MétaCan
Menu
Back to cohort
Record W4406315931 · doi:10.35882/jeeemi.v7i1.571

A Novel Encoder Decoder Architecture with Vision Transformer for Medical Image Segmentation

2025· article· en· W4406315931 on OpenAlexaff
Saroj Bala, Kumud Arora, R Jeevitha, Rini Chowdhury, Prashant Kumar, C. Shobana Nageswari

Bibliographic record

VenueJournal of Electronics Electromedical Engineering and Medical Informatics · 2025
Typearticle
Languageen
FieldNeuroscience
TopicBrain Tumor Detection and Classification
Canadian institutionsArtificial Intelligence in Medicine (Canada)
Fundersnot available
KeywordsComputer scienceArtificial intelligenceSegmentationUpsamplingEncoderConvolutional neural networkImage segmentationDeep learningComputer visionOverfittingPattern recognition (psychology)Artificial neural networkImage (mathematics)

Abstract

fetched live from OpenAlex

Brain tumor image segmentation is one of the most critical tasks in medical imaging for diagnosis, treatment planning, and prognosis. Traditional methods for brain tumor image segmentation are mostly based on Convolution Neural Network (CNN), which have been proved very powerful but still have limitations to effectively capture long-range dependencies and complex spatial hierarchies in MRI images. Variability in the shape, size, and location of tumors may affect the performance and may get stuck into suboptimal outcomes. In these regards, new encoder-decoder architecture with the VisionTranscoder(ViT) is proposed, to enhance brain tumor detection and classification. The proposed VisionTranscoder exploits a transformer's ability in modeling global context through self-attention mechanisms, providing more inclusive interpretation of the intricate patterns in medical images and classification by capturing both local and global features. The proposed VisionTranscoder maintains the Vision Transformer in its encoder for processing images as sequences of patches to capture global dependencies often outside the view of traditional CNNs. Then the segmentation map is rebuilt at a high level of fidelity with the decoder through upsampling and skips connections to maintain detailed spatial information. The risk of overfitting is hugely reduced by design and advanced regularization techniques with extensive data augmentation. The dataset contains 7,023 human brain MRI images, all of which are in four different classes: glioma, meningioma, no tumor, and pituitary. Images from the 'no tumor' class, indicating an MRI scan without any detectable tumor, were taken from the Br35H dataset . The results show the efficiency of VisionTranscoder over a wide set of brain MRI scans, producing an accuracy of 98.5% with a loss of 0.05. This performance underlines the ability of it to accurately segment and classify a brain tumor without overfitting.

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 imitation

Not 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.

metaresearch head score (Codex)0.001
metaresearch head score (Gemma)0.002
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Bench or experimental · Consensus signal: none
GenreCandidate signal: Methods · Consensus signal: none
Teacher disagreement score0.898
Threshold uncertainty score0.535

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0010.002
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.000
Science and technology studies0.0000.000
Scholarly communication0.0000.000
Open science0.0000.000
Research integrity0.0000.001
Insufficient payload (model declined to judge)0.0000.000

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 head0.005
GPT teacher head0.259
Teacher spread0.254 · 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 teacher head, not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designBench or experimental
Domainnot available
GenreMethods

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

Citations4
Published2025
Admission routes1
Has abstractyes

Explore more

Same venueJournal of Electronics Electromedical Engineering and Medical InformaticsSame topicBrain Tumor Detection and ClassificationFrench-language works237,207