Deep Learning-Based Skin Disease Detection and Classification: A Systematic Literature Review

This review paper examines the current state of lung disease diagnosis based on deep learning (DL) methods. Lung diseases, such as Pneumonia, TB, Covid-19, and lung cancer, are significant causes of morbidity and mortality worldwide. Accurate and timely diagnosis of these diseases is essential for effective treatment and improved patient outcomes. DL methods, which utilize artificial neural networks to extract features from medical images automatically, have shown great promise in improving the accuracy and efficiency of lung disease diagnosis. This review discusses the various DL methods that have been developed for lung disease diagnosis, including convolutional neural networks (CNNs), deep neural networks (DNNs), and generative adversarial networks (GANs). The advantages and limitations of each method are discussed, along with the types of medical imaging techniques used, such as X-ray and computed tomography (CT). In addition, the review discusses the most commonly used performance metrics for evaluating the performance of DL for lung disease diagnosis: the area under the curve (AUC), sensitivity, specificity, F1-score, accuracy, precision, and the receiver operator characteristic curve (ROC). Moreover, the challenges and limitations of using DL for lung disease diagnosis, including the limited availability of annotated data, the variability in imaging techniques and disease presentation, and the interpretability and generalizability of DL models, are highlighted in this paper. Furthermore, strategies to overcome these challenges, such as transfer learning, data augmentation, and explainable AI, are also discussed. The review concludes with a call for further research to address the remaining challenges and realize DL's full potential for improving lung disease diagnosis and treatment.

Accurate detection of skin diseases is essential for timely intervention and treatment. This article proposes a patch-based, interpretable deep learning framework for skin disease detection using wearable sensors and clinical data. Specifically, a fully convolutional residual neural network (FCRN) is employed to extract local features from high-resolution skin images captured via wearable sensors, using a patch-level training approach. Pre-processing techniques—including image resampling, intensity normalization, and noise reduction—standardize the input data to ensure consistency across sensor variations. To enhance local feature learning, the FCRN incorporates residual modules, which mitigate gradient vanishing and improve model performance. The framework generates disease probability maps that visualize regions of high diagnostic risk, providing interpretable insights into skin anomalies. In the proposed methodology, a convolutional neural network (CNN) integrates image-derived features with clinical data such as patient demographics, symptoms, and medical history. This CNN-based multimodal fusion approach improves the model’s ability to capture spatial relationships and enhances classification performance. Experimental evaluations demonstrate that the proposed framework achieves state-of-the-art results across multiple evaluation metrics, including accuracy, sensitivity, and specificity. The interpretable disease probability maps highlight affected skin regions, enhancing model transparency and clinical usability. This approach demonstrates the potential of combining wearable sensor technology with deep learning for efficient, scalable, and explainable skin disease detection, laying the foundation for real-time clinical applications.

In the face of rapid technological advancements, innovations in wave forecasting are increasingly essential for effectively addressing the complex impacts of climate change. This study aims to explore technological developments in wave forecasting that can manage the complexities related to climate change and enhance the accuracy and efficiency of predictions in dynamic marine environments. Employing a qualitative approach through a Systematic Literature Review (SLR) methodology, the research focuses on literature from databases such as Scopus, DOAJ, and Google Scholar, specifically targeting publications from 2014 to 2024. Recent findings reveal that advancements in machine learning technologies, including deep learning, ensemble learning, transfer learning, and data augmentation, have significantly improved the precision and efficiency of wave forecasting models. Techniques like Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) have been particularly effective in capturing complex, non-linear patterns within wave data, enhancing the overall prediction accuracy. Ensemble learning methods have further contributed by increasing the stability and robustness of forecasts. Moreover, transfer learning and data augmentation play vital roles in adapting these models to rapidly changing environmental conditions, making them highly relevant in the context of climate change. These approaches are crucial for models to remain adaptable and responsive to dynamic oceanic conditions influenced by climate variability. The insights derived from this study are expected to provide valuable direction for the future development of machine learning-based wave forecasting models, emphasizing the need for innovative techniques that can accommodate the complexities and uncertainties brought about by climate change.

