IC‐P‐127: Statistical Analysis of Automated Hippocampal Volumes in ADNI Dataset Reveals Center and Group Variability
Bibliographic record
Abstract
Hippocampal (HC) atrophy is a key diagnostic marker for Alzheimer's disease (AD). While manual segmentation by trained raters is the reference standard procedure for assessing HC volumes, a number of computer-based techniques have been proposed to automate this task. It is known in manual studies that different anatomical definitions and protocols result in heterogeneous estimates of normal HC volumes, from 2 to 5.3 cm3 (Geuze et al., Mol Psychiatry 2005;10:147-59). Our goal was to assess this variability by Centres and diagnostic Groups (AD, mild cognitive impairment (MCI) and controls (CTRL)) for automatically generated HC volumes. We downloaded automatically generated left and right HC volume data from the ADNI dataset (last access: Nov. 2009) for three different Centers: UCSF (FreeSurfer); UCSD (Semi-automated diffeomorphism); and U of Az. (SPM). Statistical analysis was performed using SAS (Cary, NC, USA). The total number of subjects available in the study was 766, from which 531 subjects had data from all three Centers. Summary statistics are presented in Table 1. Two-way ANOVA for left HC volumes showed statistically significant Center (p < 0.0001) and Group effects (p < 0.0001), as well as a weak interaction (Figure 1) leading to a significant Center by Group effect (p = 0.0001). Similar testing on right HC volumes showed equally significant Center (p < 0.0001) and Group effects (p < 0.0001), with smaller interaction (Figure 2) yet significant Center by Group effect (p = 0.0045).
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.008 | 0.017 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.002 | 0.003 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.003 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.005 | 0.002 |
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 source (direct Gemma or distilled Codex), 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".