acusim: a synthetic dataset for cervicocranial acupuncture points localisation

Sun, Qilei1 ; Ma, Jiatao 1 ; Craig, Paul1 ; Dai, Linjun2 ; Lim, EngGee1

Published Mar 28, 2025 on Dryad. https://doi.org/10.5061/dryad.zs7h44jkz

Data files

Mar 28, 2025 version files 13.04 GB

acuSim.zip
13.04 GB
README.md
3.05 KB

Abstract

The locations of acupuncture points (acupoints) differ among human individuals due to variations in factors such as height, weight, and fat proportions. However, acupoint annotation is expert-dependent, labour-intensive, and highly expensive, which limits the data size and detection accuracy. In this paper, we introduce the "AcuSim" dataset as a new synthetic dataset for the task of localising points on the human cervicocranial area from an input image using an automatic render and labelling pipeline during acupuncture treatment. It includes the creation of 63,936 RGB-D images and 504 synthetic anatomical models with 174 volumetric acupoints annotated, to capture the variability and diversity of human anatomies. The study validates a convolutional neural network (CNN) on the proposed dataset with an accuracy of 99.73% and shows that 92.86% of predictions in the validation set align within a 5mm threshold of margin error when compared to expert-annotated data. This dataset addresses the limitations of prior datasets and can be applied to applications of acupoint detection and visualization, further advancing automation in Traditional Chinese Medicine (TCM).

Dryad DOI: https://doi.org/10.5061/dryad.zs7h44jkz

Dataset Overview

A multi-view acupuncture point dataset containing:

64x64, 128x128, 256x256, 512×512 and 1024x1024resolution RGB images
Corresponding JSON annotations with:
- 2D/3D keypoint coordinates
- Visibility weights (0.9-1.0 scale)
- Meridian category indices
- Visibility masks
174 standard acupuncture points (map.txt)
Occlusion handling implementation

Key Features

Multi-view Rendering: Generated using Blender 3.5 with realistic occlusion simulation
Structured Annotation:
- Default initialization for occluded points ([0.0, 0.0, 0.5])
- Meridian category preservation for occluded points
- Weighted visibility scoring
ML-Ready Format: Preconfigured PyTorch DataLoader implementation

Dataset Structure

dataset_root/
├── map.txt                 # Complete list of 174 acupuncture points
├── train/
│   ├── image/img_512/      # Training images (*.png)
│   └── label/label/        # JSON annotations (*.json)
└── test/                   # [Optional] Similar structure for testing

Quick Start

Dependencies

Python 3.10
PyTorch 2.0+
TensorFlow 2.60 (GPU recommended)
CUDA 12.4
cuDNN 9.5.0
blender 3.6.n LTS under python 3.10

Basic Usage

from acuSim_dataloader_modified import AcuPointsDataset, create_category_encoding

# Initialize dataset
dataset = AcuPointsDataset(
    image_dir="path/to/images",
    json_dir="path/to/labels",
    target_size=(512, 512),
    transform=transforms.Compose([
        transforms.Resize((512, 512)),
        transforms.ToTensor()
    ])
)

# Create DataLoader
train_loader = DataLoader(
    dataset,
    batch_size=32,
    shuffle=True,
    num_workers=4
)

Advanced Features

Key Components

Occlusion Handling:

Preserves meridian category for occluded points
Automatic mask generation (create_mask())
Weighted visibility scores in JSON annotations

Code Availability

Script	Purpose	Requirements
`script.py`	Blender rendering pipeline	Blender 3.5
`xuewei5.py`	Visualization examples	Python 3.10
`dataloader_modified.py`	PyTorch/TF data loading	TF 2.60+
`occlusion_detection.py`	Occlusion generation	CUDA 12.4

Citation

Please cite our work when using this dataset:

Sun, Qilei; Ma, Jiatao; Craig, Paul et al. (2025). acusim: a synthetic dataset for cervicocranial acupuncture points localisation [Dataset]. Dryad. https://doi.org/10.5061/dryad.zs7h44jkz