Deep Convolutional Neural Networks in Medical Image Analysis: A Review

Information, Год журнала: 2025, Номер 16(3), С. 195 - 195

Опубликована: Март 3, 2025

Deep convolutional neural networks (CNNs) have revolutionized medical image analysis by enabling the automated learning of hierarchical features from complex imaging datasets. This review provides a focused CNN evolution and architectures as applied to analysis, highlighting their application performance in different fields, including oncology, neurology, cardiology, pulmonology, ophthalmology, dermatology, orthopedics. The paper also explores challenges specific outlines trends future research directions. aims serve valuable resource for researchers practitioners healthcare artificial intelligence.

Язык: Английский

---

A multi-verse optimizer-based CNN-BiLSTM pixel-level detection model for peanut aflatoxins

Food Chemistry, Год журнала: 2024, Номер 463, С. 141393 - 141393

Опубликована: Сен. 24, 2024

Язык: Английский

---

Non-invasive diagnosis of lung diseases via multimodal feature extraction from breathing audio and chest dynamics

Computers in Biology and Medicine, Год журнала: 2025, Номер 191, С. 110182 - 110182

Опубликована: Апрель 10, 2025

Язык: Английский

---

A Content-Based Medical Image Retrieval System for Lung Diseases Using Mask AttnRCNNpro Segmentation and Hybrid Distance Approach

Research Square (Research Square), Год журнала: 2025, Номер unknown

Опубликована: Апрель 14, 2025

At present, Content-Based Medical Image Retrieval Systems (CBMIRS) are a novel and potentially useful technology though they lack clinical validation. The study aims to assess how CBMIRS helps in interpretation of chest X-ray (CXR) images patients who have lung disease. This paper proposes Lung-CBMIR, new hybrid model that enhance retrieval precision computational complexity for disease images. system combines Mask AttnR-CNNpro, an improved segmentation uses attention mechanisms precisely segment areas. Feature extraction is done through Local Binary Patterns (LBP) texture features, shape descriptors geometric pattern, DenseNet+, which utilizes three dense blocks strategic pooling methods achieve deep feature extraction. Bobcat-Fish Hybrid Optimizer (BFHO) method proposed this integrates Bobcat Optimization exploration ability with the exploitation capability Catch Fish optimal selection features. There also distance metric, combining Mahalanobis Cosine distances, improves image similarity measurement. Furthermore, rank based on their relevance query compile them into vector. Lastly, DeepCL-Net classifier, combination Convolutional Neural Networks (CNN) Long Short-Term Memory (LSTM) networks, facilitates effective classification illnesses like pneumonia, infiltrates, nodules. Lung-CBMIR found attain accuracy 98.75%, F1-score 98.13%, MCC 0.9801, better than state-of-the-art models CNN-AE 95.58% VGG-19 96.81%. results confirm greatly accuracy, lowers complexity, yields strong tool diagnosis CBMIR tasks. abbreviation concern description manifested Table 1.

Язык: Английский

---

Audio-based digital biomarkers in diagnosing and managing respiratory diseases: a systematic review and bibliometric analysis

European Respiratory Review, Год журнала: 2025, Номер 34(176), С. 240246 - 240246

Опубликована: Апрель 1, 2025

Advances in wearable sensors and artificial intelligence have greatly enhanced the potential of digitised audio biomarkers for disease diagnostics monitoring. In respiratory care, evidence supporting their clinical use remains fragmented inconclusive. This study aimed to assess current research landscape digital medicine through a bibliometric analysis systematic review (PROSPERO CRD 42022336730). MEDLINE, Embase, Cochrane Library CINAHL were searched references indexed up 9 April 2024. Eligible studies evaluated accuracy sound diagnosing managing obstructive (asthma COPD) or infectious diseases, excluding COVID-19. A narrative synthesis was conducted, QUADAS-2 tool used quality risk bias. Of 14 180 studies, 81 included. Bibliometric identified fundamental ( e.g. “diagnostic accuracy”+“machine learning”) emerging “developing countries”) themes. Despite methodological heterogeneity, generally achieved moderate (60–79%) high (80–100%) accuracies. 80% (eight out ten) reported sensitivities specificities asthma diagnosis, 78% (seven nine) 56% (five COPD, 64% eleven) sensitivity specificity values pneumonia diagnosis. Breathing coughing most common biomarkers, with neural networks being technique. Future on should focus testing validity clinically diverse populations resolving algorithmic If successful, hold promise complementing existing tools enabling more accessible applications telemedicine, communicable monitoring, chronic condition management.

Язык: Английский

---

Classification and Recognition of Lung Sounds Using Artificial Intelligence and Machine Learning: A Literature Review

Big Data and Cognitive Computing, Год журнала: 2024, Номер 8(10), С. 127 - 127

Опубликована: Окт. 1, 2024

This review explores the latest advances in artificial intelligence (AI) and machine learning (ML) for identification classification of lung sounds. The article provides a historical overview from invention electronic stethoscope to auscultation sounds, emphasizing importance rapid diagnosis diseases post-COVID-19 era. classifies including wheezes stridors, their pathological relevance. In addition, deeply feature extraction strategies, measurement methods, multiple advanced models classification, such as deep residual networks (ResNets), convolutional neural combined with long short-term memory (CNN–LSTM), transformer (transformer). discusses problems insufficient data replicating human expert experience proposes future research directions, improved utilization, enhanced extraction, using spectrograms. Finally, emphasizes expanding role AI ML sound potential further development this field.

Язык: Английский

---

Deep Learning Classification Algorithms Applications: A Review

Indonesian Journal of Computer Science, Год журнала: 2024, Номер 13(3)

Опубликована: Июнь 15, 2024

This paper examines the recent articles on classification tasks, particularly focusing deep learning Algorithms. The process of categorizing data into distinct classes based specific features is essential for tasks such as image recognition, sentiment analysis, disease diagnosis, and more. article fundamental concepts learning, including neural network architectures like Convolutional Neural Networks (CNNs), Recurrent (RNNs), their variants. It explores significance feature selection techniques in improving model performance. Additionally, this provides a detailed literature review, aiming to foster development more effective efficient algorithms methodologies highlighting applications fields healthcare, agriculture, disaster response, beyond. Through underscores transformative impact approaches enabling automated decision-making, pattern data-driven insights, offering valuable insights researchers, practitioners, policymakers involved aims facilitate methodologies.

Язык: Английский

---

Lightweight attention temporal convolutional network based on multi-scale feature fusion for respiratory prediction in tumor radiotherapy

Biomedical Signal Processing and Control, Год журнала: 2024, Номер 100, С. 106923 - 106923

Опубликована: Сен. 26, 2024

Язык: Английский

---