Pose-estimated 3D data of infant spontaneous activity from Helsinki and Pisa

DOI10.5281/zenodo.14269867Zenodo14269867MaRDI QIDQ6707673

Dataset published at Zenodo repository.

Author name not available (Why is that?)

Publication date: 3 December 2024

Copyright license: No records found.

Description

This dataset contains pose-estimated 3D (and 2D proxy) data as well as trained models for generating infant Kinetic Age, collected from research trials in Helsinki and Pisa. The dataset is organized into separate archives for metadata, data streams, trained models, and predictions. Below is a detailed breakdown of the dataset contents: Metadata metadata/combined.csv - test_id: Unique identifier for each infant. - corrected_age: Corrected age of the infant in days. - outcome: Neurodevelopmental outcome labels (0 typical, 1 weak impairment, 2 MNI). Data data/features.csv - Handcrafted movement features computed for each, by experts annotated, useful video segment. data/streams/combined/*.feather - 3D recording segment, with 18 J, B, V, A streams over 600 time steps. data/streams_2d/combined/*.feather - 2D recording segment, with 18 J, B, V, A streams over 600 time steps. Results results/model/fold_n/... - train_predictions.npy: Predictions made on the training set. - val_predictions.npy: Predictions made on the validation set. - best_model.ckpt: Checkpoint file containing the saved model weights. - metadata.json: Metadata describing the fold, including training parameters and validation segments. - scatter_*.png: Regression results. Predictions predictions/jb-aagcn-coord-xy_predictions.csv - Model predictions for the 2D model on MNI samples. predictions/jb-aagcn-coord_predictions.csv - Model predictions for the 3D model on MNI samples.

This page was built for dataset: Pose-estimated 3D data of infant spontaneous activity from Helsinki and Pisa