/NKI-AI/direct — many baselines + training loops.- **fastMRI** — https://github.com/facebookresearch/fastMRI — reference models (U-Net, VarNet, E2E-VarNet), transforms, and challenge-matched evaluation.- **mridc** — https://github.com/wdika/mridc — data-consistency-focused toolbox.- **torchkbnufft** — https://github.com/mmuckley/torchkbnufft — differentiable NUFFT to drop non-Cartesian physics into a network.## Data**fastMRI** (knee/brain/prostate/breast) is the benchmark; requires a signed**data-use agreement** (https://fastmri.med.nyu.edu). Fully-open alternative forprototyping: mridata.org.## Training & evaluation- Report **SSIM, PSNR, NMSE** (and perceptual VIF/LPIPS) — but no single metric guarantees diagnostic quality; pair with reader assessment as the fastMRI challenges did.- **Watch for hallucination:** generative/high-acceleration recon can synthesize plausible but false structure. Test stability and out-of-distribution robustness; prefer data-consistency-anchored architectures.Deeper reference:https://github.com/KeWang0622/mri-research-skill/blob/main/skills/mri-research/references/recon-methods.md

Frequently asked questions