Integrated ensemble CNN and explainable AI for COVID-19 diagnosis from CT scan and X-ray images

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Oct. 23, 2024

In light of the ongoing battle against COVID-19, while pandemic may eventually subside, sporadic cases still emerge, underscoring need for accurate detection from radiological images. However, limited explainability current deep learning models restricts clinician acceptance. To address this issue, our research integrates multiple CNN with explainable AI techniques, ensuring model interpretability before ensemble construction. Our approach enhances both accuracy and by evaluating advanced on largest publicly available X-ray dataset, COVIDx CXR-3, which includes 29,986 images, CT scan dataset SARS-CoV-2 Kaggle, a total 2,482 We also employed additional public datasets cross-dataset evaluation, thorough assessment performance across various imaging conditions. By leveraging methods including LIME, SHAP, Grad-CAM, Grad-CAM++, we provide transparent insights into decisions. model, DenseNet169, ResNet50, VGG16, demonstrates strong performance. For image sensitivity, specificity, accuracy, F1-score, AUC are recorded at 99.00%, 0.99, respectively. these metrics 96.18%, 0.9618, 0.96, methodology bridges gap between precision in clinical settings combining diversity explainability, promising enhanced disease diagnosis greater

Language: Английский

---

An enhanced tree-seed algorithm for global optimization and neural architecture search optimization in medical image segmentation

Biomedical Signal Processing and Control, Journal Year: 2025, Volume and Issue: 104, P. 107457 - 107457

Published: Jan. 8, 2025

Language: Английский

---

SG-UNet: Hybrid self-guided transformer and U-Net fusion for CT image segmentation

Journal of Visual Communication and Image Representation, Journal Year: 2025, Volume and Issue: unknown, P. 104416 - 104416

Published: Feb. 1, 2025

Language: Английский

---

MDSTransUNet: Multi-Scale Deep Supervised Transformer U-Net for COVID-19 Lung and Infection Segmentation

Published: Jan. 1, 2025

Language: Английский

---

Landscape of 2D Deep Learning Segmentation Networks Applied to CT Scan from Lung Cancer Patients: A Systematic Review

Deleted Journal, Journal Year: 2025, Volume and Issue: unknown

Published: March 4, 2025

The increasing rates of lung cancer emphasize the need for early detection through computed tomography (CT) scans, enhanced by deep learning (DL) to improve diagnosis, treatment, and patient survival. This review examines current prospective applications 2D- DL networks in CT segmentation, summarizing research, highlighting essential concepts gaps; Methods: Following Preferred Reporting Items Systematic Reviews Meta-Analysis guidelines, a systematic search peer-reviewed studies from 01/2020 12/2024 on data-driven population segmentation using structured data was conducted across databases like Google Scholar, PubMed, Science Direct, IEEE (Institute Electrical Electronics Engineers) ACM (Association Computing Machinery) library. 124 met inclusion criteria were analyzed. LIDC-LIDR dataset most frequently used; finding particularly relies supervised with labeled data. UNet model its variants used models medical image achieving Dice Similarity Coefficients (DSC) up 0.9999. reviewed primarily exhibit significant gaps addressing class imbalances (67%), underuse cross-validation (21%), poor stability evaluations (3%). Additionally, 88% failed address missing data, generalizability concerns only discussed 34% cases. emphasizes importance Convolutional Neural Networks, UNet, analysis advocates combined 2D/3D modeling approach. It also highlights larger, diverse datasets exploration semi-supervised unsupervised enhance automated diagnosis detection.

Language: Английский

---

Image segmentation with Cellular Automata

Heliyon, Journal Year: 2024, Volume and Issue: 10(10), P. e31152 - e31152

Published: May 1, 2024

Image segmentation is a computer vision technique that involves dividing an image into distinct and meaningful regions or segments. The objective was to partition the areas share similar visual characteristics. Noise undesirable artifacts introduce inconsistencies irregularities in data. These severely affect ability of most algorithms distinguish between true features, leading less reliable lower-quality results. Cellular Automata (CA) computational concept consists grid cells, each which can be finite number states. cells evolve over discrete time steps based on set predefined rules dictate how cell's state changes according its own states neighboring cells. In this paper, new approach CA model introduced. proposed consisted three phases. initial two phases process, primary eliminate noise interfere with identification exhibiting To achieve this, designed modify value cell pixel elements. third phase, element assigned chosen from directly represent final values for corresponding method evaluated using different images, considering important quality indices. experimental results indicated produces better-segmented images terms robustness.

Language: Английский

---