MétaCan
Menu
Back to cohort
Record W4414015774 · doi:10.11159/mhci25.001

Improving Medical Imaging Efficiency with AI Algorithms Running on Dedicated AI Chips

2025· article· en· W4414015774 on OpenAlexvenueno aff
Dalila B. Megherbi

Bibliographic record

VenueProceedings of the World Congress on Electrical Engineering and Computer Systems and Science · 2025
Typearticle
Languageen
FieldMedicine
TopicRadiomics and Machine Learning in Medical Imaging
Canadian institutionsnot available
Fundersnot available
KeywordsComputer scienceMedical imagingArtificial intelligenceAlgorithm

Abstract

fetched live from OpenAlex

The application of Artificial Intelligence (AI) in Medical imaging has been facing several challenges, including issues related to medical misdiagnoses and delays primarily caused by inaccurate data and a lack of interoperability, which hampers effective data sharing.The enormous volume of images generated daily can overwhelm traditional central processing units (CPUs), making it challenging to manage and analyze this data efficiently.Additionally, ensuring the quality and consistency of images used in clinical tests presents another obstacle.To address these issues, AI is being increasingly implemented in the field of medical imaging.AI technology enhances diagnostic accuracy and streamlines workflows by utilizing specialized AI chips to process medical images more effectively.These advanced algorithms can rapidly analyze images, detecting patterns and abnormalities that might be missed by human observers.As a result, AI enables earlier disease detection, improves diagnostic precision, and enhances treatment planning, among others, providing a significant advantage over conventional processing methods.In relatively recent years, machine learning (ML) and, more specifically, deep learning have emerged as a significant trend in medical imaging.However, deploying machine learning in this field necessitates fast and optimized hardware to handle the extensive data processing required by these models.CPUs have limitations in their computational processing, which makes them inadequate for effectively executing machine learning (ML) algorithms.Advancements in AI chip technology have enhanced the capabilities of these algorithms.Neural Network Accelerators (NNAs) and Neural Network Processors (NPUs) are specialized processors optimized to handle specific neural network capabilities at the hardware level.These technologies have shown promising applications in medical imaging.In this presentation, we will conduct a comparative analysis of CPUs and specialized AI chips, including Neural Processing Units (NPUs).We will highlight their differing performance capabilities in terms of speed and precision in medical imaging applications.Additionally, we will analyze the advantages of NPUs compared to CPUs and explore the hardware architecture of a System-on-a-Chip (SoC) that utilizes distributed multi-soft processors specifically for medical imaging.We will also discuss a hardware architecture implementation of reinforcement learning, which is highly sought after in the fields of AI and machine learning.

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)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.977
Threshold uncertainty score0.484

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0010.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.001
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 head0.003
GPT teacher head0.228
Teacher spread0.225 · 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 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
Published2025
Admission routes1
Has abstractyes

Explore more

Same venueProceedings of the World Congress on Electrical Engineering and Computer Systems and ScienceSame topicRadiomics and Machine Learning in Medical ImagingFrench-language works237,207