MétaCan
Menu
Back to cohort
Record W4366826390 · doi:10.3390/bdcc7020082

Analysis of 2D and 3D Convolution Models for Volumetric Segmentation of the Human Hippocampus

2023· article· en· W4366826390 on OpenAlexfundno aff
You Sheng Toh, Carol Anne Hargreaves

Bibliographic record

VenueBig Data and Cognitive Computing · 2023
Typearticle
Languageen
FieldComputer Science
TopicMedical Image Segmentation Techniques
Canadian institutionsnot available
FundersNational Institute of Biomedical Imaging and BioengineeringCanadian Institutes of Health ResearchNational Institutes of HealthGenentechIXICOH. Lundbeck A/SServierEisaiNorthern California Institute for Research and EducationBioClinicaF. Hoffmann-La RocheUniversity of Southern CaliforniaBiogenU.S. Department of DefenseMeso Scale DiagnosticsAlzheimer's Disease Neuroimaging InitiativeNovartis Pharmaceuticals CorporationPfizerEli Lilly and CompanyBristol-Myers SquibbNational Institute on AgingAlzheimer's AssociationFoundation for the National Institutes of Health
KeywordsHippocampusSegmentationComputer scienceConvolution (computer science)Task (project management)Similarity (geometry)Artificial intelligenceHippocampal formationPattern recognition (psychology)AtrophyDiceMarket segmentationNeurosciencePsychologyMedicinePathologyMathematicsStatisticsArtificial neural network

Abstract

fetched live from OpenAlex

Extensive medical research has revealed evidence of a strong association between hippocampus atrophy and age-related diseases such as Alzheimer’s disease (AD). Therefore; segmentation of the hippocampus is an important task that can help clinicians and researchers in diagnosing cognitive impairment and uncovering the mechanisms behind hippocampal changes and diseases of the brain. The main aim of this paper was to provide a fair comparison of 2D and 3D convolution-based architectures for the specific task of hippocampus segmentation from brain MRI volumes to determine whether 3D convolution models truly perform better in hippocampus segmentation and also to assess any additional costs in terms of time and computational resources. Our optimized model, which used 50 epochs and a mini-batch size of 2, achieved the best validation loss and Dice Similarity Score (DSC) of 0.0129 and 0.8541, respectively, across all experiment runs. Based on the model comparisons, we concluded that 2D convolution models can surpass their 3D counterparts in terms of both hippocampus segmentation performance and training efficiency. Our automatic hippocampus segmentation demonstrated potential savings of thousands of clinician person-hours spent on manually analyzing and segmenting brain MRI scans

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 imitation

Not 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.

metaresearch head score (Codex)0.001
metaresearch head score (Gemma)0.003
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.017
Threshold uncertainty score0.033

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0010.003
Meta-epidemiology (narrow)0.0010.000
Meta-epidemiology (broad)0.0000.001
Bibliometrics0.0010.000
Science and technology studies0.0000.000
Scholarly communication0.0010.001
Open science0.0010.000
Research integrity0.0010.001
Insufficient payload (model declined to judge)0.0010.000

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.

Opus teacher head0.117
GPT teacher head0.351
Teacher spread0.235 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designSimulation or modeling
Domainnot available
GenreEmpirical

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".

Quick stats

Citations0
Published2023
Admission routes1
Has abstractyes

Explore more

Same venueBig Data and Cognitive ComputingSame topicMedical Image Segmentation TechniquesFrench-language works237,207