{
  "id": 5398501,
  "title": "MIND the gap: methodological considerations and guidance for structural MRI similarity network analysis with MIND",
  "url": "https://urgent.news/2026/09/03/mind-the-gap-methodological-considerations-and-guidance-for",
  "topic": "science",
  "section": "Science",
  "published": "2026-09-03T00:00:00.000Z",
  "source": {
    "name": "bioRxiv",
    "slug": "biorxiv",
    "url": "https://www.biorxiv.org/content/10.64898/2026.08.28.747811v1?rss=1"
  },
  "original_language": "en",
  "account": "Structural similarity networks measure the likeness of structural traits across brain regions, giving researchers a broad perspective on the organization of cortical structure. Morphometric inverse divergence, known as MIND, is a multivariate similarity score for cortical areas. This score relies on Kullback-Leibler (KL) divergence, comparing the distributions of multiple MRI features or morphometric variables measured at the level of voxels or vertices. MIND is known for being both technically robust and biologically meaningful, and it is frequently used to gauge cortico-cortical similarity in clinical and developmental network neuroscience.\n\nThis article offers a detailed explanation of the underlying principles of KL divergence and MIND, identifying potential sources of bias, key decision points in designing a MIND processing pipeline, and advice for minimizing technical risks when using MIND as a metric of cortical similarity. The authors use simulated data and MRI datasets from adults (UK Biobank, consisting of 500 T1-weighted and diffusion scans) and neonates (Developing Human Connectome Project, with 752 T2-weighted scans) to illustrate how the KL divergence estimator in MIND can be affected or biased by five properties of input MRI feature maps. These properties include smoothness, proportion of identical values, the choice between native or common space analysis and vertex mesh resolution, parcellation decision, and covariance between input features.\n\nThe researchers provide practical guidance for investigators looking to develop the MIND processing pipeline that best suits the constraints and opportunities of their MRI data. These recommendations clarify which pipeline steps should be used sparingly—such as smoothing vertex maps—and which should be employed with informed caution, like choosing a parcellation or resampling vertex mesh. Additionally, the authors suggest using principal component analysis to preprocess multivariate MRI features as a new method for more accurate MIND estimation. To advance the field of structural MRI similarity network analysis and promote the adoption of reliable MIND methods, the article also makes the code used to generate the study results available to the public as an open resource.",
  "summary": "Structural similarity networks quantify the similarity of structural properties across cortical regions, providing a macroscopic window onto the organisation of cortical architecture. Morphometric inverse divergence (MIND) is a multivariate metric of similarity between cortical areas, based on the Kullback-Leibler (KL) divergence between areal distributions of multiple MRI features or…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}