On Improving Malaria Parasite Detection from Microscopic Images: A Comparative Analytics of Hybrid Deep Learning Models
Bibliographic record
Abstract
Malaria, a life-threatening mosquito-borne disease, contributes to a significantly high number of fatalities in tropical/sub-tropical regions due to inadequate detection technology, lack of laboratory experience, and other barriers. From the design perspective of a general-purpose point-of-care solution for detecting malaria along with other tropical diseases, malaria parasite detection from blood work needs to integrate accurate and fast detection capabilities. In this vein, in this paper, we develop three hybrid data-driven models in this paper that combine a convolutional neural network (CNN) with long short-term memory (LSTM), bi-directional LSTM (BiLSTM), and gated recurrent unit (GRU), respectively. CNN is employed in all three proposed models to extract the relevant features that are passed to two cascaded layers of Recurrent Neural Networks (RNNs) in each model that acts as a classifier. Based on the experiments conducted with a public dataset, we demonstrate that our designed CNN-GRU-GRU hybrid model outperformed the other models in terms of accuracy (96.01%), less type-I error rate (1.81%), and type-II error rate (2.18%). On the other hand, the CNN-LSTM-LSTM model was attributed to a low computing (training) time of just 4 minutes and 46 seconds. Our findings clearly elucidate the potential of combining classifiers in biomedical analytics research and pave the way for portable point-of-care devices with reasonable accuracy and fast computation times, enabling them to be used for collaborative learning for large-scale, real-time disease modeling.
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 | 0.002 | 0.003 |
| Meta-epidemiology (narrow) | 0.002 | 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 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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.
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".