A pneumonia detection system based on MobileNetv2 network and model callback
Bibliographic record
Abstract
This study focuses on automatically classifying radiographic images of the chest region into standard classification, COVID-19 classification, and viral pneumonia classification by utilizing complex neural networks. Using a chest radiograph dataset from the academic bastion Université de Montréal, the study introduces an innovative paradigm rooted in MobileNetV2. We conducted a comparative analysis to evaluate the efficacy of this avant-garde model by juxtaposing it with the typical DenseNet121 and RESNET50 popular in the field of medical image classification. This exploration revealed MobileNetV2 as an ingenious model distinguished by its tiny scale and commendable accuracy. Using DW convolution design greatly reduces the computational complexity and parameter count. Regarding the composition of the architecture, transfer learning is used to attach global mean pooling and fully connected layers on top of MobileNetV2, and is customized for nuanced tripartite classification work. Comparative evaluation shows that the MobileNetV2 model has only 2,643,187 parameters, completes training in just 37 seconds, and has an accuracy of 98.48%. In contrast, although DenseNet121 and RESNET50 demonstrate commendable proficiency, their large model dimensions and lengthy training intervals limit their usefulness in resource-limited environments. The findings highlight the superior performance of the MobileNetV2 model in the field of chest X-ray classification, providing a simplified and efficient alternative for deployment on mobile and embedded devices.
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 | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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".