Pneumonia occurs frequently and can be life-threatening all over the world, mainly in locations where radiologists are hard to find, so chest X-ray imaging is widely used and correctly interpreting it is difficult.
Pneumonia remains a leading cause of morbidity and mortality worldwide, and timely radiological screening is essential for effective treatment, particularly in settings where specialist radiologists are scarce. Convolutional neural networks (CNNs) classify chest X-ray images as normal or pneumonic with high accuracy, but their computational cost limits deployment on portable, low-power clinical devices. This paper presents an end-to-end pipeline that trains a ResNet-50 network, using transfer learning, to classify chest radiographs and deploys the trained model on a Xilinx ZCU104 Zynq UltraScale+ FPGA board through the Vitis AI toolchain. Images from a public pediatric pneumonia dataset (5216 training, 624 test, 16 validation images) are resized, normalized, and evaluated under several classical filtering operations before being passed to the network, which combines residual convolutional blocks, ReLU activations, and a sigmoid output layer. The model is trained for 30 epochs with the Adam optimizer and binary cross-entropy loss, reaching a validation accuracy in the 90–92% range. After quantization and compilation for the on-board Deep-Learning Processing Unit, the FPGA implementation achieves an inference throughput of 1250 frames per second at 4.1 W, compared with 44 frames per second at 19.9 W on a general-purpose CPU baseline — approximately a 28-fold gain in throughput and a 138-fold gain in energy efficiency. These results indicate that SoC-FPGA platforms are a practical route to real-time, energy-efficient pneumonia screening at the point of care.
V. M. Vinaya, R. P. Vishwanath, N. P. Nainika et al.· Indian Journal of Electronic...· 0 citations
An attention-enhanced deep learning framework for clinically accurate pneumonia identification from chest imaging radiology that combines a self-attention mechanism with a pretrained VGG16 backbone is proposed and tested against many cutting-edge convolutional neural network architectures.
Mohini Gahlot, Pinaki Ghosh· International journal of com...· 0 citations
Background: Pneumonia remains a leading cause of mortality worldwide, with chest X-ray serving as the primary diagnostic tool. However, manual interpretation is subject to inter-observer variability, and existing deep learning models often require substantial computational resources that limit deployment in resource-constrained clinical environments.
Objective: This study aimed to develop Light CNN, a novel lightweight convolutional neural network that integrates depthwise separable convolution, inverted residual blocks, channel shuffle mechanism, and lightweight attention for efficient and accurate pneumonia classification from chest X-ray images.
Methods: Light CNN was designed with seven progressive feature extraction stages that incorporate the four aforementioned optimization techniques. The model was trained and evaluated on the publicly available Chest X-Ray Images (Pneumonia) dataset from Kaggle, comprising 5,856 images stratified into training (70%), validation (15%), and test (15%) subsets with patient-level splitting to prevent data leakage.
Preprocessing included CLAHE contrast enhancement, normalization, and data augmentation. Training employed the AdamW optimizer with cosine annealing scheduling and class-weighted cross-entropy loss over 50 epochs. The performance of Light CNN was benchmarked against three baseline models — MobileNetV2 (2.23 M parameters), ResNet-18 (11.18 M parameters), and EfficientNet-B0 (4.01 M parameters) — using identical preprocessing and training protocols. Evaluation metrics included accuracy, precision, recall, F1-score, AUC-ROC, parameter count, model size, and inference time.
Results: LightCNN achieved 95.56% accuracy, 0.9556 recall, 0.9584 precision, 0.9562 F1-score, and 0.9875 AUC-ROC on the test set, outperforming all baseline models. The model contains 2.52 million parameters (9.63 MB), representing a 77.4% reduction compared to ResNet-18, with an inference time of 0.25 ms per image — approximately four times faster than the nearest competitor. Ablation study results confirmed that each architectural component contributed incrementally to overall performance; depth wise separable convolution provided the largest efficiency gain, and inverted residual blocks contributed the most substantial accuracy improvement.
