# MrGiovanni/UNetPlusPlus

[IEEE TMI Best Paper Award] Official Implementation for UNet++

Repository: https://github.com/MrGiovanni/UNetPlusPlus
Canonical: https://ross.abutalabs.com/products/unetplusplus
Homepage: https://www.zongweiz.com/research
Language: Python
License: NOASSERTION
License Family: other
Topics: medical-imaging, segmentation, unet, biomedical-imaging, ct, keras, medical-image-segmentation, mri, pytorch, semantic-segmentation, skip-connections
Last push: 2026-08-25T17:17:31+00:00

## Health v2 (maintenance only)
Score: 77/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 35, longevity 100
- inputs: {"age_days": 3061, "days_push": 8, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2679, forks 565 (observed 2026-08-28T04:07:09.960810+00:00)

## What it is
Official implementation of UNet++, a nested U-Net architecture for medical image segmentation, in both Keras and PyTorch. It redesigns skip connections with dense decoder connections to exploit multiscale features and address the unknown optimal depth of the original U-Net.

## Use cases
- segment medical images like CT and MRI scans
- implement UNet++ for biomedical image segmentation
- train a nested U-Net with redesigned skip connections
- segment organs or tumors in medical imaging datasets
- reproduce results from the UNet++ TMI paper
- compare U-Net variants for semantic segmentation research

## When to choose
- you need the canonical, paper-backed UNet++ implementation for medical image segmentation
- you work with CT or MRI data and want dense skip connections across decoder depths
- you want both Keras and PyTorch reference code from the original authors
- you are researching segmentation architectures and need a well-cited baseline

## When to avoid
- you need a general-purpose segmentation toolkit beyond the UNet++ architecture
- you require a permissively licensed library for commercial embedding, since the license is not a standard OSI license
- you want a maintained production pipeline with pretrained weights and inference APIs rather than research code
- your project is not image segmentation related

## Facets
- artifact type: library
- maturity: stable
- function: machine-learning, deep-learning, image-processing, computer-vision
- domain: deep-learning, computer-vision, healthcare
- platform: python, cross-platform
- tags: unet-plus-plus, semantic-segmentation, medical-image-segmentation, nested-unet, skip-connections, biomedical-imaging, keras, pytorch, research-code, ieee-tmi, medical-imaging

## Member repositories
- MrGiovanni/UNetPlusPlus (main) score 77

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:09.960810+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:17:57.066960+00:00, confidence not recorded.
  - readme: https://github.com/MrGiovanni/UNetPlusPlus (fetched 2026-08-28T04:07:09.960810+00:00, sha 1426a1af7d4d)
  - homepage: https://www.zongweiz.com/research (fetched 2026-08-29T10:00:14.850626+00:00, sha 44f817137c6e)
- Data as of 2026-08-30T08:39:29.467469+00:00.