ICHD3 Primary Headache Syndromes Represented as a Diagnostic Matrix

A matrix representation of primary headache syndromes from the International Classification of Headache Disorders (ICHD3) enables automated diagnosis. The dataset encodes headache phenotypes and characteristics as a biadjacency matrix, with results showing diagnoses can be represented in a 63-dimensional vector space. It was created by Pengfei Zhang and last updated on 2026-05-11.

Use Cases

• Developing automated diagnostic tools based on the logical criteria structure of ICHD3.
• Analyzing relationships between headache phenotypes and characteristics using bipartite projection and Markov clustering.
• Investigating the fragmentation of specific diagnoses, such as paroxysmal hemicrania, which can be broken into 9 clusters.
• Exploring the basis vectors spanning the headache phenotype space derived from row reduction of the matrix.

Strengths

• Based on the ICHD3 gold standard for clinical trials and research in headache medicine.
• Enables systematic investigation of relationships between headache and phenotypes through mathematical representation.
• The matrix row reduction yields 63 basis vectors, providing a compact representation of the diagnostic space.

Limitations

• Dataset scope is limited to primary headache diagnoses only, excluding secondary headaches.
• The dataset is very small at 62.9 KB, indicating limited raw data content.
• Column-level documentation is absent; field semantics must be inferred after download.

Provenance

Source

figshare

Collection Method

Manual translation of ICHD3 diagnostic criteria into true/false statements and matrix encoding.

Freshness

Last updated 2026-05-11 14:55:31; freshness should be verified.