Medicine image classification using deep learning: highlighting the MedNet-MoBiL hybrid model
Bibliographic record
Abstract
Deep learning has transformed image classification tasks across many domains, including medical diagnostics. Medicine wrappers, boxes, and strips often contain valuable but complex information that can be difficult to read and comprehend manually. This complexity drives users to seek additional knowledge online. However, traditional search engines often present a large volume of results, requiring users to manually filter through multiple links to find relevant information, which can be time-consuming. To address this issue, we propose a lightweight pipeline that leverages MobileNetV2 for image classification and Optical Character Recognition (OCR) for extracting text content from medicine packaging. The extracted text is processed using the RAKE algorithm to identify significant keywords, which are then matched with relevant URLs through a Google Search API. To ensure relevance, retrieved links are ranked using ROUGE scores. Performance metrics demonstrate the model's efficiency, with ROUGE-1 achieving 90% Recall, 95% F1-Score, and 90% Accuracy, and ROUGE-L achieving 83% Recall, 91% F1-Score, and 83% Accuracy. The pipeline was trained and validated on a curated dataset of 3,000 real-world medicine packaging images, publicly available on GitHub. These results highlight the novelty and practicality of our solution for automating medical information retrieval from packaging, using an interpretable and scalable deep learning-driven approach.
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 | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.004 | 0.003 |
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".