Enhancing Performance of Linked Independent Component Analysis: Investigating the Influence of Subjects and Modalities
Bibliographic record
Abstract
In recent years, neuroimaging studies have increasingly been acquiring multiple modalities of data. The benefit of integrating multiple modalities through fusion lies in its ability to combine the unique strengths of each modality when analyzed collectively, as opposed to examining each one individually. In 2011, Adrian R. Groves proposed the Linked independent component analysis (LICA) method, which simultaneously models and discovers common features across multiple modalities. LICA has emerged as a powerful technique for analyzing multivariate data, particularly in neuroimaging and biomedical signal processing. The performance of LICA can be affected by the number of subjects and modalities. However, the detailed influence of the number of subjects and modalities on its performance remains an open question. In this study, we test the effects of the number of subjects and modalities on the performance of LICA using both simulated multimodal MRI data and the real multimodal MRI datasets from Autism Brain Imaging Data Exchange II (ABIDE II). Simulated data were utilized to evaluate the influence of subjects and modalities' variabilities. Real multi-site MRI data were used to demonstrate the advantages of multimodal fusion in identifying site-related components and removing site effects. Based on the simulation results, we found that increasing the number of modalities and subjects can improve the results when LICA can not recover the spatial maps or subject courses well. The correlation among subject courses from various modalities, the number of modalities, and the choice of components for decomposition all affect the linking performance of LICA. Our results from real-world datasets also demonstrated the advantages of multimodal fusion by LICA: 1) identify more site-related components; 2) remove more site effects.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".