toderian/autism-detector

Autism Spectrum Disorder Screening Model

Model Description

A feedforward neural network for autism spectrum disorder (ASD) risk screening using 8 structured clinical input features.

Important: This is a screening tool, NOT a diagnostic instrument. Results must be interpreted by qualified healthcare professionals.

Intended Use

• Primary use: Clinical decision support for ASD screening
• Users: Healthcare professionals, clinical software systems
• Out of scope: Self-diagnosis, definitive diagnosis

Input Features

Field	Type	Valid Values	Description
developmental_milestones	categorical	N, G, M, C	Normal, Global delay, Motor delay, Cognitive delay
iq_dq	numeric	20-150	IQ or Developmental Quotient
intellectual_disability	categorical	N, F70.0, F71, F72	None, Mild, Moderate, Severe (ICD-10)
language_disorder	binary	N, Y	No / Yes
language_development	categorical	N, delay, A	Normal, Delayed, Absent
dysmorphism	binary	NO, Y	No / Yes
behaviour_disorder	binary	N, Y	No / Yes
neurological_exam	text	non-empty string	N for normal, or description

Output

{
  "prediction": "Healthy" | "ASD",
  "probability": 0.0-1.0,
  "risk_level": "low" | "medium" | "high"
}

Risk Level Thresholds

• Low: probability < 0.4
• Medium: 0.4 ≤ probability < 0.7
• High: probability ≥ 0.7

How to Use

import json
import torch
from pathlib import Path
from huggingface_hub import snapshot_download

# Download model
model_dir = Path(snapshot_download("toderian/autism-detector"))

# Load config
with open(model_dir / "preprocessor_config.json") as f:
    preprocess_config = json.load(f)

# Load model
model = torch.jit.load(model_dir / "autism_detector_traced.pt")
model.eval()

# Preprocessing function
def preprocess(data, config):
    features = []
    for feature_name in config["feature_order"]:
        if feature_name in config["categorical_features"]:
            feat_config = config["categorical_features"][feature_name]
            if feat_config["type"] == "text_binary":
                value = 0 if data[feature_name].upper() == feat_config["normal_value"] else 1
            else:
                value = feat_config["mapping"][data[feature_name]]
        else:
            feat_config = config["numeric_features"][feature_name]
            raw = float(data[feature_name])
            value = (raw - feat_config["min"]) / (feat_config["max"] - feat_config["min"])
        features.append(value)
    return torch.tensor([features], dtype=torch.float32)

# Example inference
input_data = {
    "developmental_milestones": "N",
    "iq_dq": 85,
    "intellectual_disability": "N",
    "language_disorder": "N",
    "language_development": "N",
    "dysmorphism": "NO",
    "behaviour_disorder": "N",
    "neurological_exam": "N"
}

input_tensor = preprocess(input_data, preprocess_config)
with torch.no_grad():
    output = model(input_tensor)
    probs = torch.softmax(output, dim=-1)
    asd_probability = probs[0, 1].item()

print(f"ASD Probability: {asd_probability:.2%}")
print(f"Prediction: {'ASD' if asd_probability > 0.5 else 'Healthy'}")

Training Details

• Dataset: 315 ASD patients + 100 healthy controls (415 total)
• Preprocessing: Min-max normalization for numeric, label encoding for categorical
• Architecture: Feedforward NN (input → 64 → 32 → 2)
• Loss: Cross-entropy
• Optimizer: Adam (lr=0.001)

Evaluation

Metric	Value
Accuracy	0.9759
F1 Score	0.9839
ROC-AUC	0.9913
Sensitivity	0.9683
Specificity	1.0000

Confusion Matrix (Test Set, n=83)

	Predicted Healthy	Predicted ASD
Actual Healthy	20	0
Actual ASD	2	61

Limitations

• Trained on limited dataset (415 samples)
• Healthy controls are synthetically generated
• Not validated across diverse populations
• Screening tool only, not diagnostic
• Requires all 8 input fields

Ethical Considerations

• Results should always be reviewed by qualified professionals
• Should not be used as sole basis for clinical decisions
• Model performance may vary across different populations
• False negatives (2 in test set) may delay intervention

Files

File	Description
autism_detector_traced.pt	TorchScript model (load with torch.jit.load())
config.json	Model architecture configuration
preprocessor_config.json	Feature preprocessing rules (JSON, no pickle)
model.py	Model class definition
requirements.txt	Python dependencies

Citation

@misc{asd_detector_2024,
  title={Autism Spectrum Disorder Screening Model},
  year={2024},
  publisher={HuggingFace},
  url={https://huggingface.co/toderian/autism-detector}
}