SVM Model Doc

Overview

SVM is a classical non-deep-learning method for neuroimaging-based disease classification.

• Model family: non-deep-learning supervised classification method
• Typical objectives:
  • classify patient vs control groups from neuroimaging features
  • build discriminative models from ROI features or tabular summaries
  • export predictive scores and evaluation metrics
• Primary input: preprocessed fMRI / sMRI derived features, labels, optional covariates
• Primary output: class predictions, decision scores, cross-validation metrics

In NeuroClaw, this document is model-level guidance for SVM-based disease classification workflows rather than deep learning phenotype prediction.

Upstream preparation should usually be delegated to:

• fmri-skill for fMRI preprocessing and ROI / voxel feature preparation
• smri-skill for structural feature extraction when disease classification uses sMRI
• nilearn-tool for concrete SVM fitting on prepared feature tables

Research use only.

---

Quick Start

1) Prepare disease classification inputs

Expected inputs:

• subject-level labels such as patient / control
• preprocessed imaging features
• optional covariates such as age, sex, site
• optional train / validation / test split definition

If features are not ready, delegate preprocessing to fmri-skill or smri-skill first.

2) SVM route

Representative operations:

• prepare ROI-wise or tabular neuroimaging features
• standardize features within the training fold
• fit linear or kernel SVM for disease classification
• export predictions, decision scores, and performance metrics

Example execution route:

# delegated through claw-shell after features are prepared
python skills/nilearn-tool/scripts/svm_classifier_reference.py \
  --features path/to/features.csv \
  --labels path/to/labels.csv \
  --target diagnosis \
  --cv 5 \
  --output-dir run_models_output/svm

---

Input / Output Contract

Required inputs

• subject-level labels for disease classification
• feature table or ROI summary matrix

Optional inputs

• confounds or covariates table
• train / validation / test split file
• hyperparameter settings such as kernel, C, or number of CV folds

Produced outputs

• predicted labels and decision scores
• cross-validation metrics such as accuracy, AUC, sensitivity, specificity
• fitted model artifact or coefficient table

---

Recommended Delegation

• imaging preprocessing and feature preparation -> fmri-skill and/or smri-skill
• concrete implementation of SVM -> nilearn-tool
• shell execution and logging -> claw-shell

No execution before explicit plan confirmation.

---

When to Use SVM

• The user wants classical disease classification instead of a deep learning model.
• The dataset size is moderate and model interpretability matters.
• ROI-level features are already prepared and SVM is sufficient.
• The task is case-control prediction, diagnosis support, or cross-validated disease discrimination.

---

Limitations and Notes

• SVM performance depends strongly on feature engineering, scaling, and leakage-free cross-validation.
• Site effects and confounds can dominate disease classification if not controlled properly.
• Small sample sizes can lead to optimistic estimates unless split strategy is rigorously managed.

---

Reference

• Cortes C, Vapnik V. Support-vector networks.
• Nilearn decoding documentation: https://nilearn.github.io/stable/decoding/index.html

Created At: 2026-04-14 00:34 HKT Last Updated At: 2026-04-14 00:45 HKT Author: chengwang96