Content-Based Image Retrieval (CBIR) has become a critical technology for efficiently searching and retrieving images from large datasets based on their visual content. Traditional CBIR systems, which rely on low-level features like color, texture, and shape, often struggle with semantic gaps and scalability issues. With the rapid advancements in deep learning and cloud computing, there is a growing need to enhance CBIR performance for real-world applications. In order to tackle the issues, an enhanced CBIR process leveraging advanced neural networks, particularly Convolutional Neural Networks (CNNs), Siamese Networks, and attention mechanisms is proposed. It integrates multimodal data, including text and audio, to improve retrieval accuracy. Additionally, cloud-based infrastructure is employed to support large-scale image processing, enabling faster retrieval times and real-time performance. Edge computing techniques are also incorporated to reduce latency in applications requiring immediate responses. The proposed model demonstrates significant improvements in retrieval accuracy and efficiency compared to traditional CBIR methods. Deep learning models, particularly CNNs with transfer learning and attention mechanisms, effectively capture high-level semantic features. The integration of cloud infrastructure enhances scalability and real-time processing capabilities, while multimodal retrieval improves search relevance. The use of explainable AI techniques adds transparency to the decision-making process, increasing user trust. Hence, the advanced neural networks, coupled with cloud and edge computing, can significantly optimize CBIR systems, making them more robust, scalable, and applicable to a wide range of industries such as healthcare, security, and e-commerce.

Recent advances in deep learning have aided the well-being business in Medical Imaging of numerous disorders like brain tumours, a serious malignancy caused by unregulated and aberrant cell portioning. The most frequent and widely used machine learning algorithm for visual learning and image identification is CNN. In this article, the convolutional neural network (CNN) technique is used. Augmentation of data and processing of images is used to classify scan imagery of brain MRI as malignant or benign. The performance of the proposed CNN model is compared with pre-trained models: VGG-16, ResNet-50, and Inceptionv3 using the technique which is transfer learning. Even though the experiment was conducted on a relatively limited dataset, the experimental results reveal that the suggested scratched CNN model accuracy achieved is 94 percent, VGG-16 was extremely effective and had a very low complexity rate with an accuracy of 90 percent, whereas ResNet- 50 reached 86 percent and Inception v3 obtained 64 percent accuracy. When compared to previous pre-trained models, the suggested model consumes significantly less processing resources and achieves significantly higher accuracy outcomes and reduction in losses.

The leather industry plays a crucial role in the production of high-quality leather products to remain competitive in the market. Due to various reasons such as material handling and turning, leather defects can appear at different stages of the production process. This paper proposed deep learning techniques, specifically transfer learning, deep convolutional neural networks (D-CNN), and ensemble learning for automating leather defect detection and classification. The dataset was taken from the open source Kaggle. The dataset was preprocessed from the beginning. However, we performed Error Level Analysis (ELA) to detect possible changes or fraud in the image. We also used data augmentation techniques (resizing, rescaling, flipping, rotation, zooming, and contrasting) on the dataset to improve the model performance. This study investigated deep learning architectures for automatic leather defect detection, comparing them with seven deep convolutional neural networks (D-CNN) models and transfer learning architectures. Furthermore, we introduce and evaluate two novel ensemble models: MER built on transfer learning architecture (MobileNetV2, EfficientNetB0, and ResNet50) and AZL built on D-CNN architecture (AlexNet, ZfNet, and LeNet). A key contribution of this study is the demonstration of significant performance improvements achieved by strategically integrating squeeze and excitation and label smoothing techniques into transfer learning, D-CNN, and ensemble frameworks. Our proposed MER ensemble method achieves a state-of-the-art of 96.39% for leather defect detection. The proposed model was evaluated using performance metrics including precision, recall, F1-score, AUC, ROC, and Matthews Correlation Coefficient (MCC). Additionally, Grad-CAM was employed for explainable AI to visualize and interpret the model’s decision-making process. Comparative analysis against separate transfer learning and D-CNN models fails to establish the superior performance of our MER ensemble, which highlights the potential to significantly advance automated quality inspection in leather production. This study provides insights into evaluating the best deep learning techniques for leather defect classification, paving the way for more accurate, efficient, and reliable industrial applications. Proposes a robust deep learning-based framework for automated leather defect detection using transfer learning, deep CNNs, and two novel ensemble models (MER and AZL), enhanced with squeeze-and-excitation and label smoothing techniques. Achieves state-of-the-art performance with the proposed MER ensemble model, achieve 96.39% accuracy, and demonstrates superior results across precision, recall, F1-score, AUC–ROC, and MCC. Integrates Error Level Analysis (ELA), extensive data augmentation, and Grad-CAM–based explainable AI to ensure data reliability, improve generalization, and provide transparent visual interpretation of defect regions for industrial quality inspection.