Conclusion: Light CNN demonstrates that systematic integration of lightweight architectural techniques can achieve clinically relevant diagnostic performance with minimal computational overhead, supporting its potential deployment in mobile and edge computing scenarios for point-of-care pneumonia diagnosis.
W. Swastika, Heri Kristianto, Paulus Lucky Tirma Irawan et al.· Health Sciences Investigatio...· 0 citations
The proposed deep learning models provide an efficient and accurate tool for multiclass pneumonia detection from CXR images and have the potential to support healthcare professionals in making more accurate diagnoses.
Timothy Karani, Stephen Waithaka· Journal of the Kenya Nationa...· 0 citations
Pneumonia is a leading cause of infectious disease mortality worldwide, accounting for approximately 2.5 million deaths annually and 15% of deaths in children under five. Chest X-ray imaging remains the primary diagnostic tool, but accurate interpretation requires radiological expertise that is disproportionately concentrated in high-income settings, creating a diagnostic gap where disease burden is highest. Automated deep learning offers a scalable complement to specialist-dependent diagnosis, yet clinical adoption requires both high accuracy and transparent, interpretable reasoning. Convolutional neural networks (CNNs) have shown strong potential for pneumonia detection from chest X-rays, but two barriers impede clinical translation: the interpretability of black-box models and the computational feasibility of large architectures in resource-constrained settings. Explainable AI (XAI) methods such as Grad-CAM, Grad-CAM++, and Score-CAM address the interpretability barrier, yet systematic quantitative comparisons across multiple CNN architectures remain scarce. Furthermore, CNN architectures widely used for medical image classification carry high parameter counts that limit feasibility in resource-constrained settings, motivating architectures that achieve competitive accuracy with substantially fewer parameters. Here we propose a parameter-efficient deep learning framework for pneumonia detection based on transfer learning, evaluated across three CNN architectures representing distinct architectural families: EfficientNet-B0 with fine-tuning (proposed method), ResNet50, and DenseNet121, trained under identical conditions on the Kaggle chest X-ray dataset (5,863 images). Our method achieved 90% classification accuracy, outperforming both baselines while requiring 4.8x fewer parameters than ResNet50. To evaluate explainability, Grad-CAM, Grad-CAM++, and Score-CAM were applied across all three architectures and compared quantitatively using Intersection over Union against manually annotated lung segmentation masks, Insertion score, and Deletion score, with pairwise statistical validation via Wilcoxon signed-rank tests and Bonferroni correction. Findings show that classification accuracy and XAI explanation quality must be evaluated independently, and that the proposed parameter-efficient architecture offers a favorable trade-off for resource-constrained clinical deployment.
B. Mahtabi, E. Nasr-Esfahani, S. Yaraghi· medRxiv· 0 citations
Abstract Pneumonia diagnosis via chest X-rays is critical for effective clinical intervention, yet conventional approaches often fail to capture subtle pathologies. This article introduces a new Pneumonia Multi-Scale Attention Network (PMSAN). This deep learning framework synergistically integrates multiscale feature extraction, channel-wise attention mechanisms, and a novel edge-aware loss (EAL) function to improve pneumonia classification accuracy and interpretability. Extensive experiments on a Kermany pediatric chest X-ray data set (5,856 images) demonstrate that PMSAN, when trained with the EAL, achieves a test set accuracy of 96.4%, precision of 97.9%, recall of 97.2%, F1-score of 97.6%, and an area under the curve (AUC) of 0.988. In addition, fivefold cross-validation shows consistent performance with an accuracy of 96.4% ± 0.5% and AUC of 0.988 ± 0.003, outperforming baseline models such as ResNet18 and VGG16. The model's enhanced interpretability is supported by visualizations including receiver operating characteristic curves, attention maps, and edge maps.
M. R. Mahdiani· Indian Journal of Radiology...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
