Skip to content
Projekte

jibikbam/CNN-3D-images-Tensorflow

Auf GitHub ansehen

jibikbam/CNN-3D-images-Tensorflow has 267 GitHub stars and 110 forks. GitHot measured 0 stars in 24 hours. Last checked 2026-11-14 18:29 UTC. Last push 2019-10-29.

So werden diese Zahlen gemessen

267 Sterne110 Forks

Gemessener Sternzuwachs

in 24h
0
in 7 days
Verlauf wird gesammelt
in 15 days
Verlauf wird gesammelt
in 1 month
Verlauf wird gesammelt

Nettoänderungen in überlappenden Zeiträumen, keine Tagessummen. Fehlende Messwerte bleiben leer. Sterne zeigen Interesse, nicht Qualität.

Über uns

3D image classification using CNN (Convolutional Neural Network)

Maschinelles Lernen

README.md

Original

CNN 3D Images using Tensorflow

  • Goal: MRI classification task using CNN (Convolutional Neural Network)

  • Code Dependency: Tensorflow 1.0, Anaconda 4.3.8, Python 2.7

  • Difficulty in learning a model from 3D medical images

    1. Data size is too big. e.g., 218x182x218 or 256x256x40
    2. There is only limited number of data. In other words, training size is too small.
    3. All image looks very similar and only have subtle difference between subjects.

  • Possible solutions

    1. Be equipped with good machine especially the RAM
    2. Downsample images in the preprocessing
    3. Data augmentation e.g., rotate, shift, combination
    4. Transfer learning