Bird species identification and classification are challenging yet crucial for research and conservation. Traditional methods are labor-intensive, require specialized expertise, and can be prone to error. Recent advances in deep learning offer an automated solution to this complex problem. This study evaluates a convolutional neural network (CNN) model for classifying images of 525 bird species. The model employs transfer learning using the EfficientNetB0 architecture and was trained on over 84,000 images. With data augmentation, the model achieved 87% validation accuracy and 86.7% test accuracy, demonstrating its ability to overcome limited data and generalize well. Difficulties in obtaining balanced, high-quality data for each species were addressed through transfer learning and augmentation.

Dogs are beloved pets and loyal companions to millions of people worldwide. Unfortunately, dogs can also suffer from a variety of skin diseases that can cause discomfort, pain, and even life-threatening complications. Some dog skin diseases can be transmitted to humans through direct contact. Early detection and treatment of these skin diseases are crucial for the health and well-being of dogs and humans. Our project aims to provide a quick and precise approach to identifying various types of skin diseases in dogs. To expedite the process of identifying and diagnosing infections related to canine interaction, we plan to utilize a machine-learning model. This approach aims to reduce the time and expertise required for accurate and consistent diagnosis, which can otherwise be challenging and time-consuming. Two models, InceptionV3 and MobileNetV2, were utilized and compared in our implementation. In the case of InceptionV3, a training accuracy of 0.99 and a validation accuracy of 0.98 were achieved. For MobileNetV2, we attained a validation accuracy of 96 and a categorical accuracy of 97. Key Words: Dermatophytosis, zoonosis, image classification, deep learning, Transfer learning, InceptionV3, MobileNetV2, CNN, DNN.

The diseases affecting orange fruits are one of the major challenges to citrus farming around the worldwide. Conventional techniques for detecting these diseases are laborious, time-consuming and may result in erroneous outcomes. This study proposes an advanced deep learning approach to enable fast and accurate classification of various orange fruit diseases through images. This work utilizes transfer learning and involves tuning three pre-trained Convolutional Neural Network (CNN) models: Inception-V3, Inception-ResNet-V2, and EfficientNet-V2-B3 on a large dataset of orange fruit images. The proposed approach is tested for five values of dropout percentages (10% to 50%), accompanied by extensive data augmentation strategies to improve generalization capabilities. The performance of machine learning models is assessed by evaluating parameters such as accuracy rate, precision, recall value, and confusion matrix. The effectiveness of deep learning is shown through high classification: The EfficientNet-V2-B3 achieved the highest accuracy of 99.1% with a loss of 0.055 at a 30% dropout rate. Meanwhile, Inception-ResNet-V2 attained 96.7% accuracy, and Inception-V3 reached 96.5% accuracy. These outcomes significantly outperform classical approaches. This study offers a strong and scalable approach to detecting various diseases at an early stage to encourage sustainable citrus growth and reduce economic challenges associated with farming. A web application was also developed based on the best-model approach for easy activation of AI in farming.

Recent advancements in deep learning, particularly convolutional networks, have rapidly become the preferred methodology for analyzing medical images, facilitating tasks like disease segmentation, classification, and pattern quantification. Central to these advancements is the capacity to leverage hierarchical feature representations acquired solely from data. This comprehensive review meticulously examines a variety of deep learning techniques applied across diverse healthcare domains, delving into the intricate realm of medical imaging to unveil concealed patterns through strategic deep learning methodologies. Encompassing a range of diseases, including Alzheimer’s, breast cancer, brain tumors, glaucoma, heart murmurs, retinal microaneurysms, colorectal liver metastases, and more, the analysis emphasizes contributions succinctly summarized in a tabular form.The table provides an overview of various deep learning approaches applied to different diseases, incorporating methodologies, datasets, and outcomes for each condition.Notably, performance metrics such as accuracy, specificity, sensitivity, and other crucial measures underscore the achieved results. Specifically, an in-depth discussion is conducted on the Convolutional Neural Network (CNN) owing to its widespread adoption as a paramount tool in computer vision tasks. Moreover, an exhaustive exploration encompasses deep learning classification approaches, procedural aspects of medical image processing, as well as a thorough examination of key features and characteristics. At the end, we delve into a range of research challenges and put forth potential avenues for future improvements in the field.

