Advancing Pneumonia Classification and Detection: Comparative Analysis of Deep Learning Models Using Convolutional Neural Networks
Bibliographic record
Abstract
Pneumonia is a fatal disease that arises from a bacterial infection in the lungs. If it is not detected in an early stage, it can cause death among young children. Early detection of this disease can play an important role in the effectiveness of the treatment process. The diagnosis is usually monitored from chest X-ray images by an expert radiologist. Due to a lack of confidence in the diagnosis process regarding ambiguous X-ray images or being mistaken for other medical diseases, the application of computer vision is needed to assist radiologists in the decision-making process. In this study, we utilized techniques from transfer learning alongside three architectures based on convolutional neural networks (CNNs) to facilitate the detection of pneumonia and improve the interpretability of diagnostic outcomes. To achieve this, we used a publicly available dataset of 5,863 grayscale chest X -ray images. These images include standard anterior-posterior (AP) and lateral views obtained from unique patients from Guangzhou Women and Children’s Medical Center in China (1,583 normal and 4,280 pneumonia images). Before the training, validation, and testing phases, data preprocessing techniques including image resizing and data augmentation were used to prepare the dataset for binary classification. To enhance the generalizability and efficacy of our findings, we utilized high-performing pre-trained models such as DenseNet121, DenseNet169, and ResNet101, evaluating the performance of each architecture against an external validation, and test set. The evaluation results of deep learning models for binary classification of pneumonia demonstrated that DenseNet121 outperformed its counterparts achieving the highest validation accuracy of 98.68% and the lowest loss value of 0.04.
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 imitationNot 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.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 teacher head, 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".