An AI-assisted explainable mTMCNN architecture for detection of mandibular third molar presence from panoramic radiography

However, orange orchards are frequently threatened by a plethora of diseases that can drastically reduce yield and quality. Early and accurate disease detection is paramount for effective management and mitigation, preventing widespread crop loss and ensuring sustainable agricultural practices. Traditional disease identification methods often rely on visual inspection by experts, which can be time-consuming, subjective, and prone to human error. Moreover, the rapid spread of certain diseases necessitates swift and reliable diagnostic tools. In this context, the application of advanced technologies like deep learning offers a promising avenue for automating and enhancing disease classification in orange fruits. The advent of deep learning has revolutionized various fields, including computer vision, enabling the development of highly accurate image recognition systems. Convolutional Neural Networks (CNNs), a class of deep learning models, have demonstrated exceptional performance in extracting intricate features from images, making them well-suited for tasks like disease classification. In the domain of agricultural disease detection, CNNs have shown remarkable potential in identifying and classifying various plant diseases based on visual symptoms captured in images. . This project will encompass the entire pipeline of developing a robust disease classification system, from dataset acquisition and preprocessing to model training, evaluation, and potential deployment considerations. A comprehensive dataset of orange fruit images, encompassing various disease types and healthy samples, will be assembled. Rigorous preprocessing techniques, including data augmentation, will be employed to enhance dataset diversity and model generalization. The hybrid model will be trained and fine-tuned using appropriate optimization algorithms and evaluated using relevant performance metrics. The ultimate goal is to create a practical and effective tool that can assist farmers and agricultural experts in the early detection and management of orange fruit diseases, contributing to improved crop health and productivity.

Application of deep learning for autonomous detection and localization of colorectal polyps in wireless colon capsule endoscopy

Emotion detection, hence, is an important step toward making human-computer interaction a more enhanced process, where systems are made capable of identifying and responding to the emotional state of users. In fact, multimodal emotion detection systems in which both auditory and visual information are fused are emerging, and these approaches toward expressive emotional states are complementary and robust. Multimodal systems enhance the quality of interacting and, through many applications, can diagnose emotional disorders, monitor automotive safety, and improve human-robot interactions. In nature, the high-dimensional space and dynamic threats have resulted in obtaining low accuracy and high computational cost in applying the traditional models based on single-modality data. On the other hand, multimodal systems explore the synergy between audio and visual data, giving better performance and higher accuracy in inferring subtle emotional expressions. The latest improvement was done on these systems using recent advancements in transfer learning and deep learning techniques.That being said, this research Proposal devises a multimodal emotion recognition system integrating speech and face information through transfer learning for improved accuracy and robustness. Serving this purpose, the objectives of this research entail the effective comparison among different transfer-learning strategies, including the impact of pre-trained models in speech-based emotion recognition, and to introduce the role of voice activity detection in the process. Advanced neural network architectures like Spatial Transformer Networks and bidirectional LSTM in facial emotion recognition will also be tested. Early and late fusion strategies will also be used to find the best strategy for combining speech and facial data.This research will target several challenges that involve the complexity of data, balancing of the model performance-robustness balance, computational limitations, and standardization of evaluations in developing a working and robust emotion recognition system to enhance digital interaction and apply in practical areas. The goal is to create a system that oversteps the limitation of single-modality models through state-of-the-art advances in deep learning, as well as front-line improvements in transfer learning, in the manner of emotion detection performance.

Recent advancements in deep learning techniques tailored for autonomous driving in Indian road conditions are crucial for revolutionizing transportation systems. Indian roads present unique challenges, including unpredictable traffic patterns, diverse road infrastructures, and challenging weather conditions. Deep learning is pivotal in addressing these challenges, focusing on perception, decision-making, and control. Various architectures, including Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and Deep Reinforcement Learning (DRL), are analyzed for their efficacy in autonomous driving. Sensor fusion techniques enhance perception capabilities, such as lidar, radar, and camera data integration. Additionally, advancements in semantic segmentation and object detection algorithms improve scene understanding and obstacle recognition. Collaborative efforts between academia, industry, and government agencies are essential for accelerating the deployment of autonomous driving technology in Indian road conditions. In conclusion, leveraging cutting-edge deep learning methodologies enables safe and efficient navigation of autonomous vehicles, ushering in a transformative future in transportation systems.